Skip to main content
Comms AI Pentathlon 2026 · Event 3 of 5

The marathon: comms strategy

The brief, all eight entries exactly as the judges received them, every judge’s score and justification, and the human judge’s rank and note for each entry.

The brief

Pasted word for word into a fresh chat in every app.

You are a communications consultant. Westbrae Buses, a fictional regional bus operator serving a mix of small towns and rural communities, has hired you. The brief from the managing director: "Next spring we are making the biggest changes to our network in twenty years: twelve routes revised, four low-usage routes withdrawn, two new express routes added, and a full move to capped contactless fares. Independent analysis says 70 per cent of current passengers will be better off, but the four withdrawn routes serve some of our most vocal communities and two local councillors have already started petitions based on leaked partial information. Our own drivers heard about the leak before we briefed them, and morale is poor. We have a comms team of two, a modest budget, eight months until the changes take effect, and a local press that enjoys a 'war on rural passengers' headline. We need a communications strategy." Your task: produce a communications strategy for the eight months to launch, covering objectives, audiences, key messages, channel choices, phasing, risk handling and how success will be measured. It should be a working document the comms team of two can actually run, not a textbook exercise.

The three criteria

  • Strategic coherence (objectives, audiences, messages and channels genuinely connect)
  • Practicality (deliverable by the stated team and budget, honestly measurable)
  • Insight (choices are argued, not listed; recognises the hostility problem).

Notes on this event

  • The judging packs for this event ran to about 118,000 characters. GLM judged it across four fresh chats and Siri across two, with the same prompt in each part and the presentation order kept.
  • Copilot's reply stopped on screen mid-plan. Only the first reply is judged, as returned by Copilot's own copy button (which ran to section 8). The "Continue" turn that followed was not judged.
  • Claude produced its entry as a Claude Docs document. The document's link and the byline were removed before judging; its "[embed: node/...]" line was left in, so the judges saw it.
  • ChatGPT searched the web while producing its entry.
  • Siri's closing line ("By the way, I'm an assistant powered by AI and may make mistakes. Always verify important details.") was removed before judging.

Results

Mean of the eight AI judges’ scores, to one decimal place, with the unrounded mean beside it. The human judge ranked the same entries blind, 1 to 8. Word counts are whitespace-separated tokens in the text the judges saw.

Results for event 3
Rank Assistant Letter Mean Unrounded Medal Human judge Words
1 Claude G 9.1 9.1125 Gold 1st 4,989
2 GLM D 8.8 8.7875 Silver 2nd 2,415
3 GPT H 8.5 8.5375 Bronze 4th 2,195
4 Kimi F 8.4 8.3875 3rd 1,799
5 DeepSeek A 8.1 8.1125 7th 2,202
6 Copilot C 6.3 6.2750 6th 1,460
7 Gemini B 5.8 5.8125 5th 1,373
8 Siri E 5.3 5.2500 8th 642

Tied means share the medal. Human judge’s rank correlation with the AI judges’ order in this event: 0.88.

All 64 scores

Rows are judges, columns are entries. The outlined cell in each row is the judge scoring its own entry, which it could not identify.

Every judge’s score for every entry in event 3
Judge A DeepSeekB GeminiC CopilotD GLME SiriF KimiG ClaudeH GPT
GPT 6.23.65.67.54.57.38.5Own entry: 9.1
Claude 7.44.65.28.63.87.8Own entry: 8.87.6
Gemini 9.2Own entry: 6.56.89.05.58.89.48.5
DeepSeek Own entry: 8.46.85.48.74.67.99.28.1
GLM 9.46.87.0Own entry: 9.36.89.39.49.2
Kimi 8.35.66.19.06.4Own entry: 8.49.28.2
Copilot 8.54.8Own entry: 7.69.45.98.49.29.6
Siri 7.57.86.58.8Own entry: 4.59.29.28.0
Mean 8.15.86.38.85.38.49.18.5

The entries

In blind-letter order, as the judges received them. Shown here from Markdown, with headings reduced in size for the page; the exact text is in the download. Judges are listed in lane order.

Entry A · DeepSeek Mean 8.1 · 5th with the AI judges · 7th with the human judge · 2,202 words Show

chat.deepseek.com, default settings

Here is a working communications strategy for Westbrae Buses, designed to be run by a team of two over eight months. It is deliberately practical: fewer, bigger actions, clear owners, and pre-written materials to reduce daily load.


Westbrae Buses: Network Change Communications Strategy

Period: Eight months to launch (Month 0 = now; Month 8 = changes take effect, “Spring”)
Owner: Managing Director (accountable), Communications Lead (day-to-day), Comms Officer (delivery)
Budget: Modest. Priority spend: one part-time local press/community liaison freelancer for Months 5–8; printing for affected-route mailings; accessibility formats; small paid social boost in the final six weeks.


1. The situation in one paragraph

We are making the biggest network change in twenty years. Most passengers (70%) will be better off, but four routes are being withdrawn, serving vocal communities. Partial, leaked information has already triggered two councillor-led petitions and damaged driver morale. The local press default is a “war on rural passengers” frame. We have two comms staff, limited money, and eight months. We cannot win a argument about whether change is happening. We can win the argument about how fairly and openly it is handled, and we must give our own people the truth before anyone else gets it.


2. Objectives

Primary

  1. Ensure every affected passenger knows exactly what is changing, when, and what their alternatives are, at least six weeks before launch.
  2. Protect trust and licence to operate: no ambush, no misinformation vacuum, no staff learning from the press.

Secondary
3. Convert the conversation from “war on rural passengers” to “honest reallocation of a limited network, with support for those who lose out.”
4. Maintain driver morale and enable them to be credible frontline ambassadors.
5. Build a reusable contact list and channel base for future changes.

Explicitly not an objective: persuading the four withdrawn-route communities to support withdrawal. That is not achievable and chasing it wastes the budget. The goal there is fair process and visible support, not agreement.


3. Audiences (in priority order)
PriorityAudienceWhat they needWho owns relationship
1Drivers and depot staffTruth first, talking points, certainty about jobsMD + Comms Lead
2Passengers on the four withdrawn routesPersonal notice, alternatives, supportComms Officer
3Passengers on the twelve revised routesRoute-specific “what changes for you”Comms Officer
4Local councillors (all, not just the two)Early briefing, constituent answersMD
5Local press and radioFacts, access, a human story that isn’t only angerComms Lead
6New express-route catchmentsAwareness and trialComms Officer
7Schools, employers, health/social care, community transportCoordination on alternativesComms Lead
8Regional transport body / fundersAssurance of processMD

Two councillors have already started petitions. Treat them as a known, permanent channel, not a fire to put out. Brief them early and repeatedly, even when they will still oppose.


4. Key messages

Core narrative: The network has to change to stay viable. Most passengers gain. Where routes are withdrawn, we are being open about it, we are explaining why, and we are putting real support in place. No one should find out from a leak or a headline.

Message discipline — use these, in this order:

  1. Honesty: “We are telling you everything now, including the difficult parts.”
  2. Scale and benefit: “This is the biggest change in twenty years. Around seven in ten current passengers will be better off — better frequency, better connections, capped fares.”
  3. Fares: “Contactless capped fares mean you will never pay more than the cap, however many journeys you make.”
  4. Withdrawals — name them, don’t hide them: “Four routes are being withdrawn because usage is very low and the subsidy per passenger is unsustainable. Here is exactly what replaces them.”
  5. Support: “No withdrawn route is being left without an alternative. We have worked with community transport, and we will publish the alternatives for each community.”
  6. Process: “We are consulting, listening, and where we can improve the plan we will.”

Forbidden phrases: “minority of passengers”, “tough choices”, “we appreciate your patience”, “some disruption is inevitable.” These are heard as contempt.

Message for staff only: “You will know before the public. If you don’t know something, say so and take a name — don’t guess.”


5. Channel choices — and the discipline around them

The team of two cannot run ten channels well. Choose five.

1. Direct mail to affected addresses (primary, non-negotiable).
Every household on the four withdrawn and twelve revised routes gets a printed, route-specific leaflet. This is the single highest-trust channel for rural and older passengers. Cost is real but this is where the budget goes. Postcode-level targeting via the electoral roll or a mailing house. Include a QR code and a freephone number.

2. Website “network change” hub (single source of truth).
One URL, printed on everything. Route-by-route pages, an interactive map, a fare calculator, a full FAQ, and a “last updated” date on every page. If a fact changes, it changes here first.

3. Face-to-face engagement (highest value per pound).
Twelve to fifteen drop-in sessions: one in each withdrawn-route community, one in each town centre, plus two depot staff sessions. Two people, a table, printed maps, and a laptop. Book village halls early. Advertise via parish councils and noticeboards.

4. Local press and radio (managed, not feared).
Proactive: offer the MD for a briefing with the editor of each title before launch. Reactive: same-day response to any story. Do not wait to be asked. The “war on rural” headline will come; the goal is that the article also contains our alternatives and the 70% figure because we gave them.

5. Staff channels (internal first).
Depot briefings, a printed staff Q&A, a WhatsApp/Teams group, and a laminated “what to say” card for every driver. Drivers must hear news before passengers, every time. This is the rule that repairs morale.

Optional, low-cost only: paid social boost (final six weeks, targeted at affected postcodes); parish council and community newsletter submissions; posters on buses and at stops.

Do not attempt: a glossy consultation app, a big TV ad, a rebrand, a celebrity endorsement, a separate app.


6. Phasing — the eight months

Phase 1: Stabilise and brief internally (Months 1–2)

  • Day 1–3: MD briefs all staff at every depot, in person. Acknowledge the leak directly: “You should have heard this from us. Here is the full picture now.”
  • Publish the full plan internally before anything external.
  • Appoint a single named contact for councillors and press.
  • Build the route-by-route fact base: for each of the 18 routes, the change, the reason, the alternative, the affected passenger numbers.
  • Start the website hub (even if password-protected or “coming soon”).
  • No proactive external comms yet. Resist the urge to respond to the petitions with a statement. Prepare instead.

Phase 2: Go public with the whole plan (Months 3–4)

  • Publish everything at once: all 18 routes, all four withdrawals, the fares change, the two new express routes. Partial release caused the current problem. Do not repeat it.
  • Press briefing and embargoed packs to local titles, with the MD available.
  • Direct mail to affected households — first wave.
  • Open the first drop-in sessions.
  • Brief all councillors in writing, including the two petitioners, with an offer of a face-to-face.
  • Brief schools, major employers, health and social care, and community transport.

Phase 3: Consult, listen, adjust (Months 4–6)

  • Continue drop-ins, including evenings and weekends.
  • Publish a “you said, we did” log — even small changes build credibility. If a withdrawal cannot be reversed, say so plainly and say why.
  • If any change can be made in response to feedback (a timing, a stop, a connection), make it visibly and announce it. This is the cheapest trust purchase available.
  • Second direct mail wave with any changes.
  • Fare campaign begins: “capped contactless fares” explained simply, with a calculator.

Phase 4: Countdown and launch (Months 6–8)

  • Four weeks out: “changes take effect on [date]” campaign across all channels.
  • Two weeks out: final reminder mailing to affected routes; posters on buses and at stops.
  • Launch week: staff on the ground at key stops; MD visible; daily social and website updates; press invited to ride the new express routes.
  • Week after launch: rapid feedback capture, fix obvious problems fast, publish a “first week” update.

7. Risk handling

Risk 1: The leak continues / new partial information emerges.
Response: publish the full plan earlier if necessary. Never respond to a leak with a partial statement. Have a standing line: “The full plan is at [URL]. We are not going to brief this in fragments.”

Risk 2: “War on rural passengers” headline.
Response: pre-brief editors with the 70% figure, the alternatives, and a named rural passenger who benefits. Offer a rural route that gains frequency as the counter-story. Do not complain about the coverage; out-fact it.

Risk 3: Councillor petitions escalate.
Response: treat the councillors as a channel to their constituents, not as opponents to defeat. Offer a joint drop-in. If they will not engage, publish exactly what we offered and when. Keep the tone factual.

Risk 4: Driver morale and misinformation at the wheel.
Response: the “you heard it first” rule; a laminated Q&A card; a named internal contact; a weekly internal note for the first three months. If a driver is asked something they don’t know, the script is: “I don’t know, but here’s the number and the website — they’ll tell you.”

Risk 5: The four withdrawn routes have genuinely poor alternatives.
Response: this is an operational risk, not a comms one, but comms must not paper over it. Escalate to the MD now. If the alternative is a community transport minibus twice a week, say that plainly and put the operator’s number on the leaflet. Overpromising here will destroy trust for years.

Risk 6: Low attendance at drop-ins.
Response: go where people already are — market days, supermarkets, GP surgeries, church halls, school gates. A table at a Saturday market reaches more people than a Tuesday evening village hall.

Risk 7: Budget runs out before the final mailing.
Response: protect the final mailing and the launch-week staffing above all else. Cut paid social and printing of generic leaflets first.


8. What “good” looks like — measurement

Measure a small number of things honestly.

Reach

  • % of affected households mailed (target: 100% of the four withdrawn routes, 95% of revised routes)
  • Drop-in attendance and unique visitors to the network change hub
  • Press coverage: % of articles containing the 70% figure and the alternatives (target: 60%+)

Understanding

  • Short pulse survey at drop-ins and online: “Do you know what is changing for your route?” (target: 80% of affected passengers say yes by Month 7)
  • Website FAQ page views and freephone call volume — spikes are fine, silence is not

Trust and sentiment

  • Driver morale pulse (two questions, anonymous, at Months 1 and 6)
  • Councillor and parish council feedback log
  • Sentiment of press coverage, tracked simply as positive/neutral/negative

Operational (the real test)

  • Complaints and enquiries in the two weeks after launch vs. a comparable previous change
  • No-show / confusion incidents at stops in launch week
  • New express route trial usage in first month

One honest caveat to give the MD: if the four withdrawn routes genuinely lack alternatives, no communications strategy will make that fair. Comms can deliver honesty and support, but it cannot manufacture a bus that isn’t there. The strategy should be judged on whether people felt informed and treated fairly — not on whether they were happy.


9. The first ten working days (so the team can start tomorrow)
  1. MD briefs all depot staff in person; acknowledge the leak.
  2. Confirm the single named press/councillor contact.
  3. Build the 18-route fact table (change, reason, alternative, affected numbers).
  4. Register the network change URL and put up a holding page with the launch date.
  5. Book village halls and market pitches for Months 3–6.
  6. Request quotes for the two mailings.
  7. Draft the staff Q&A card and the standard press line.
  8. Write to all councillors offering a briefing, including the two petitioners.
  9. Identify one rural passenger who benefits from the changes, and one who loses out but has an alternative — both willing to speak.
  10. Diarise the Phase 2 publication date and work back from it.

Bottom line for the MD: the leak and the petitions are symptoms of a vacuum. Fill the vacuum with the whole truth, brief your drivers first, mail the affected households directly, and put real alternatives on paper. You will not be loved on the four withdrawn routes. You can be believed everywhere else — and that is what protects the network for the next twenty years.

What the judges said

  • GPT: 6.2

    Route-specific information and the refusal to promise public agreement are strong choices. Two mailing waves and 12–15 drop-ins strain the resources, while its messages promise alternatives that have not been established.

  • Claude: 7.4

    The single-owner channel discipline, direct mail to affected households, and the ten-day start list make this easy to run, and the closing caveat about mitigation shows real understanding of the hostility problem. Holding all external comms until months 3 to 4 leaves the vacuum the leak created open for too long, and the promise that "no withdrawn route is being left without an alternative" contradicts its own later warning against overpromising.

  • Gemini: 9.2

    Entry A offers a masterclass in practicality and insight, ruthlessly prioritizing five channels for the two-person team and explicitly abandoning the unwinnable goal of convincing withdrawn communities to like the cuts. Its coherence is exceptional, directly linking the immediate repair of driver morale and a hyper-local print strategy to the overarching objective of restoring trust.

  • DeepSeek: 8.4 (its own entry)

    Highly practical and refreshingly honest, with a strong "fewer, bigger actions" discipline and a clear-eyed refusal to chase unachievable objectives such as persuading withdrawn-route communities to support the cuts. The insight is strong—direct mail as the highest-trust channel, the "you said, we did" log, and the candid caveat that comms cannot manufacture a bus that is not there—though it is slightly less comprehensive than Entry G on phasing detail and risk escalation.

  • GLM: 9.4

    Exceptionally coherent: the publish-everything-at-once logic, prioritised audiences with named owners, and forbidden-phrase message discipline all flow directly from the leak and hostility diagnosis, and it is genuinely runnable by two people (a first-ten-days list, a budget cut-order, two-question pulse surveys, drop-ins at markets instead of empty halls). The insight is outstanding — treating councillors as a channel rather than a fire, declaring that winning over withdrawn-route communities is explicitly not an objective, and the honest caveat that comms cannot manufacture a bus that isn't there — with only minor strain in the sheer workload of mailings, drop-ins and website builds.

  • Kimi: 8.3

    Genuinely strategic: five channels chosen and defended, "persuading withdrawn communities" explicitly rejected as an objective, and an honest closing caveat that comms cannot manufacture a bus that isn't there. Measurement is honest but less rigorous than the best entries, and the phasing is coarser than the quality of the thinking deserves.

  • Copilot: 8.5

    Strategically coherent and highly usable, with strong prioritisation, candid handling of hostility and practical first actions. It loses marks for assuming every withdrawn route has an alternative and for a resource-heavy programme that includes extensive mailing, numerous events and a freelancer on a modest budget.

  • Siri: 7.5

    This entry offers a sensible, grounded approach with a clear focus on direct mail and face-to-face engagement, which is appropriate for the rural demographic. While strategically coherent, it lacks the deeper insight needed to fully neutralise the political hostility, treating the petitions as a permanent fixture rather than a dynamic to be managed.

  • Human judge (Michael MacLennan): ranked 7th of 8

    Decent skeleton, too many audiences without weighting, copy generic and lacks sharpness, a lot needs filled out, do like 'what good looks like' and first working days

Entry B · Gemini Mean 5.8 · 7th with the AI judges · 5th with the human judge · 1,373 words Show

Gemini in Google Workspace, 3.6 Thinking

WESTBRAE BUSES: NETWORK TRANSFORMATION COMMUNICATIONS STRATEGY

Operational Period: 8 Months Pre-Launch to Launch+1 Month

Resourcing: 2 FTE Comms Team | Modest Operational Budget

1. Core Objectives

  1. Rebuild Internal Trust (Months 1–2): Achieve 100% driver coverage via face-to-face briefings, restoring internal sentiment and turning frontline staff into brand advocates before public launch.

  2. Shift Political & Media Narrative (Months 2–4): Pivot local press coverage from a “war on rural passengers” to a “£Xm modernization of regional transit,” securing neutral or positive quotes from at least one key local politician.

  3. Drive Public Awareness & Readiness (Months 5–8): Ensure 80%+ of regular passengers are aware of route changes and contactless capped fares prior to Day 1.

  4. Protect Passenger Retention: Retain existing patronage while achieving a target 15% growth on new express corridors within six months of launch.

2. Stakeholder Matrix & Prioritisation
Given the two-person capacity, engagement is strictly prioritized by operational risk and influence.

Stakeholder GroupPrimary Risk / OpportunityCore ObjectiveKey Channels
Frontline Drivers & Depot StaffHigh risk: Low morale, leak-driven distrust, operational disruption.Inform first, validate frustration, equip to answer passenger queries.Face-to-face depot briefings, driver app, pay-slip inserts, depot posters.
Local Councillors (esp. Petition Organisers)High risk: Political grandstanding, press amplification of route closures.Neutralise hostility; reframe as evidence-based network rebalancing.embargoed 1-on-1 briefings, data packs, mitigation workshops.
Affected Rural Communities (4 Withdrawn Routes)High risk: High vocal opposition, local press sympathy.Direct, honest communication; highlight feeder services or DRT alternatives.Bus stop notices, targeted community drop-ins, parish council newsletters.
Core Commuters & 70% “Gainers”High opportunity: Express travel, lower capped fares, speed benefits.Mobilise support, drive contactless adoption, build brand loyalty.On-bus digital screens, targeted social media, local press release.
Local MediaHigh risk: Sensationalist headlines (“War on Rural Passengers”).Proactive backgrounding; focus on total network investment & 70% savings.Exclusive background briefings, press packs, site visits to new express stops.

3. Key Messaging Framework
Master Narrative
”Westbrae Buses is delivering the biggest upgrade to regional transport in 20 years. We are investing in faster express links, simpler capped contactless fares, and revised schedules so 70% of our passengers get a faster, cheaper journey. To fund these improvements and protect our long-term network, we have had to make hard choices on four unsustainable routes, working with local partners to ensure alternative connections where possible.”

Tailored Segment Messaging

  • For Drivers:

    • “We hear you, and we apologise that you learned about early proposals through a leak. You are the face of Westbrae Buses, and we are committing to updating you first on every stage of this rollout.”
  • For Affected Rural Communities:

    • “We understand the frustration caused by route changes. These four services carried an average of fewer than [X] passengers per trip, making them financially unviable. We are working directly with local councils to signpost alternative community transport and demand-responsive transit options.”
  • For Express Commuters & General Public:

    • “Faster routes, cheaper travel. Tap on, tap off, and never pay more than the daily cap. Two new express corridors will cut journey times into regional hubs by up to 20 minutes.”
  • For Local Politicians & Press:

    • “Independent data shows 7 per cent of local passengers will see direct improvements or lower fares under this rebalancing. We are open to constructive dialogue on how to mitigate impacts on low-density corridors.”

4. Channel Strategy & Resource Allocation
With a two-person team, do not attempt high-maintenance channels (e.g., daily live Q&As, unmoderated open forums, or extensive print advertising).

                  +-----------------------------------+  
                  |   RESOURCE DISTRIBUTION (2 FTE)   |
                  +-----------------------------------+  
                                    |
  +---------------------+-----------+-----------+---------------------+  
  |                     |                       |                     |

Internal & Ops Political & PR Digital & Social On-Ground/Collateral
(35%) (25%) (20%) (20%)
• Depot Briefings • 1-on-1 Briefings • Website Landing • Bus Stop Notices
• Driver Q&A Packs • Media Releases Page • Network Maps
• Pay-slip Inserts • Press Backgrounders • Targeted Ads • Seat Drop Leaflets

  • Digital Hub (Single Source of Truth): Create a dedicated landing page (westbraebuses.co.uk/future) featuring a Route Checker Tool (enter origin/destination to see new route/fare), downloadable maps, and an interactive FAQ.

  • In-Person Engagement: Limit public sessions to 4 focused, drop-in information surgeries (one in each affected rural hub) jointly run with operations staff.

  • Printed Assets: Hyper-local bus stop posters, print-on-demand driver toolkits, and simple seat-drop flyers across the existing fleet.

5. Phased Implementation Roadmap
+---------------------------------------------------------------------------------------------------+
| MONTH 1: INTERNAL REPAIR & POLITICAL RESET | | • Execute Driver Apology & Briefing Roadshow across depots. | | • Hold confidential, embargoed 1-on-1 briefings with the 2 petitioning councillors. | | • Launch internal ‘Driver First’ feedback channel. | +---------------------------------------------------------------------------------------------------+
| +---------------------------------------------------------------------------------------------------+
| MONTHS 2–3: MITIGATION & CONTROLLED ANNOUNCEMENT | | • Issue formal public press release focused on 70% fare/speed gains & contactless rollout. | | • Publish dedicated website & Route Checker tool. | | • Host 4 targeted community drop-in surgeries in affected rural zones. | +---------------------------------------------------------------------------------------------------+
| +---------------------------------------------------------------------------------------------------+
| MONTHS 4–5: PASSENGER EDUCATION & FARE PROMOTION | | • Launch “Tap & Go” educational campaign (video demos on social, bus back posters). | | • Distribute revised network maps to libraries, council offices, and community hubs. | | • Issue driver “Customer FAQ Cards” for daily route queries. | +---------------------------------------------------------------------------------------------------+
| +---------------------------------------------------------------------------------------------------+
| MONTHS 6–7: COUNTDOWN & FINAL PREPARATION | | • Install timetable notices at all affected stops. | | • Run digital targeted social campaigns highlighting new Express routes. | | • Conduct depot refresher sessions with drivers on route changes & fare capping rules. | +---------------------------------------------------------------------------------------------------+
| +---------------------------------------------------------------------------------------------------+
| MONTH 8: LAUNCH & HYPER-CARE | | • Day 1-5: Comms team present at major interchanges and depots. | | • Daily media monitoring and rapid response on social media channels. | | • Post-launch review at Day 30. | +---------------------------------------------------------------------------------------------------+
6. Risk Management & Crisis Protocols
Risk 1: Local Press “War on Rural Passengers” Campaign

  • Mitigation: Offer the local paper an exclusive interview with the MD before the general press release, providing access to independent passenger volume data showing the unviability of the 4 routes alongside the £X savings for the remaining 70% of passengers.

  • Protocol: If negative headlines run, respond with a standardized, empathetic statement emphasizing hard facts: “Continuing to run empty buses on four routes threatened the survival of the entire regional network. Rebalancing allows us to protect transport for 70%+ of local residents.”

Risk 2: Councillor Petitions Gaining Momentum

  • Mitigation: Invite affected councillors to a closed-door “Network Options Working Group.” Present the financial reality and offer concrete micro-mitigations (e.g., tweaking a revised route to cover a medical centre, or co-funding a demand-responsive minibus pilot with the local authority).

  • Protocol: Reframe councillors publicly as “working with us to address local impacts” rather than opponents.

Risk 3: Frontline Driver Pushback / Misinformation to Passengers

  • Mitigation: Provide drivers with a simple, laminated pocket card listing key changes, rationale, and a QR code directing passengers to the digital Route Checker. Provide a £50 voucher incentive for the depot with the highest customer service ratings during Month 8.

  • Protocol: Establish an operational escalation channel so drivers can report customer friction points directly to the comms team for immediate messaging adjustments.
    7. Success Measurement Dashboard

Metric CategoryTarget KPIMethod / Tool
Internal Morale>75% of drivers report feeling “well informed” ahead of launch.Short paper survey distributed at depot sign-in (Month 6).
Media SentimentNeutral or positive coverage in ≥60% of local press mentions.Monthly press clippings log and sentiment tagging.
Passenger Awareness>80% awareness of contactless capping & route changes.On-bus digital surveys / short pop-up web survey.
Digital Adoption>15,000 unique uses of the online Route Checker Tool.Web analytics tracking (westbraebuses.co.uk/future).
Operational Ease<5% fare-query delays during Week 1 of launch.Driver feedback logs & dispatch reports.

What the judges said

  • GPT: 3.6

    Its priorities are visible, but invented investment and journey-time claims, a mistaken “7 per cent” figure, and promises of alternatives undermine the messages. The route checker and extensive activity also stretch the stated budget and team.

  • Claude: 4.6

    The structure is there, but the messages ("empty buses", "financially unviable", a "£Xm modernisation" reframe) would feed the hostile frame rather than defuse it, and the 7 per cent typo in the councillor and press line would be damaging if used. Some choices, such as a £50 depot voucher and a closed-door working group, are thin or counterproductive, and the phasing leads on the gains before the losses are properly handled.

  • Gemini: 6.5 (its own entry)

    While this entry organizes its stakeholders well and keeps the two-person constraint in mind, its messaging leans heavily on corporate jargon that is unlikely to survive contact with hostile rural communities. It lacks the deep, pragmatic insight of other entries, attempting to smoothly "reframe" anger rather than authentically mitigating it.

  • DeepSeek: 6.8

    A genuinely runnable strategy with a sound internal-first sequence, a clear stakeholder prioritisation and a phased roadmap that a two-person team could execute. It loses marks on insight: the messaging slips into "7 per cent" and other minor errors, and the treatment of the four withdrawn communities is more damage-limitation than genuine recognition of the hostility problem.

  • GLM: 6.8

    A complete and well-sequenced framework, but it meets the leak with narrative control rather than candour — no public statement for weeks, "neutralising" critics, and a modernisation frame that outruns the facts — and the press-facing message misstates the core statistic as 7 per cent. Several success measures (route-checker tool, 15 per cent express growth, on-bus digital surveys) assume budget, infrastructure and outcomes the brief does not grant.

  • Kimi: 5.6

    It ticks every required heading, but a key message states "7 per cent of local passengers will see direct improvements" — a corrupting error in the campaign's central statistic — and several other figures (£Xm, 15% growth, 15,000 tool uses) are invented without basis. The response lines lean on the 70% in ways that would read as dismissive to losing communities, so insight into the hostility problem is shallow.

  • Copilot: 4.8

    The structure covers the required components and limits some high-maintenance activity, but invented claims such as “£Xm”, “up to 20 minutes” and the erroneous “7 per cent” seriously weaken credibility. Its emphasis on neutralising politicians, manufacturing supportive quotes and incentivising drivers also misreads the trust problem and creates avoidable risks.

  • Siri: 7.8

    This entry offers a highly structured and logical approach, clearly connecting its objectives to specific stakeholder needs and measurable outcomes. However, it slightly underestimates the political difficulty of neutralising hostile councillors through embargoed briefings alone.

  • Human judge (Michael MacLennan): ranked 5th of 8

    Too short, begins too technically, feels slightly dashed off - resource distribution breakdown is pretty nice, mitigations/methods/etc slim and would need fleshed out

Entry C · Copilot Mean 6.3 · 6th with the AI judges · 6th with the human judge · 1,460 words Show

M365 Copilot (Premium), fresh Business Premium trial, Auto

Below is a practical communications strategy designed for a two-person team with limited budget and eight months to prepare. It assumes the network changes are already decided in principle and that the communications challenge is to maintain trust, minimise misinformation, support staff, and maximise passenger understanding and adoption.

Westbrae Buses
Network Change Communications Strategy
Eight Months to Launch
1. Situation Summary

Westbrae Buses is introducing its largest network redesign in twenty years:

  • 12 routes revised
  • 4 low-usage routes withdrawn
  • 2 new express routes introduced
  • Full transition to capped contactless fares
  • Independent analysis shows 70% of existing passengers will be better off
    However:
  • Information has leaked before formal announcement
  • Petitions are already active in some affected communities
  • Driver morale has been damaged by learning details externally
  • Local media is likely to frame changes as service cuts rather than network improvements
    The communications task is therefore not simply to announce changes. It is to maintain trust through a potentially controversial transition.
2. Communications Objectives
Strategic Objectives
  1. Protect organisational credibility throughout the process.
  2. Ensure drivers and frontline staff become informed advocates rather than critics.
  3. Help passengers understand exactly what the changes mean for them.
  4. Reduce misinformation and speculation.
  5. Shift media coverage from “cuts” to “network modernisation”.
  6. Maximise uptake of contactless capped fares.
  7. Manage opposition respectfully while demonstrating that decisions are evidence-based.
Behavioural Objectives

By launch:

  • 90%+ of drivers report feeling adequately informed.
  • Majority of passengers understand how to find their new journey.
  • Contactless usage exceeds previous ticketing adoption forecasts.
  • Local media coverage is balanced rather than overwhelmingly negative.
  • Complaints peak early and decline before implementation.
3. Audience Priorities
Tier 1: Internal Staff
Drivers

Most important audience.
Passengers trust drivers more than press releases.
Current challenge:

  • Feel blindsided by leak.
  • Likely to become source of rumours.
    Needed outcome:
  • Drivers informed first.
  • Drivers confident answering basic questions.
Depot and Customer Service Teams

Need:

  • Detailed FAQs.
  • Escalation procedures.
  • Consistent messaging.
Tier 2: Existing Passengers
Beneficiaries (70%)

Potentially supportive but often silent.
Communications goal:

  • Turn silent beneficiaries into visible supporters.
  • Highlight faster journeys, better connections and fare savings.
Passengers Directly Affected by Withdrawals

Not expected to support changes.
Communications goal:

  • Demonstrate respect.
  • Explain rationale honestly.
  • Offer alternatives wherever possible.
Tier 3: Stakeholders
Councillors

Some will oppose changes regardless.
Goal:

  • Avoid surprises.
  • Ensure they hear directly from Westbrae before press reports.
Community Groups

Especially in affected villages and towns.
Goal:

  • Give opportunities to ask questions.
  • Show evidence rather than corporate slogans.
Regional Media

Goal:

  • Move narrative from “service cuts” to:
    • reliability
    • investment
    • simplified fares
    • new express links
4. Core Messages
Overall Narrative

Do not communicate thirty separate changes.
Communicate one story:
“Westbrae is redesigning the network so it works better for today’s travel patterns, provides better value for most passengers and protects the long-term sustainability of local bus services.”

Supporting Message 1
Most customers benefit

“Independent analysis shows around 70% of current passengers will be as well off or better off under the new network.”

Supporting Message 2
Investment is accompanying change

“This is not simply a reduction exercise. We are introducing new express services, simpler fares and better connections.”

Supporting Message 3
Difficult decisions were necessary

“Some routes carry very low passenger numbers and can no longer be sustained in their current form.”
Avoid:

  • “Hard choices.”
  • “Optimisation.”
  • Corporate jargon.
    Be specific and human.
Supporting Message 4
No surprises

“We will provide journey information, community events and individual support well before launch.”

5. Channel Strategy

Given limited resources, focus on channels that directly reach passengers.

Internal
Driver Roadshows

Managing Director and Operations Manager visit every depot.
Purpose:

  • Acknowledge frustration over leak.
  • Explain rationale.
  • Take questions.
    Frequency:
  • Initial briefings.
  • Follow-up sessions at T-3 months.
Weekly Staff Bulletin

Simple format:

  • What has happened
  • What customers are asking
  • This week’s key messages
    Two pages maximum.
Passenger Communications
Dedicated Change Website

Single source of truth.
Must include:

  • Interactive route finder
  • Maps
  • FAQs
  • Fare calculator
  • Timeline
  • Myth-vs-fact section
    All communications point here.
On-Bus Advertising

Most cost-effective channel.
Use:
“New Network Coming Next Spring”
Then progressively introduce specifics.

Direct Passenger Information

Affected route users should receive:

  • Printed leaflets
  • Clear maps
  • Alternative journey guidance
    Avoid forcing elderly passengers online.
Social Media

Purpose is clarification, not argument.
Content types:

  • Route explainers
  • Fare examples
  • Video walkthroughs
  • Questions and answers
    Never engage in lengthy comment wars.
Community Engagement
Local Drop-In Sessions

Particularly in areas losing routes.
Format:

  • Small venues
  • Maps displayed
  • Staff available
    Goal:
  • Listen visibly
  • Explain consistently
Councillor Briefings

Private briefing before public launch.
Not to seek permission.
To prevent surprise.

Media
Media Pack

Provide:

  • Maps
  • Passenger benefit data
  • Independent analysis summary
  • Quotes
  • Case studies
    Make positive angles easy to write.
6. Phasing Plan
Phase 1
Recovery and Preparation
T-8 to T-6 Months

Objectives:

  • Repair internal confidence.
  • Stop speculation dominating.
    Actions:
  • Driver briefings.
  • Stakeholder mapping.
  • Develop FAQs.
  • Build website.
  • Create response protocols.
    Success measure:
  • Rumours decline.
  • Staff understand core rationale.
Phase 2
Public Announcement
T-6 Months

Objectives:

  • Take ownership of narrative.
    Actions:
  • Full announcement.
  • Website launches.
  • Community letters issued.
  • Media interviews.
    Headline:
    “Westbrae announces biggest improvement programme in twenty years.”
    Not:
    “Major route changes announced.”
Phase 3
Consultation and Explanation
T-6 to T-3 Months

Objectives:

  • Increase understanding.
  • Demonstrate listening.
    Actions:
  • Public meetings.
  • Online Q&A content.
  • Route-by-route information.
  • Stakeholder engagement.
    Key principle:
    Don’t debate whether concerns are valid.
    Show people have been heard.
Phase 4
Practical Information
T-3 to T-1 Month

Objectives:

  • Shift from policy discussion to individual journeys.
    Actions:
  • Journey planners.
  • Timetable distribution.
  • Driver refresher briefings.
  • Passenger mailings.
    Main question becomes:
    “What does this mean for my journey?”
Phase 5
Launch
Launch Month

Objectives:

  • Minimise confusion.
    Actions:
  • Extra customer support.
  • Social media monitoring.
  • Senior managers visible on network.
  • Rapid myth correction.
Phase 6
Stabilisation
First 8 Weeks

Objectives:

  • Demonstrate benefits.
    Actions:
  • Publish usage figures.
  • Share customer stories.
  • Highlight fare savings.
  • Show performance improvements.
7. Issues and Risk Management
Risk: “War on Rural Passengers” Headlines

Mitigation:

  • Acknowledge withdrawn routes openly.
  • Lead with overall benefit.
  • Provide alternatives.
  • Use independent evidence.
    Never claim nobody loses out.
Risk: Councillor Campaigns

Mitigation:

  • Brief early.
  • Provide detailed route information.
  • Correct inaccuracies quickly.
    Remain factual and respectful.
Risk: Staff Opposition

Mitigation:

  • Internal-first principle.
  • Managers visible.
  • Regular updates.
    The best external communications campaign will fail if drivers feel ignored.
Risk: Misinformation Online

Mitigation:

  • Dedicated myth-busting section.
  • Standard response templates.
  • Rapid correction within hours, not days.
Risk: Contactless Fare Confusion

Mitigation:

  • Heavy use of examples.
  • Short videos.
  • On-bus signage.
  • “You’ll never pay more than the daily cap” messaging.
8. Measurement Framework
Internal Measures
  • Staff briefing attendance
  • Driver confidence surveys
  • Staff engagement scores
  • Number of internal questions received
Communications Measures
  • Website visits
  • Route lookup usage
  • Video views
  • Social media sentiment
  • FAQ traffic
Media Measures

Track:

  • Positive coverage
  • Neutral coverage
  • Negative coverage
    Success is not eliminating criticism.
    Success is achieving a balanced narrative.
Passenger Measures
  • Contactless adoption
  • Customer complaints volume
  • Call centre enquiries
  • Passenger satisfaction
  • Usage on new express routes
Critical Success Principle

The biggest mistake would be treating this as a publicity exercise. This is a change-management exercise. The deciding factor will not be the quality of press releases; it will be whether drivers, affected communities and regular passengers feel informed, respected and given enough time to adapt.
For a team of two, the priority order should be:

  1. Staff communications
  2. Dedicated information website
  3. Community engagement on affected routes
  4. Passenger journey information
  5. Media relations
    If those five elements are executed consistently, Westbrae can move the debate from “service cuts” to “network modernisation with some difficult trade-offs”, which is the most realistic and achievable communications outcome.

What the judges said

  • GPT: 5.6

    It recognises the leak, driver trust and affected communities, with a sensible sequence from announcement to journey advice. Much of it remains generic, and its measures and interactive tools are insufficiently specified for a working plan.

  • Claude: 5.2

    It correctly identifies drivers as the priority and ends with a sound point about change management over publicity. It remains largely a generic list with no owners, budget or weekly rhythm, and its measures and targets are too vague to show whether it worked.

  • Gemini: 6.8

    Though it correctly identifies the situation as a change-management challenge rather than a publicity exercise, the entry relies too heavily on listed generic tactics rather than arguing for hard, strategic choices. It struggles with practicality, proposing an expansive suite of materials, videos, and roadshows that would likely overwhelm a comms team of two.

  • DeepSeek: 5.4

    Competent and well-organised, with a sensible phase structure and a reasonable channel hierarchy, but it is largely descriptive. Objectives are broad and often unmeasurable ("protect organisational credibility"), the insight into the hostility problem is generic, and the document never quite becomes the working tool a two-person team could pick up and run. It is a solid textbook exercise rather than a strategy with teeth.

  • GLM: 7.0

    A competent, well-structured strategy with a sound internal-first priority, clean phasing, a disciplined no-jargon message set, and a genuinely wise closing principle that this is change management rather than publicity. It is weakened by ceding the public field for two months while petitions built on leaked partial information set the narrative, launching that announcement under an "improvement programme" headline that invites the exact press framing it fears, and listing channels and measures (interactive route finder, sentiment tracking, engagement scores) without owners, baselines or collection methods the stated team could actually operate.

  • Kimi: 6.1

    Sound instincts — drivers as the most trusted channel, "change-management not publicity", never claim nobody loses — but the document is mostly unargued lists in fragmented formatting, with behavioural objectives too vague to steer by. It recognises the hostility problem without ever really engaging with how the petitions, the leak and the press frame interact.

  • Copilot: 7.6 (its own entry)

    This is comprehensive and logically organised, with sensible internal-first communication, audience segmentation and phased journey information. However, the interactive tools, weekly bulletins, roadshows, events and extensive content programme strain two-person capacity, while several success measures lack baselines or precise definitions.

  • Siri: 6.5

    This entry provides a solid, logical framework with a strong emphasis on internal communications and phased messaging, which is highly practical for a small team. However, it lacks specific, actionable detail on how to genuinely mitigate the hostility surrounding the withdrawn routes, relying instead on standard corporate engagement tactics.

  • Human judge (Michael MacLennan): ranked 6th of 8

    Like this but feels more like a slide deck than a written strategy, feels needing fleshed out. Like tiers for audience and communicating one core story. Like mitigations with risks, structure solid

Entry D · GLM Mean 8.8 · 2nd with the AI judges · Silver · 2nd with the human judge · 2,415 words Show

chat.z.ai, GLM-5.3, Deep Think Max

Westbrae Buses — Spring Network Change

Communications Strategy: Eight Months to Launch (+ Six Weeks Stabilisation)

Working document for the comms team. v1.0. T = months until launch day.


The strategy in one paragraph: The leak means the clock has already started, so we tell the whole truth early — to our drivers before anyone else, then to councils and communities, then to the public — and we make sure nobody ever learns about their own route from a newspaper or a Facebook group. We give the local press a better story than the petitions can offer: real people saving real money, two new express routes, and an operator willing to sit in a village hall and take the anger. We will take our worst fortnight of coverage now, on complete information, rather than in spring on partial information.


1. Situation: four facts that shape everything
  1. Partial information is already public and being organised against. Eight months of silence or drip-feeding hands the narrative to others. Full disclosure, early, is now the only defensive play that is also an honest one.
  2. Our most credible spokespeople are currently our most disaffected. Drivers heard from a leak. Every bus is a rolling public meeting: if drivers are sour, hundreds of passenger conversations a day go against us. Internal comms is not a subsection of this strategy — it is the foundation.
  3. The winners are diffuse; the losers are concentrated. The 70% figure is true and useful, and almost meaningless to a village whose bus is going. Anger organises; satisfaction doesn’t. Plan for concentrated opposition however well we communicate.
  4. Two people cannot do everything. Every audience and activity below is tiered, and the bottom tier is a deliberate don’t.
2. Objectives (by launch date)
  1. Every passenger on a withdrawn route knows the change, the date and their alternative before it happens — measured, not assumed.
  2. Drivers are the best-informed audience at every stage: staff hear every announcement 48 hours before the public, without exception.
  3. The public conversation covers the whole package (caps, expresses, revisions), not just withdrawals; negative-led coverage is trending down by T-3.
  4. Every council and both petitioning councillors in affected areas have been offered a full briefing — with the offer on record before we announced.
  5. 80%+ of surveyed passengers can say how the capped fare affects them.
  6. No surprises: every significant fact is published by us before it can be leaked.
3. What comms needs from the business

These are dependencies, not requests. The strategy cannot run without them:

#Decision neededDeadline
1Jobs answer for drivers on withdrawn routes (redeployment, roles, timeline). Do not hold the first depot briefing without it — it will backfire.Week 1
2Mitigation package per withdrawn route — even modest: shuttle link to interchange, demand-responsive service, community transport support, concessionary ticketing on alternatives. If there is genuinely nothing, say so now: the strategy survives, the story gets much harder.Week 2
3Approval to publish the independent fare analysis (summary + methodology). A secret report saying “most people better off” is a worse headline than an open one.Week 2
4Cash policy at and after launch. If cash is accepted during transition, we shout it — it defuses the digital-exclusion attack. If not, we need a hardship plan and it becomes a priority message.T-6
5Launch date and registration/variations timeline confirmed.T-7
6A named route planner as comms liaison (~2 hrs/week, plus attendance at drop-ins).Week 1
7Ten real before/after fare examples for common journeys.T-5
4. Audiences, tiered
TierAudienceWhat they needLead channels
1Drivers & depot staffJob security, facts, scripts, to be heardDepot briefings, fact cards, fortnightly bulletin, named “driver champions”
1Passengers on the 4 withdrawn routes (and materially changed routes)Their specific change, the date, their alternative, a human to askHousehold letters, route cards, drop-ins, dedicated phone line + hub page
2All current passengersWhat changes for me; how capping works; the cash answerOn-bus posters/leaflets, hub, social, press
2Councillors, council officers, MPs, parish/town councils in affected areasThe full picture before the public; honest answers; a route to raise casesEmbargoed briefings, 1:1s, written updates
3Local mediaStories with humans and numbers; access to the MD; fast answersMonthly story drops, exclusives at key moments
3Community Facebook groupsCorrect facts, visiblyMonitor; one factual reply + link; log themes (these groups are also our early-warning radar)
4Non-users / express-route prospectsAwareness that the express existsPress coverage and geo-targeted paid only — no active programme
5. Key messages

Core narrative: One network, honest choices: simpler capped fares that most passengers will pay less under, faster express links, and straight answers about the routes too few people use.

Four pillars (each with proof points to be filled from ops data):

  1. Simpler and cheaper for most. Tap and go with automatic daily/weekly caps; you’ll never pay more than the cap. Independent analysis published in full: seven in ten passengers pay less. Real examples: [journey A: £X → £Y].
  2. Faster where travel is growing. Two new express routes — [town] to [town] in [X] minutes — and twelve routes revised to serve the hospital, colleges and employment sites people actually travel to.
  3. Straight answers for affected communities. We will tell you plainly what’s changing, why, and what your alternatives are — in person, in your town, before it happens.
  4. Built to last. [X]% of the network is unchanged or improved. The four withdrawn routes carried [X] passengers last year; keeping them as they were puts the routes most people rely on at risk.

Tone rules: Plain English — ban “network optimisation” and never call a withdrawal an “enhancement.” Say the hard thing first: the first sentence we give the press will be the headline, so make it the honest one. Never dismiss a petition or petitioner. One voice: MD on controversy, comms lead on process, drivers never volunteered as spokespeople on withdrawals.

Lines to take (the four questions we will face constantly):

  • “70% better off — so the rest of us lose?” → “Seven in ten passengers pay less under the caps. For those facing bigger changes, the honest position is: route [X] can’t be sustained as it is — and here is exactly what we’re doing instead.”
  • “War on rural passengers” → “We’re adding two express routes and revising twelve to run where people actually travel. We’re not walking away from those communities — [mitigation] — but we won’t pretend the old network was working.”
  • “Why should we trust contactless?” → Cap explanation + cash answer + hands-on help at travel clinics.
  • “Why did drivers hear from a leak?” → “That shouldn’t have happened, and it’s on us. From now on, staff hear everything first.”
6. Phasing

Phase 1 — Stabilise and seize the narrative (Weeks 1–4, T-8 to T-7)

  • Days 1–3: holding statement with a date certain for the full announcement (“Full details on [date]”), so we stop the vacuum without drip-feeding. Factbook v1 (single source of truth, numbered, versioned).
  • Week 1–2: depot briefings at every depot — jobs answer in hand, Q&A, driver fact cards, feedback channel, and a pledge: you will never learn about your job from the press again.
  • Week 2: embargoed briefings for council officers, all councillors (not just the two critics), parish clerks, MPs, colleges, major employers. Written 1:1 offers to the two petitioning councillors — the offer itself goes on record.
  • Week 3: full public announcement. MD on the record; before/after network map; per-route one-pagers; published analysis summary; mitigation detail; timeline for full timetables. Hub page live. On-bus posters the next morning. Letters to every household on withdrawn routes. Analyst available for interview.
  • Brief the board now: this fortnight will be our worst coverage. We chose it deliberately. Early pain on full facts beats spring pain on half-facts.

Phase 2 — Show up (T-7 to T-4)

  • Drop-in sessions in every affected town (minimum 8), plus the roadshow bus parked in market squares on market days — free venue, visible, photogenic, on-brand. Route planner present at every session.
  • Consult honestly with clear boundaries: network shape and withdrawals are decided; timings, stop siting, connections and mitigation design are genuinely open. Publish a “you said, we did” summary at the end — fake consultation is worse than none.
  • Fortnightly driver bulletin; monthly positive press drop (driver profiles, fare-cap savings stories with real consented families, first look at the express routes).

Phase 3 — The new offer (T-4 to T-2)

  • Fare cap campaign built on real examples (“the most you’ll ever pay is £X”). Contactless education plus travel clinics — in libraries and on the roadshow bus — for anyone anxious about cards, phones or cash.
  • Timetable publication (~T-6 weeks, aligned to registration deadlines): second major press moment, and a positive one — new maps, faster journeys, named planners (“meet the person who designed your route”).

Phase 4 — Countdown (final 6 weeks)

  • On-bus saturation and SMS/email alerts: “Your route, your date.” Geo-targeted paid social to affected postcode districts only.
  • MD does a local radio phone-in — accept it, front it, do it honestly.
  • Launch-week plan: help points at key interchanges (requires ops staffing — dependency), daily 8am standup, rapid-response cell, driver launch pack with de-escalation guidance (“hand them the card, contact the office, we’ve got your back”).

Phase 5 — Live and fix (launch + 6 weeks)

  • Daily issues log in week one; fix-forward notes published (“changed today: …”). Six-week review. Post-implementation report to every council — demonstrating we did what we said is the best long-term political protection this company can buy. Transition to business-as-usual comms.
7. Channels — and what we won’t do

The bus itself is our biggest medium: captive, relevant, cheap. Priority order: on-bus print → hub page (one URL, everything routes there) → household letters → drop-ins/roadshow → press → Facebook groups → radio → small geo-targeted paid.

Press rhythm: one exclusive at the announcement (to the largest outlet — exclusives buy goodwill), then a monthly drop, then the timetable moment. Never an unattributable briefing — “sources said” stories feed the conflict narrative.

We will not: run a brand campaign; chase national media; open new social platforms; argue in comment threads (one factual reply, then link and log); attend every council meeting (brief them all in writing; attend in person only in the three most affected areas); or promise mitigation that doesn’t exist.

8. Risk and issues handling
RiskResponse
Further leaksCompress the timeline (announcement week 3); numbered factbook; tiny distribution list
Driver morale spiral / attritionInternal-first rule for all eight months; job clarity; visible MD at depots; driver champions
High-profile hardship case (hospital, school, care)Empathy-first protocol: no templates, no glibness, MD signs every individual response, ops fixes fast
”Secret report” / 70% weaponised as “30% lose”Publish the analysis; per-route impact sheets so people check their own situation, not an average
Council motions / political escalationBrief the whole council membership, not just critics; officers aligned before members
Contactless failure or cash passengers stranded at launchTransition comms, clinics, interchange help points, pre-approved technical holding statement
Timetable slippage looks like chaosBuffer built in; if it slips, say so early — slippage announced is survivable, slippage discovered is not
Team overload / illnessThis document’s tiers are the triage; materials bank pre-approved in advance

Protocol: Severity 1 (routine) — comms lead handles. Severity 2 (media/political) — MD informed same day, joint line. Severity 3 (crisis) — MD leads, board informed. First-hour rule on social issues. Write six holding statements now, while calm.

9. Running it
  • Comms Lead: strategy, press, MD and political engagement, internal briefings, risk.
  • Comms Officer: hub and digital, community groups, events logistics, print, measurement, the issues log.
  • Rhythm: Monday 30-minute plan with the ops liaison; Friday one-page dashboard; monthly MD review; fortnightly driver bulletin.
  • Budget split: print & on-bus 35% · events & venues 20% · household letters 10% · geo-targeted paid 15% · survey/tools 5% · contingency 15%.
10. Measurement

Baseline (weeks 2–4): on-bus pulse survey (~300 responses, six questions), driver pulse, six-month press audit.

MetricTarget
Withdrawn-route passengers aware of change + alternative70% at T-2; 85% by launch week
Passengers who can state their fare impact80% by launch
Negative-led local coveragePeaks at announcement, below 25% of stories by T-2, majority neutral/positive by launch
Drivers feel informed / confident answering passengers80% / 70% (pulses at T-6 and T-1)
Councils & both petitioning councillors offered briefings100% (the offer is the KPI, not their acceptance)
Post-launch: contactless share, express ridership vs forecast, complaint volume declining by week 4Reviewed at six weeks against objectives
First fortnight — checklist
  • Days 1–3: holding statement approved and issued; factbook v1; jobs answer chased; 1:1 letters to both councillors.
  • Days 4–7: all depot briefings complete; embargoed stakeholder pack issued; announcement date locked.
  • Days 8–10: public announcement (press + hub + posters + household letters); MD broadcast round; analyst interviews.
  • Days 11–14: first drop-in booked and advertised; baseline survey fielded; dashboard v1 to the MD.

A closing note to the MD: comms can win the argument about facts; only the network can win the argument about consequences. The single biggest determinant of how the spring goes is whether the mitigation package for those four routes is real. Everything in this document is built to make an honest story land well — it cannot invent one.

What the judges said

  • GPT: 7.5

    It has sharp insight into concentrated opposition and makes the business’s mitigation decisions explicit. Its roadshow, household letters and many events are demanding, and the conflicting announcement dates weaken its value as a working schedule.

  • Claude: 8.6

    Framing business decisions as dependencies (the jobs answer, a mitigation package, permission to publish the analysis), choosing to take the worst coverage early, and giving specific lines to take show strong insight tied to practical tiering, a budget split and measurable targets. The only small weakness is that it sometimes treats "better off" as simply "pay less", and the drop-in programme may stretch two people.

  • Gemini: 9.0

    By bravely front-loading the negative news to seize control of the narrative, this strategy shows profound insight into the mechanics of local outrage and driver morale. It is highly practical, clearly tiering audiences to protect the two-person team's capacity and demanding specific operational dependencies before communications can succeed.

  • DeepSeek: 8.7

    Excellent strategic judgement: the decision to take the worst coverage early, on complete information, is the sharpest single insight in the field, and the dependencies section—jobs answer, mitigation package, cash policy, analysis publication—shows a rare understanding that comms cannot succeed without operational decisions. The tiering and "what we won't do" discipline are strong. It is slightly less detailed than Entry G on the mechanics of running the team week to week, which is the only thing keeping it from the top score.

  • GLM: 9.3 (its own entry)

    Choices are argued rather than listed: the decision to take the worst coverage early on complete information, the inviolable internal-first rule, and the explicit "what we won't do" tiering all follow directly from its four situational facts, while the dependency table honestly names what comms cannot control. Slightly optimistic awareness targets and a heavy events calendar for two people are the only practical strains.

  • Kimi: 9.0

    The sharpest strategic thinking of the set: "take our worst fortnight now", a jobs answer required *before* the first depot briefing, honest consultation boundaries, and KPIs like "the offer is the KPI, not their acceptance" that show real understanding of political hostility. It is fully runnable — roles, rhythm, budget split, severity protocol — with only marginally less operational depth than the most exhaustive entry.

  • Copilot: 9.4

    Exceptionally coherent and operational, connecting the leak, concentrated opposition and driver trust to explicit dependencies, triage choices, owners, budgets, protocols and measurable outcomes. A few activities, particularly eight-plus events, a roadshow bus and intensive stakeholder handling, may still require more borrowed operational capacity than the working plan fully secures.

  • Siri: 8.8

    A very strong, realistic strategy that correctly identifies the need to take "early pain" by fully disclosing the facts rather than letting partial information dictate the narrative. It is highly coherent and practical, though slightly ambitious in the sheer volume of face-to-face engagement expected from a two-person team.

  • Human judge (Michael MacLennan): ranked 2nd of 8

    Like that this has measurement, first fortnight and note for MD at end, good breakdown on who running, pillars for key messaging, tiered audiences, starts strongly with summary and situation – a solid basis for getting going

Entry E · Siri Mean 5.3 · 8th with the AI judges · 8th with the human judge · 642 words Show

Siri app, iOS 27 beta

Here is a practical, eight-month communications strategy designed for a two-person team to manage the upcoming network changes at Westbrae Buses. It focuses on neutralising early opposition, rebuilding driver morale, and clearly explaining the benefits of the new capped contactless system to the majority of your passengers.
Communications Strategy: Westbrae Network Redesign
OBJECTIVES

  • Rebuild Internal Trust: Ensure 100% of drivers and frontline staff are fully briefed and feel heard before any further public announcements.
  • Reframe the Narrative: Shift the local press focus from “rural cuts” to “modernisation and passenger savings.”
  • Educate on Value: Achieve a 75% understanding among regular passengers of how capped contactless fares will save them money.
  • Mitigate Vocal Opposition: Provide clear, factual counter-narratives to local councillors regarding the four withdrawn routes, focusing on alternative solutions.
    AUDIENCES & KEY MESSAGES
AUDIENCEKEY MESSAGE
Drivers & StaffYou are the face of this change. We are giving you the tools and information first so you can confidently answer passenger questions.
The 70% (Better Off)Your daily commute is getting faster, easier, and cheaper with new express routes and capped fares.
Affected CommunitiesWe understand the impact of withdrawing these four routes. We are working on community-led alternatives and demand-responsive solutions.
Local CouncillorsWe are modernising a network that hasn’t changed in twenty years to ensure long-term sustainability for the whole region.

CHANNEL CHOICES

  • Internal: Weekly driver briefings, a dedicated Q&A inbox, and “lunch and learn” sessions with the Managing Director.
  • External (Digital): A dedicated landing page on the website with a “Fare Calculator” to show passengers exactly how they will be better off.
  • External (Print): Simple, bold on-bus posters and take-away leaflets focusing on the new express routes and contactless benefits.
  • Community: Town hall meetings in the four affected communities, specifically inviting the petitioning councillors to discuss alternatives.
    PHASING (EIGHT MONTHS TO LAUNCH)
  • Months 1–2: Internal Focus & Stabilisation
    • Brief all drivers and staff immediately. Apologise for the leak and share the full, unvarnished plan.
    • Launch an internal FAQ document.
    • Meet privately with the two petitioning councillors to share the full data behind the route withdrawals.
  • Months 3–5: The “Better Off” Campaign
    • Launch the public website with the Fare Calculator.
    • Begin local press briefings, leading with the “70% better off” statistic and the new express routes.
    • Host town hall meetings in the affected communities.
  • Months 6–7: Education & Countdown
    • Heavy on-bus advertising for capped contactless fares.
    • “How-to” videos on social media showing how the new fare system works.
    • Final press push highlighting the launch date.
  • Month 8: Launch & Support
    • “Street teams” at major hubs to help passengers with the new contactless system.
    • Daily monitoring of social media and press to quickly correct misinformation.
      RISK HANDLING
  • Risk: Local press runs “War on Rural Passengers” headline.
    • Mitigation: Proactively pitch a feature on the new express routes and the sustainability of the network. Provide case studies of passengers who will save money.
  • Risk: Drivers remain demoralised and unhelpful to passengers.
    • Mitigation: Incentivise them. Create a “Change Champion” program with rewards for drivers who actively help passengers adapt to the new system.
  • Risk: Councillors continue to agitate.
    • Mitigation: Invite them to co-design the demand-responsive alternatives for their areas. Give them a win to take back to their constituents.
      MEASUREMENT OF SUCCESS
  • Internal: 85% of staff report feeling “informed” or “very informed” in a pre-launch survey.
  • External: Positive or neutral sentiment in 60% of local press coverage.
  • Adoption: 50% of passengers using capped contactless within the first month of launch.
  • Engagement: High usage of the online Fare Calculator (target: 5,000 unique uses).

What the judges said

  • GPT: 4.5

    It connects staff, passengers and councillors to relevant channels, but treats opposition mainly as a narrative problem. The fare calculator, street teams and ambitious targets lack a credible plan for a two-person team.

  • Claude: 3.8

    It covers the headings, but objectives such as "neutralise" opposition and a "modernisation" reframe underestimate the hostility, and it gives almost no reasoning for its choices. Measurement targets look arbitrary, there are no owners or capacity planning, and the risk responses are thin.

  • Gemini: 5.5

    This entry reads too much like a textbook exercise, proposing resource-heavy tactics like street teams and incentive programs that ignore the strict two-person, modest-budget constraints. It lacks a nuanced understanding of the hostility problem, optimistically assuming that vocal opposition can be neutralized simply by highlighting the benefits to other passengers.

  • DeepSeek: 4.6

    Covers the expected headings but reads as a template rather than a working document. Objectives are asserted without a credible route to measurement, channels are listed rather than argued, and the insight into the hostility problem is shallow: town hall meetings and "give them a win" do not grapple with the fact that the decisions are painful and the leak has already damaged trust. It would need substantial development before a two-person team could run it.

  • GLM: 6.8

    Covers all required elements in a sensible order (staff first, campaign, countdown, launch) with some genuinely smart touches — giving councillors a co-design win, apologising to drivers for the leak — but choices are more listed than argued, and several items (street teams, a rewards programme, a fare-calculator build, verifying "75% understanding") strain a two-person team and modest budget. Reliance on town halls, posters and digital over direct mail under-serves the rural, older households at the centre of the hostility problem, and measurement targets like 5,000 calculator uses arrive without basis.

  • Kimi: 6.4

    Coherent, complete and deliverable by two people, with sensible phasing that puts drivers first; but most choices are listed rather than argued, the hostility problem gets standard mitigations rather than real analysis, and targets like "5,000 fare calculator uses" are asserted rather than reasoned.

  • Copilot: 5.9

    It provides a clear basic framework and recognises the need to repair staff trust, explain fares and engage affected communities. It remains too generic for execution, relies on unconfirmed alternatives, and treats councillors and drivers somewhat transactionally through “neutralising”, “giving them a win” and rewards rather than addressing substantive concerns.

  • Siri: 4.5 (its own entry)

    This strategy feels more like a textbook exercise than a working document, relying heavily on generic marketing tactics like "lunch and learn" sessions and "street teams" that are impractical for a two-person team. It significantly underestimates the hostility of the situation and fails to provide a realistic approach to the withdrawn routes.

  • Human judge (Michael MacLennan): ranked 8th of 8

    Most bare and generic

Entry F · Kimi Mean 8.4 · 4th with the AI judges · 3rd with the human judge · 1,799 words Show

kimi.com Plus, K3 High

Westbrae Buses — Network Change Communications Strategy

Period: October → May (launch) | Team: 2 comms staff + MD sponsorship | Status: Working document v1


1. Situation in one paragraph

We’re announcing the biggest network change in 20 years from a position of weakness: the story is already half-told by a leak, drivers heard it third-hand, two councillors are running petitions, and the press has a ready-made “war on rural passengers” frame. The honest truth is that the package is good for most passengers — 70% better off, capped contactless fares, new express routes — but four communities will lose their bus. No amount of spin fixes that. Our strategy is therefore not to “sell” the changes but to reclaim the narrative through radical transparency, fix the internal breach of trust first, and give losing communities something real rather than spin.

2. Strategic principles
  1. Inside before outside. No further public announcement until staff are briefed face-to-face. Every external message fails if a driver contradicts it at the bus stop.
  2. Tell the whole story once, in public, in writing. Publish the full analysis — including the 30% worse off and the four withdrawn routes — before anyone else leaks it. The leak wounded us because it was partial; the antidote is completeness.
  3. Never argue with grief; mitigate it. For the four withdrawn routes, lead with what we’re doing about it, not why they’re losing the service.
3. Objectives

By launch day:

  1. Awareness: ≥85% of current passengers know changes are coming and where to check their route (measured via on-board and online survey waves).
  2. Understanding: ≥60% of passengers correctly identify whether they are better/worse off (survey).
  3. Trust: Staff confidence score (“I understand the changes and can answer passenger questions”) ≥70% by month 6, from a baseline taken in month 1.
  4. Containment: Majority of local media coverage from month 3 onward is neutral or positive in tone; zero further leaks.
  5. Adoption: ≥50% of fares paid contactless within 3 months of launch; cash-paying passengers given a clear, well-used alternative.
  6. Political: No formal council motion opposing the package passes; at least one petitioning councillor moves to neutral (“we got a fair hearing”) by month 5.
4. Audiences (priority order)
AudienceWhat they care aboutPrimary channels
Drivers & frontline staffJob security, timetable changes, being embarrassed in front of passengersDepot briefings, manager cascade, printed FAQs
Passengers on the 4 withdrawn routes”How do I get to work/shops/the GP?”Door-to-door leaflet, village meetings, parish councils
The 70% better offLittle — they don’t know it yetOn-bus, social, local radio
Councillors & MPsPostboxes full of complaints; looking effectiveDirect briefings, written offers, one-to-ones
Local press & radioA clean, verifiable story; accessPress pack, briefings, named spokesperson
Older / cash-paying / non-smartphone passengersContactless anxiety, digital exclusionPrint, libraries, community groups, drivers
Community & accessibility groupsVulnerable users falling through gapsDirect outreach, forums
5. Key messages

Master narrative (all audiences):

“We’re rebuilding the network around how people actually travel today. It’s the biggest improvement in 20 years: seven in ten current passengers get a faster, cheaper, or more frequent service, and fares are capped so no one pays more than £X a day with contactless. Four routes with very low usage are being withdrawn — we know that’s a real loss for those communities, and we’re working with them directly on alternatives.”

Tailored variants:

  • Drivers: “You heard about this wrong, and that’s on us. Here’s the full picture, here’s what it means for your rota, and here’s what to say when passengers ask.” (Job security question must be answered explicitly and honestly in the first briefing — whatever the truth is.)
  • Withdrawn-route communities: “Your route carried an average of X passengers a day and costs £Y per passenger journey. We can’t pretend that’s sustainable — but here’s what we’re putting in its place / exploring with you.” Lead with alternatives, not the deficit.
  • Contactless anxious: “Nothing changes about how you board. Cash still works [if true — see risk 4]. Contactless just means you’ll never pay more than the daily cap, automatically.”
  • Press/councillors: “Everything is published. Here’s the data, here’s the analysis, here’s the person to call.”

Message discipline: the 70% figure must always be sourced (“independent analysis by [consultant]”) and never used to dismiss the 30%. Banned phrases: “only four routes,” “the majority benefit,” “improvements across the board.”

6. Channels (chosen for a 2-person team — depth over breadth)
  • Depot briefings (15 min, every depot, run by depot managers with a script we write). Highest-leverage channel we own.
  • A single “Big Change” webpage as the source of truth: route-by-route “what changes for me” lookup, full analysis PDF, FAQ, timeline. Every other channel points here.
  • Printed leaflets/posters on buses, at stops, libraries, GP surgeries, post offices — essential for the cash-and-no-smartphone audience.
  • Two local radio interviews (MD, not comms staff) at announcement and launch — radio over-indexes in rural areas and is harder to frame negatively than print.
  • Proactive press pack + named spokesperson (the MD). One comms person owns all media contact; the other owns the webpage and social.
  • Facebook only for social (rural demographics), plus parish council newsletters and community noticeboards for the four affected communities — free, trusted, local.
  • What we are NOT doing: no advertising campaign, no Twitter/X presence, no app, no roadshow bus. Can’t staff them.
7. Phasing

Phase 0 — Stop the bleeding (Weeks 1–2, October)

  • Baseline staff and passenger surveys (short, cheap — clipboard + online).
  • Brief MD and depot managers; agree the honest answer on jobs/rotas.
  • Quiet one-to-ones with the two petitioning councillors: acknowledge the leak, offer a full private briefing before anything public. Don’t ask them to drop petitions — offer them the information first.
  • Build the webpage, press pack, FAQ, and driver script in the background.

Phase 1 — Staff first (Weeks 3–4)

  • Depot briefings at every location, MD attends at least one session per depot. Driver FAQ cards printed. Feedback logged; questions answered in writing within 5 working days.

Phase 2 — Full public announcement (Week 5, early November)

  • Publish everything: full network map, route-by-route impacts, the independent analysis (warts and all), and the mitigation offers.
  • Press pack out under embargo; MD radio interview same day.
  • Brief all councillors/MPs 24 hours ahead, in writing.

Phase 3 — Community engagement (November–January)

  • One public meeting per withdrawn-route community, held in the village, evening, MD or ops director present. Format: 10 min honest presentation, 40 min listening, 10 min on alternatives.
  • Explore mitigations in public: demand-responsive transport pilot, community transport partnership, revised school/shopper services, minibus contracts. Even one saved or replaced service changes the whole story.
  • Monthly “you said / we did” updates on the webpage.

Phase 4 — Build the positive (February–March)

  • Shift messaging to the gains: new express route previews, fare-cap explainers (“a week of commuting for £X”), worked examples.
  • Driver ambassadors: volunteers filmed/quoted on why the changes help passengers.
  • Countdown begins on-bus: “Your route changes in 8 weeks — check yours.”

Phase 5 — Launch (April–May)

  • 6-week countdown with route-specific alerts.
  • Contactless transition push: demonstrations at bus stations, driver prompting, helpline number on every poster.
  • Launch week: MD visible at key interchanges; press invited to ride the new express routes with real passengers.
8. Risk handling
RiskLikelihoodResponse
Petition story escalates to regional/national mediaMediumDon’t fight the petition; invite the councillors into the mitigation process publicly. If they refuse, that becomes visible. Hold a standing “response statement” ready.
Another leakMediumPhase 2 publishes everything, so there is nothing left to leak. Internally: brief managers on why, not just what; leaks come from excluded people.
Driver hostility / industrial frictionMedium-HighUnion engagement from Phase 0, not Phase 3. Rotas and job questions answered in writing. Never let drivers learn timetable specifics from a passenger.
”Contactless = war on the elderly” storylineHighDecide early and state plainly whether cash is retained. If yes, lead with it. If no, prepare a serious answer (onboard top-up cards, shop-based top-ups) before announcement — this is the story most likely to beat the route withdrawals in the press.
Meeting turns hostile / viral clipMediumMD chairs, stays calm, never says “only X passengers.” No filming restrictions (looks guilty). Publish our own full recording afterwards.
Analysis is challenged (“70% is rigged”)MediumPublish the methodology and the consultant’s name at announcement. Pre-brief one trusted local journalist on the numbers.
Mitigation for withdrawn routes fails to materialiseMediumNever promise what isn’t funded. Commit to “exploring with you,” report honestly. An honest “we tried, here’s what we found” beats a broken promise.
Comms team overwhelmedCertain at peaksPre-write everything in Phases 0–2 (statements, FAQs, social posts). MD does the broadcast media. Deprioritise social replies — point to webpage.
9. Measurement
  • Monthly pulse (months 1–7): 100-passenger on-board/online survey — awareness, understanding, sentiment. ~2 hrs/month to run.
  • Staff: confidence survey at months 1, 4, 7; question volume from driver FAQs as a proxy for confusion.
  • Media: tone log of every article/broadcast (one comms person, 30 min/week) — target neutral-or-better majority from month 3.
  • Digital: webpage visits, route-lookup usage, FAQ search terms (free analytics).
  • Political: petition signature trajectory; councillor statements logged.
  • Launch: contactless adoption %, complaint volumes vs. baseline, helpline call themes.
  • Review: 30-minute team stand-up every Monday against this document; formal review at end of each phase — adjust channels and messages based on the pulse data, not on whoever shouted loudest that week.
10. What success looks like on launch day

The change happens, it isn’t a surprise to anyone, drivers are answering questions correctly rather than complaining with passengers, at least one withdrawn-route community has a real alternative in place or in trial, the petitioning councillors are on record saying the process was fair even if they disagree with the outcome, and the launch-day headline is about the new express routes and fare caps — not the withdrawals.


First actions this week: agree the honest internal answer on jobs/rotas, book depot briefing dates, and request the baseline surveys. Everything else sequences from those three.

What the judges said

  • GPT: 7.3

    The staff-first approach, direct work with withdrawn-route communities and limited channel selection show sound judgement. It assumes the remaining 30% are worse off and sets political and leak-prevention targets outside the comms team’s control.

  • Claude: 7.8

    The principles are sharp and well argued ("never argue with grief"), it spots the contactless and cash risk as a potential bigger story than the withdrawals, and it keeps channels realistically lean for two people. Some objectives, such as no hostile council motion or a councillor moving to neutral, sit outside the team's control, and the budget and role detail is lighter than a working document needs.

  • Gemini: 8.8

    Entry F delivers a highly coherent, no-nonsense strategy that beautifully aligns its messaging and channel choices with the reality of an under-resourced team. Its insight into the hostility problem is superb, correctly identifying that the company must transparently mitigate grief rather than trying to spin route withdrawals as network optimizations.

  • DeepSeek: 7.9

    A clear, well-structured and genuinely practical strategy, with strong principles—inside before outside, tell the whole story once, never argue with grief—and a realistic channel set for a two-person team. The insight into contactless anxiety and the "war on the elderly" risk is valuable. It is a little less rigorous than the best entries on measurement baselines and phasing precision, and the political handling of the petitioning councillors is thinner than Entry G's.

  • GLM: 9.3

    This is a genuinely runnable strategy: three stated principles drive every subsequent choice — channels are explicitly scoped to a two-person team with named trade-offs and an explicit "not doing" list, and the measurement plan is lightweight, mostly free, and tied to a weekly review cadence. The insight is the standout — reframing the leak's damage as a problem of partiality ("the antidote is completeness"), refusing to argue with grief, banning dismissive phrasings, flagging unverified facts rather than assuming them, and correctly identifying contactless/digital-exclusion anxiety as the story most likely to out-run the route withdrawals — with the only real deductions being a few objectives (zero further leaks, no council motion) that sit partly outside the team's control.

  • Kimi: 8.4 (its own entry)

    Tight, honest and well-sequenced, with real insight in treating completeness as the antidote to the leak and in flagging the "contactless = war on the elderly" storyline as potentially bigger than the withdrawals. It loses a little on the mechanics of running the plan — no budget shape and thinner week-to-week operating detail than the strongest entries.

  • Copilot: 8.4

    Strong insight drives a coherent plan built around transparency, internal repair, mitigation and disciplined channel choices, with good attention to cash-paying passengers and team overload. Marks are lost for several outcomes outside communications control, including zero leaks, preventing a council motion and converting a councillor, plus recurring surveys that may be demanding for two staff.

  • Siri: 9.2

    This is an exceptional, highly practical strategy that directly confronts the reality of the leak and the hostility with a radical transparency approach. It demonstrates outstanding insight by prioritising internal trust and offering a highly coherent, phased plan that a two-person team can actually deliver.

  • Human judge (Michael MacLennan): ranked 3rd of 8

    Like that it's chosen channels for 2-person team (depth over breadth), that it ends with a concise summary of what success looks like on launch day, perhaps overly concise but it strong and sharp

Entry G · Claude Mean 9.1 · 1st with the AI judges · Gold · 1st with the human judge · 4,989 words Show

claude.ai incognito chat, Opus 5.5 Medium

Westbrae Buses: Network Change Communications Strategy

Summary

We stop defending a leak and start leading a change: drivers first, the full plan published within three weeks, and the four withdrawn-route communities handled face to face rather than broadcast at.
The story is currently being told by other people, using partial information. That is the real problem, and it is fixable. Seven in ten passengers will be better off, but nobody is hearing that, because the only version in circulation is the one about loss. The strategy has five working principles:

  1. Staff before public, every time. Drivers heard about the leak before we told them. From now on every milestone reaches depots at least 48 hours before it reaches anyone outside the company. Drivers are our largest and most trusted channel; they cannot be our weakest.
  2. Publish everything, early. Partial information fuels petitions. We release the full plan, route by route, with the independent analysis, by week three. Nothing a councillor can “reveal” should be news.
  3. Personal, not promotional. Passengers do not care about “the network”. They care about the 07:40 from their village. Every channel answers one question: what changes for my journey?
  4. Go to the four communities; do not broadcast at them. Withdrawn routes get the MD in the room, named alternatives for every regular journey, and a real feedback window with the power to change details before the timetable is registered.
  5. Don’t lead with the 70 per cent. It is our best proof point for press and councillors, and our worst opening line for someone whose bus is being taken away. It supports the story; it is not the story.
    Core proposition: Simpler fares, faster links and a network built around how people travel now, with a plan for every passenger, including those whose route is changing most.
    This document is written for a team of two. Section 8 says who does what each week; section 9 has the lines to use when things go wrong.
Situation

We start behind: a hostile frame is already set, internal trust is damaged, and we have eight months, two people and a modest budget to reverse both.
What we have going for us

  • A plan that most passengers will benefit from, backed by independent analysis we can publish.
  • Capped contactless fares: a concrete, easy-to-grasp gain that lands in every pocket.
  • Two new express routes: a positive, visual story the press can photograph.
  • Eight months: enough time to do this properly if we move in the first three weeks.
    What is against us
  • The leak. Partial information is worse than none: it lets opponents fill the gaps. Two councillors now own the narrative and have petitions as a campaigning tool.
  • The four withdrawn routes. Small numbers of passengers, but high visibility, strong local identity and a ready-made press frame: “war on rural passengers”.
  • Driver morale. Drivers heard the news second-hand. They are asked questions by passengers every day and currently have no answers, or worse, the councillors’ answers.
  • Capacity. Two people cannot run a mass-engagement programme alone. We will have to borrow people and time from operations and customer services.
    Strategic choices that follow
We willWe will not
Publish the full plan and analysis ourselves, earlyWait for launch, or drip-feed “good news” first
Treat drivers as the primary channel and brief them firstRely on a notice on the depot board
Separate the withdrawn-route conversation from the wider network storyBury withdrawals inside a positive launch release
Engage the two councillors directly and early, in personFight them in the press or on social media
Admit what is worse for some peopleClaim everyone wins
Offer a real feedback window on detailsPretend the core decisions are open if they are not

The last row matters most. If the withdrawals are final, we say so plainly and consult on mitigation, not the decision. A fake consultation is the fastest way to turn a local grievance into a lasting one.

Objectives

Five objectives, each tied to a number we can actually collect with our own resources; targets are proposals to agree with the MD once baselines are in.

#ObjectiveTarget by launch (unless stated)
1Rebuild driver confidence so drivers can explain the changes and feel informed80% of drivers say they can explain the changes on their routes (pulse survey, month 6); zero milestones reach drivers after the public
2Make every passenger aware of what changes for them75% of regular passengers correctly name the change to their own route (onboard survey, weeks before launch)
3Give every withdrawn-route passenger a workable alternative100% of affected communities offered a face-to-face session; every identified regular journey matched to a named alternative or documented as unresolved
4Shift the public narrative from “cuts” to “a better network, with care for those losing out”Our case included in at least 80% of local press items; no story published without our response; neutral-or-better tone in at least 60% of coverage by month 6
5Land the change smoothlyContactless share of fares up by an agreed margin within 8 weeks; launch-month complaints no more than 50% above baseline and falling by week 4

Two things we are deliberately not setting as objectives: stopping the petitions (we cannot, and trying will look defensive) and winning over the two councillors (desirable, but outside our control). We aim instead to make the petitions less necessary: if people have the full facts and a way to be heard, fewer will sign, and signatures will carry less weight with the press.

Audiences

Three tiers, prioritised by how much the change hurts them and how much they can shape the story; tier 1 gets personal contact, tier 3 gets well-made mass communication.

TierAudienceWhat they needWhat we want them to doOwner
1Drivers and depot staff (incl. union reps)The full picture first, answers to passenger questions, an apology for the leak, a say in operational detailExplain changes confidently; feed back problems; stop hearing news from the pressComms Lead + MD + depot managers
1Passengers on the four withdrawn routesHonesty, a named alternative for their actual journeys, a human to talk toAttend a session; tell us about journeys we have missed; try the alternativeComms Lead + MD
1The two petitioning councillorsTo be taken seriously and not surprised againReceive the full facts; work on mitigation with us, even while campaigningMD (with Comms Lead)
1Council transport officers and the regional transport partnershipData, timelines, early sight of every changeHelp with mitigation (tendered or demand-responsive services); act as calm voicesMD
2Passengers on the twelve revised routesClear before/after for their stop and timeCheck their journey; know launch dateComms Officer
2Other councillors, community councils, MSPs/MPsA briefing pack and a named contactAnswer constituent queries accurately; not be caught outComms Lead
2Local press and radioFacts, access to the MD, real people, picturesReport our case alongside criticismComms Lead
2Journey-critical partners: schools, GP surgeries, hospital, colleges, care homes, older people’s groupsWhat changes for their users and whenPass information on to people we cannot reachComms Officer
2Customer services teamScripts, FAQs, escalation routeGive consistent answers; log themesComms Officer
3Passengers who gain (most of the 70%)Simple news: cheaper, capped, fasterSwitch to contactless; spread the wordComms Officer
3Potential new passengers on the express corridorsA reason to try the busTry an express journey in launch monthComms Officer

Groups needing specific care within the tiers: concessionary pass holders (the fare cap means little to them, so do not lead with it), people without bank cards or smartphones (explain what happens to cash and tickets), and people with disabilities whose stop or route changes. Each needs its own FAQ entry and a route into customer services that does not depend on the website.

Core narrative and key messages

One story told the same way by the MD, drivers, customer services and the website: why we are changing, what gets better, who loses out and what we are doing about it.
The narrative (for internal use, about 100 words)
Our network was designed twenty years ago, and the way people travel has changed. We have spent [period] looking at how people actually use our buses, and independent analysis shows the new network will leave seven in ten passengers better off. Fares will be simpler and capped, so nobody pays more than [£X] a day. Two new express routes will cut journey times between [towns]. Twelve routes will run in patterns that match demand. Four routes carry so few passengers that we cannot keep running them as they are. That is a hard decision for those communities, and we are working with each of them on what replaces it.
Message pillars

PillarMessageProof points
Simpler, fairer faresTap on, and you will never pay more than [£X] a day or [£Y] a weekWorked examples by town; typical saving per week
Faster linksNew express services connect [town] and [town] in [n] minutesBefore/after journey times
A network that fitsTwelve routes redesigned around when and where people travelUsage data; stop-level before/after
Honest about lossFour routes are being withdrawn; here is what each community will have insteadNamed alternatives; mitigation partners; session dates
Open and listeningEverything is published; tell us what we have missed before [date]Feedback channels; “you said, we did” updates

Messages for the withdrawn-route communities
These need particular care. Language to use and avoid:

SayAvoidWhy
”Route [n] carried an average of [x] passengers a journey""Low-usage route”Real numbers are more persuasive and less dismissive
”We know this makes some journeys harder""Nobody will be worse off”It is untrue, and one counter-example destroys trust
”Here is how you can still get to [hospital, town, school]""Alternative provision is available”Specific beats abstract
”The decision is made; the details are not""We are consulting” (if we are not)Protects credibility
”We should have told you ourselves, and we are sorry you heard it this way”Silence about the leakAcknowledging the leak removes a grievance

Messages for drivers

  • “You should have heard this from us first. You didn’t, and that was our failure.”
  • “From now on you will hear every change before the public does.”
  • “You know these routes better than anyone. Tell us what we have got wrong before timetables are fixed.”
  • Practical: what changes for rosters, duties and depots, and when. If there are job or shift implications, they must be stated plainly here, early. I do not know if there are, and this is a question for the MD.
    The 70 per cent statistic
    Use it with press, councillors and in the published analysis. Always pair it with the honest counterpart: “and for the rest, here is what we are doing.” Do not put it on posters in the four affected communities. We also need to know who the other 30 per cent are: if most are simply neutral rather than worse off, say so, as it changes the story considerably.
Channels

The bus itself is our biggest channel: drivers, stops and the inside of every vehicle reach more passengers each week than any paid media we could afford.

ChannelRoleAudienceEffortNotes
Driver briefings (depot sessions, pocket Q&A card, WhatsApp or app broadcast)Primary internal channel; drivers become messengersDrivers, then passengersMediumShort sessions at shift changes, not one big meeting. Card updated at each phase
Onboard and at-stop printRoute-specific before/after; launch date; where to get helpAll passengersHigh (logistics)Stop panels in March to April; onboard posters rotate monthly. Large type
Door-to-door leafletsTown-by-town “what changes here”Tier 1 and 2 passengers, non-digital householdsMedium, biggest costOne version per town or corridor, not one generic leaflet. Timed for Nov and April
Website: journey checker and full planSingle source of truth; analysis published in fullEveryone, press, councillorsMediumA simple route or stop lookup beats a PDF. Could be a searchable table if budget is tight
Community drop-insFace-to-face listening in the four withdrawn-route communities and major townsTier 1 communitiesHigh (people time)Village halls, a bus parked in the square on market day. MD attends the four withdrawn-route sessions
Councillor and stakeholder briefingsNo surprises; accurate information for people who get askedCouncillors, MSPs/MPs, officers, community councilsLow to mediumBriefing pack updated each phase; 48-hour embargoed sight of announcements
Local press and radioProactive announcements, MD interviews, real passenger storiesWider publicMediumRegular relationship with the two or three key editors, not just releases
Facebook, especially local community groupsWhere rural conversation happens; correcting misinformation calmlyTier 2 and 3Medium, ongoingPost in groups where admins allow; answer questions factually; never argue
Partner networks (GP surgeries, libraries, schools, pharmacies, older people’s groups)Reach people who do not read our channelsVulnerable and non-digital passengersLow per partnerProvide posters and a short email they can forward
Customer services (phone, email)Answers for people who need a personEveryone, especially older passengersMediumScripts and FAQs updated every phase; log themes weekly
Email, SMS, app (if we hold customer data)Direct notice to known users of affected routesRegistered customersLowCheck what data and consents we hold
Targeted paid socialPromote express routes and fare cap at launchTier 3, potential passengersLowSmall spend, launch month only

What we will not do

  • A glossy launch event before the withdrawals are settled. It would become the photograph for the “war on rural passengers” story.
  • Argue with the councillors or petition organisers online. We correct facts once, politely, and link to the full plan.
  • Radio or print advertising campaigns. The budget is better spent on leaflets and stop panels.
  • A single generic leaflet. It is cheaper, and it fails the “what changes for me” test.
Phasing

The first three weeks decide the whole campaign: if drivers are briefed and the full plan is public by the end of October, we are explaining rather than defending for the next seven months.
[embed: node/23487045-f587]
The registration deadline is the hinge: feedback that arrives after it cannot change the timetable, so the listening window must close early enough for planners to act on it.

PhaseKey actionsDone when
1 Reset (October, weeks 1 to 3)Week 1: MD and Comms Lead visit every depot at shift changes, apologise for the leak, share the full plan and a driver Q&A card, invite driver champions. MD phones both petitioning councillors and council transport officers and offers an in-person briefing. Week 2: briefing pack to all councillors, MSPs/MPs and community councils, 48 hours before publication. Week 3: publish the full plan, route-by-route tables, the independent analysis and an FAQ; brief local editors; MD interview. Capture all baselines.Every depot briefed before publication; plan live; baselines recorded
2 Explain (late October to December)MD-led sessions in each of the four withdrawn-route communities, logging every hard case. Open drop-ins in the main towns, including a bus parked on market day. First town-specific leaflet drop with the feedback deadline. Journey checker live. Partner packs to GP surgeries, schools, libraries and older people’s groups. Customer services scripts. December driver pulse and onboard awareness survey.Four sessions held; leaflets delivered; every hard case logged with an owner
3 Listen and adjust (January to mid March)Feedback window closes at the end of January (confirm against the registration date). Planning reviews every submission. Mitigation for withdrawn routes agreed in writing with the council. “You said, we did” published in February, covering changes made and not made, with reasons. Return visits to the four communities to present mitigation. Routes registered. March driver pulse.Changes published; routes registered; mitigation confirmed
4 Prepare (mid March to mid May)Final timetables printed. Stop panels installed a week ahead with “from [date]” banners. Onboard posters. Second leaflet drop with final times. Contactless explainer covering the cap, cash and what to do without a card. Driver training days. Customer services rehearsal. Launch volunteers recruited. April awareness survey. Press ride on an express route in early May. Go or no-go meeting two weeks out.Ticketing tested, panels in, drivers trained, lines agreed
5 Launch (late May, two weeks)Staff in tabards at 10 to 15 key stops for the first week. Daily 09:00 stand-up with operations and customer services. Monitoring twice a day; fast-fix route for errors. MD rides the network, including the replacement services. Launch story led by passengers who gain, with an honest line on the withdrawals. Paid social for the express routes and fare cap.Complaints peak then fall; no press query unanswered
6 Settle (June to July)Eight-week review with ticketing and complaints data. Report back publicly and to each of the four communities. Thank drivers publicly, by name where they agree. Log adjustments for the next registration window. Lessons-learned session.Review published; next adjustments scheduled

Running it with a team of two
Two people can run this only if they split the work cleanly, borrow people from elsewhere in the business, and protect a fixed weekly rhythm.
Division of roles

Comms LeadComms Officer
Strategy, MD liaison and sign-offContent production: leaflets, posters, web, social
Councillors, officers, MSPs/MPsJourney checker and website updates
Press relationships and interviewsSocial media monitoring and replies
Withdrawn-route communities (with the MD)Partner networks and customer services FAQs
Driver engagement (with depot managers)Print and distribution logistics
Issues and crisis handlingMeasurement: surveys, logs, monthly dashboard

People to borrow (agree these with the MD in week one)

  • Driver champions: one or two per depot, volunteered not appointed, given an hour a month and first sight of materials. They answer peer questions and tell us what drivers are really saying.
  • Depot managers: run the shift-change briefings from a pack we supply.
  • Customer services lead: weekly 15-minute call on complaint themes.
  • Planning or scheduling lead: the source of every route fact. Nothing goes out without their check.
  • Office staff volunteers: a pool of 10 to 15 people for launch fortnight to stand at key stops in branded tabards.
    Weekly rhythm
WhenWhatWho
Monday, 30 minPlan the week; review last week’s logBoth
TuesdayPress and stakeholder callsComms Lead
Wednesday, 30 minMD check-in: decisions, sign-offs, risksComms Lead + MD
ThursdayContent published; social replies batchComms Officer
Friday, 15 minCustomer services and driver champion themesComms Officer
Daily, 15 minSocial and press monitoring, morning and afternoonComms Officer (Lead covers)

Sign-off: agree now that the Comms Lead can approve routine content and reactive press lines alone, with only major announcements going to the MD. Slow sign-off is the most common reason small teams lose press cycles.
Budget priorities (proportions, since I do not know the figure)

ItemShare
Route-specific leaflets, printing and door-to-door distribution35%
Stop panels and onboard print25%
Journey checker and web updates15%
Community sessions (halls, refreshments, accessibility)10%
Launch paid social and express route promotion10%
Contingency for reactive print or corrections5%

If the budget cannot stretch to door-to-door drops everywhere, prioritise the four withdrawn-route areas and towns on the most-changed routes, and rely on stop panels elsewhere.

Risk handling

The most likely damaging story is not about the network at all: it is one named, sympathetic person who can no longer get to hospital, so we find those people before the press does.
Risk register

RiskLikelihoodImpactEarly warningResponse
Human-interest story from a withdrawn route (older passenger, hospital or school journey)HighHighComments at drop-ins; Facebook posts; councillor mentions a named caseLog every hard case at sessions. Contact them personally before they are quoted. Offer MD phone call; show the specific alternative. Respond to press with the case-specific fix, never a generic line
Further leaks of timetable or internal detailMediumMediumJournalist asks about unpublished detailPublishing everything early removes most leak value. Keep drafts to a small list. If it happens, publish the full version same day
Petitions presented with a media eventHighMediumCouncillor announces hand-in dateOffer to receive it in person; MD attends, listens, answers briefly, invites organisers to the next session. Never decline to accept
Driver discontent becomes public or turns into industrial relations issueMediumHighChampion feedback; union rep contact; social posts by staffMD depot visits in week one; early answers on rosters. Loop in HR and union reps before any roster changes are announced
Factual error in our own materialsMediumMediumCustomer services or driver flagsPlanning sign-off on all route facts. Correct publicly and quickly; thank whoever spotted it
Contactless or ticketing failure at launchMediumHighTesting results in AprilAgree a fallback now (e.g. accept payment later, fares waived for the day). Prepared line and driver instruction ready before launch
Stop panels and timetables wrong or late on day oneMediumHighPrint and installation trackerInstall one week ahead with “from [date]” banners; spot-check sample stops; staff on key stops
Political escalation (MSP/MP, regional council motion)MediumMediumParliamentary questions; council agendaMD offers briefing in advance; send full data; keep officers warm
Social media pile-onMediumLow to mediumSpike in mentionsOne factual reply with a link; move individuals to direct message or phone; do not delete critical comments

Holding lines (adapt, do not read verbatim)

  • On the withdrawals: “Route [n] carried an average of [x] people a journey. We could not keep running it as it was, and we know that is hard for the people who rely on it. We are working with each community on what replaces it, and we want to hear about any journey we have not yet solved.”
  • On the leak: “People should have heard this from us, in full, and we are sorry they didn’t. Everything is now published at [web address].”
  • On “war on rural passengers”: “Most of our passengers live in small towns and villages, and most of them will pay less and travel faster. For the four communities losing a route, we are putting in [alternatives] and meeting every one of them.”
  • On a named hard case: “We have spoken to [name] and here is how we are helping with that journey.” Only with their consent.
    Escalation: Comms Officer spots, Comms Lead decides the response within two hours in working time, MD is told immediately if a story involves a named passenger, a staff member, safety or a politician. Agree now who covers evenings and weekends in the launch fortnight.
Measurement

We measure with tools we already have or can run cheaply, report monthly on one page to the MD, and treat leading indicators as a steering wheel rather than a scorecard.

ObjectiveMeasureSourceFrequencyTarget
1 Drivers”I can explain the changes on my routes” and “I feel informed”Five-question pulse survey (paper at depots and online)Months 1, 3, 680% agree by month 6
1 DriversMilestones reaching depots before the publicComms logEach milestone100%
2 AwarenessPassengers who can name their route’s changeOnboard intercept survey, 200 to 300 passengers across routes, run by volunteersMonths 3 and 775% by month 7
2 AwarenessJourney checker visits; leaflet reachWeb analytics; distribution countsMonthlyTrend only
3 Withdrawn routesCommunities met; journeys matched to alternativesSession log; hard-case trackerMonthly4 of 4 met; every logged case resolved or explained
4 NarrativeShare of coverage including our case; toneManual coverage log (one spreadsheet row per story)Weekly log, monthly summary80% include our case; 60% neutral or better by month 6
4 NarrativePetition signature growth after the full plan is publishedPetition pagesWeeklySlowing growth after week 3
5 LaunchContactless share of transactionsTicketing dataWeekly from launchAgreed uplift within 8 weeks
5 LaunchComplaints and enquiries by themeCustomer services logDaily in launch fortnight, then weeklyPeak no more than 50% above baseline, falling by week 4
5 LaunchPatronage on express routes and revised routesTicketing dataWeekly from launchTracked against planning forecast

Baselines to capture in October: current complaint volume, contactless share, coverage tone over the last three months, and a quick driver pulse. Without these, none of the targets above means much.
The monthly dashboard is one page: five objectives, a traffic light each, three things that went well, three concerns, and decisions needed from the MD. Anything longer will not be read, and a team of two cannot afford to write it.

Assumptions and questions for the MD

This plan rests on assumptions I could not check; the answers to the questions below could change the phasing and several key messages.
Assumptions

  • Launch is late May 2027, eight months from now.
  • New registrations must be lodged with the Traffic Commissioner well ahead of launch (typically 70 days, plus earlier notice to the local transport authority). I have assumed a mid-March deadline; please confirm the exact date with the planning team, as it fixes the end of the feedback window.
  • The four withdrawals are decided; timings, stops and connections on revised routes can still change.
  • The fare cap applies across the network, and cash will still be accepted.
    Questions
  • What is the confirmed launch date and registration deadline?
  • What mitigation is realistically available for the four withdrawn routes: council-tendered replacement, demand-responsive transport, community transport partners, rerouted nearby services?
  • Is anything about the withdrawals open? If a council were to fund a reduced service, would we run it?
  • Who are the other 30 per cent: worse off, or simply no different?
  • Are there roster, shift or job implications for drivers? Is there a recognised union, and have reps been briefed?
  • What was leaked, how, and could it happen again? (Not to find someone to blame, but to know what else is out there.)
  • What customer data and contact consents do we hold (app, smartcard, email)?
  • Who is the spokesperson: the MD for everything, or an operations director for detail?
  • What is the budget figure, and can we borrow staff time as set out in section 8?
  • What are the fare cap levels and when will they be final?

What the judges said

  • GPT: 8.5

    It explains why early disclosure, driver briefings and personal journey advice matter, then assigns work and identifies staff to borrow. The plan is crowded for two comms staff, and some launch coverage and complaint targets depend heavily on events outside their control.

  • Claude: 8.8 (its own entry)

    Objectives, audiences, messages, channels and measures all connect tightly, with clear owners, a weekly rhythm, borrowed staff, a budget split, and honest measurement based on tools the team can actually use. The handling of the 70 per cent figure, the "decision made, details not" line, and hunting for hard cases before the press finds them show strong insight, although its length risks weighing a little on a team of two.

  • Gemini: 9.4

    This is a remarkably insightful and cohesive working document that understands the danger of leading with positive statistics when engaging grieving communities. It is supremely practical, offering a precise division of labor for the two-person team, clear rules on what not to do, and a phased timeline that logically prioritizes staff trust and total transparency.

  • DeepSeek: 9.2

    Outstanding. It recognises that the central problem is not the changes but the vacuum of partial information, and it argues every choice rather than listing it: publish everything early, staff before public, personal not promotional, and separate the withdrawn-route conversation from the wider good-news story. The practicality is exceptional—named owners, a weekly rhythm, borrowed staff, budget proportions and holding lines—and the measurement is honest about what cannot be controlled (petitions, councillors) while still tracking what matters.

  • GLM: 9.4

    Objectives, tiered audiences, channels, phasing and measurement interlock tightly: five working principles drive every later choice, the timeline is hinged on the registration deadline, and the plan is honestly runnable by two people (role split, weekly rhythm, borrowed staff, budget proportions, baselines, a one-page dashboard). Insight is argued rather than listed — it treats the leak as the core wound, refuses to lead with the 70 per cent, distinguishes consulting on mitigation from a fake consultation, and hunts the named hard cases before the press finds them.

  • Kimi: 9.2

    The "we will / we will not" table, the instruction not to lead with the 70 per cent, the registration-deadline hinge and the borrowed-people model show choices genuinely argued and built around the team's real capacity, with measurement tied to baselines a small team can actually collect. Only the broken embed artefact and a length that slightly strains its own "team of two" discipline keep it from the top of the range.

  • Copilot: 9.2

    This is an unusually complete working document, with excellent connections among priorities, tailored messages, operational dependencies, phasing, staffing, budget and low-cost measurement. Its main weakness is scale: the extensive channel mix, multiple surveys, visits, hard-case management and detailed monitoring depend heavily on borrowed staff and strong cross-business cooperation.

  • Siri: 9.2

    This strategy demonstrates exceptional insight into the human reality of the brief, correctly identifying that partial information and driver morale are the core problems to solve rather than just "communicating the plan." Its phased approach is intensely practical for a small team, grounding every action in the operational reality of a regional bus operator.

  • Human judge (Michael MacLennan): ranked 1st of 8

    Good that it lists assumptions and questions for MD at end, good measurement, considered risk and holding lines, named phases, really like division of roles, owners for audiences great touch, begins with a strong summary (I'm reading up from the bottom for these!)

Entry H · GPT Mean 8.5 · 3rd with the AI judges · Bronze · 4th with the human judge · 2,195 words Show

ChatGPT Plus, Work tab, GPT-6 Sol High

Westbrae Buses: network change communications strategy

Working plan | Eight months to launch | Owners: Managing Director (MD), Communications Lead (CL), Communications Officer (CO)

The job

Help people understand exactly how their journeys and fares will change, give affected communities a real chance to influence decisions that remain open, and equip drivers to answer questions confidently. Communications cannot make the withdrawal of four routes painless. Westbrae must show what it has done to reduce the harm and be candid where no workable alternative exists.
First decision for the MD: approve a one-page statement of what is final, what is proposed, and what evidence or funding could change a proposal. Do this before calling any engagement a consultation. Give the two-person comms team named contacts in network planning, operations, fares and customer service who can answer questions and correct errors.

Objectives and tests
ObjectiveTest of success
Restore trust after the leakPublish the full, verified picture promptly; correct known errors in the leaked account; keep a dated public record of changes to the plan.
Reach people most affectedBy one month before launch, at least 80% of people sampled in the four withdrawal areas know the change date and can identify their available travel option, or know whom to call if none exists.
Prepare staffBrief every driver and customer-facing colleague before the full public announcement. In a short anonymous check, at least 90% say they know where to find the latest route and fare answers by one month before launch.
Make the new network usableEvery affected route has an accurate old-to-new journey guide, timetable and stop notice at least six weeks before launch; test common journeys with passengers before publication.
Handle concerns visiblyAcknowledge enquiries within two working days; answer straightforward questions within five. Publish a “you said, we did, we could not” account after engagement closes.
Reduce launch confusionTrack wrong-stop, wrong-time and fare-cap complaints by route each day during launch week; correct recurring problems the same day.

The targets are management goals, not claims about likely public opinion. Establish a baseline during the first month and adjust any target that cannot be measured credibly.

Audiences and channels
Audience and needPrimary contact
Passengers on withdrawn routes: whether they can still reach work, health care, shops and connectionsDirect conversations at stops and community venues; notices on buses and at stops; a simple journey-by-journey leaflet; staffed phone line. Use local partners to reach people who are not online.
Passengers on revised routes: their new stop, time and connectionRoute-specific web pages and printed timetables; on-bus cards; stop notices; journey planner and third-party data updates.
Fare-paying passengers: what the cap means for their travelA plain-English fare guide with worked examples, exclusions, payment instructions and what to do if a charge looks wrong. Test it with people who have never used contactless.
Drivers and customer-facing staff: reliable answers and a route for raising problemsDepot briefings on paid time, a two-page change sheet, a searchable answer sheet and a daily correction channel near launch. Depot managers own attendance.
Councillors, local authorities and community leaders: full evidence and a route to challenge itA common briefing pack and scheduled meetings, including with petition organisers. Record requests and responses. Do not ask them to defend Westbrae’s decision.
Schools, employers, health providers and accessibility groups: effects on essential journeysTargeted calls using mapped journey impacts; invite practical alternatives and test proposed connections with them.
Local media: clear facts and accountable spokespeopleThe same public evidence pack, route maps and named spokesperson. Offer a visit to an affected community with the MD. Answer the strongest criticism directly.

The website is the single current source. Each printed item carries a version date, phone number and short web address. The phone line and printed guides remain essential for rural passengers. Provide large-print and other accessible formats on request; check language needs with local partners rather than assuming them.

Message framework
  1. What is changing: “Next spring, twelve routes will change, four are proposed for withdrawal, two express routes will begin, and capped contactless fares will be introduced.” Use proposed only where a decision is genuinely open.
  2. Why: Show the demand, cost, reliability and journey evidence for each route. Publish enough of the independent analysis for others to understand its method and limits.
  3. Who benefits and who loses: “Independent analysis estimates that 70% of current passengers will be better off under the package.” Before using this line, verify its denominator, whether “better off” means fare, time or access, and the result for each affected community. Follow it immediately with the impact on those who lose a service.
  4. What Westbrae is doing about harm: Give a specific answer for each withdrawn route: viable connections, timetable adjustments, community transport discussions or funding options. Label unconfirmed options as such.
  5. What people can influence: State the questions open for comment, the deadline, who decides, and when Westbrae will publish reasons for its decisions.
  6. What to do next: Point each person to their route guide, fare example or a human adviser.
    Tone rule: Lead with the journey a person may lose before talking about aggregate benefits. Never describe a withdrawn route as “low usage” without acknowledging the people who depend on it. Do not promise a replacement until an operator, funding and start date are confirmed.
Eight-month timetable
WhenAction and outputOwner
Now: first 72 hoursMD meets drivers and depot managers, acknowledges they heard the leak first and explains the decision status. Issue a short public statement confirming the scale of change, the date for a full pack and how urgent journey concerns can be raised. Contact both petitioning councillors and all other relevant representatives with the same verified facts. Set up one question log.MD; CL
Months 8–7Audit the 70% claim and route-level impact, including essential journeys and accessibility. Create an old-to-new route matrix and mark every fact as final, proposed or unverified. Publish the full proposal and four withdrawal-area fact sheets together. Start baseline staff and passenger checks.Planning and fares leads; CL
Months 7–6Hold a time-limited listening period, with sessions in all four withdrawal areas at usable times and an option to respond by phone, post or online. Ask about specific trips and workable alternatives. Log each substantive issue, owner and response date.CL; CO; planning lead
Month 5MD and planning lead decide and document changes. Publish a response report: what changed, what did not, why, and any still-unresolved access gap. Brief staff first, then councillors, partners, passengers and media within the same day.MD; planning lead; CL
Months 4–3Build route-by-route journey guides and fare examples. Test at least five common journeys on each changed route, including a connection and a concessionary or non-contactless case where relevant. Correct maps, stop lists, payment wording and journey-planner data.CO; operations and fares leads
Month 2Publish final timetables and route guides; begin sustained stop, bus, community and local-media notices. Target the four withdrawal areas with direct distribution and staffed sessions. Run a driver briefing at each depot.CO; depot managers
Month 1Repeat awareness and staff checks; fill gaps. Test the contactless cap with real-world scenarios, the customer help process and every public journey-planning channel. Run a launch-day rehearsal for a missed connection, inaccurate data and payment failure.CL; operations and fares leads
Launch weekStaff a daily issues meeting and update one public “latest changes” page. Put extra help at affected interchanges and stops. Correct significant errors across web, stops, staff guidance and data feeds together.Operations lead; CL
First four weeks after launchPublish weekly fixes and a four-week review covering actual journey and fare problems, especially in withdrawal areas. Commit to any timetable corrections with dates and owners.MD; planning lead; CL

The team should not run sixteen separate campaigns. Use one master route-and-fare fact sheet, from which the website, driver sheet, press answers, leaflets and stop notices are produced. CL owns the facts, stakeholder contact and approvals; CO owns publication, distribution and the enquiry log. Network planning supplies route answers, operations supplies staff time and stop access, fares supplies tested payment rules, and customer service handles individual cases. The MD approves decisions and is the public face of difficult trade-offs.

Risks and response rules
TriggerResponseLead
Another leak or false route claim spreadsCheck the claim with planning within two hours. Publish a dated correction in the common question log and give the same correction to drivers, councillors and media. Do not speculate about a source.CL
Petition or hostile headlineAcknowledge the underlying journey concern; offer the route evidence and a meeting; name the practical options being tested. The MD answers questions about withdrawals. Track factual errors for correction, not criticism for rebuttal.MD; CL
Evidence shows someone may lose access to an essential tripEscalate to planning and the relevant authority within one working day. Record the journey, possible mitigation, decision owner and deadline. Communicate the outcome directly to those affected where possible.Planning lead
Drivers report wrong information or cannot answer questionsCorrect the master sheet first, then all channels. Give depot managers a same-day update and a short script for unresolved cases.CO; operations lead
Contactless cap fails or charges appear wrongStop promotional claims about savings affected by the fault. Publish the affected payment types, interim advice, refund route and update time. The fares lead owns the fix; comms owns clarity.Fares lead; CL
Timetable, stop notice or journey-planner data disagreeTreat as a passenger-impact incident. Operations confirms the operating plan; CO corrects owned channels; the data owner updates external feeds and checks that changes appear.Operations lead
Engagement cannot change a proposalSay so before inviting views. Offer a route for evidence about harm and mitigation, and publish the decision rationale.MD

A useful opening response to the leak is: “Some details of our planned network changes have circulated before we could brief our drivers and communities. We are sorry. The information is incomplete. We will publish the full route-by-route proposals, their reasons and the ways to respond by [date]. We know the proposed withdrawals raise serious concerns, and we want to hear about the journeys people rely on.”

Measurement and weekly control

CO maintains a simple dashboard, reviewed for 30 minutes each week by CL, planning, operations and customer service:

  • Reach: leaflet delivery and stop-notice checks by route; visits to route pages; calls from affected areas. Reach figures show contact, not understanding.
  • Understanding: short, consistent questions at stops and by phone: “Does your usual journey change?”, “When?”, “What will you do instead?” Record the sample size and locations.
  • Staff readiness: briefing attendance, anonymous two-question confidence pulse, unresolved driver questions.
  • Response quality: enquiries by topic and route, answer time, corrections made, and hardship cases still open.
  • Operational reality: missed-connection and wrong-information complaints per 10,000 journeys, cap disputes, refunds, and any documented loss of access to essential trips. Compare with a pre-launch baseline where possible.
  • Decision accountability: number of material suggestions received, adopted, rejected with reasons, or awaiting a named decision.
    Escalation rule: any issue affecting an essential journey, safety, accessibility or repeated incorrect fares goes to its operational owner the same day. A rising complaint count may mean Westbrae has finally reached people who were previously unheard; investigate the cause before treating it as a communications failure.
Immediate sign-offs needed

Within one week, the MD should confirm the launch date, jurisdiction and registration timetable; the true status of each route decision; the independent analysis and permission to publish its method; named operational owners; a modest print, outreach and contingency budget; and protected depot briefing time. If these are not available, publish a clear interim timetable for providing them. Avoid announcing precise alternatives or savings that the business has not verified.
External practice check: Transport Focus advises early engagement, targeted publicity for affected passengers and visible consideration of responses in bus service reviews. For an operator in England or Wales, network changes also require a separate registration and local-authority process; in England, timetable and fare data must be kept current for journey-planning services. These operational requirements should be checked against Westbrae’s actual jurisdiction and arrangements. Transport Focus · Traffic Commissioners’ guide · Bus Open Data Service guidance

What the judges said

  • GPT: 9.1 (its own entry)

    Objectives, audiences, route-level messages, owners and measures connect tightly, especially for passengers losing essential journeys. It handles uncertain decisions honestly and gives the small team a shared fact base and clear operational support.

  • Claude: 7.6

    It is rigorous about process integrity: separating final from proposed decisions, verifying the 70 per cent claim, keeping a single master fact sheet, and publishing what was not changed and why. The practicality and honesty are excellent, but the approach to press and hostility is more procedural than persuasive, and the full plan appears later than the leak situation calls for.

  • Gemini: 8.5

    This entry demonstrates exceptional insight into the hostility problem, matching a highly practical, master-asset-led approach to the strict realities of a two-person team. Its strategic coherence shines through stringent tone rules and measurable, management-focused objectives that acknowledge the impossibility of making route withdrawals painless.

  • DeepSeek: 8.1

    The most rigorous entry on decision status, evidence integrity and measurement design, with a welcome insistence on distinguishing what is final, proposed or unverified before any consultation language is used. It is genuinely practical for two people, with a single master fact sheet feeding all channels. It is marginally weaker on the emotional and political craft of the hostility problem—the withdrawn-route communities are handled correctly but with less empathy than Entry G—and the external-practice citations, while diligent, feel slightly textbook.

  • GLM: 9.2

    An unusually disciplined and honest plan: the final/proposed/unverified distinction anchors everything from engagement validity to message wording, a single master fact sheet feeds every channel, and the measurement section is candid about what reach does and cannot prove. It loses a little on practicality through sheer density, and it never surfaces the driver jobs question that should gate the first depot briefing.

  • Kimi: 8.2

    Outstanding on integrity and operability: the final/proposed/unverified fact discipline, the demand to verify the 70% claim before using it, journey-testing, one master fact sheet and same-day escalation rules are exactly what a two-person team needs. What it lacks is a comparably strong press and narrative strategy — the media handling is correct but passive, ceding ground the brief's hostile environment makes costly.

  • Copilot: 9.6

    The strategy is exceptionally disciplined, ethically credible and practical, distinguishing genuine consultation from fixed decisions while tying every audience and channel to journey-level outcomes and operational ownership. Measurement is notably honest and actionable, though the external regulatory material adds slight complexity to an otherwise tightly focused working plan.

  • Siri: 8.0

    A highly detailed and rigorously structured document that excels in its practical approach to measurement and risk management. It demonstrates excellent strategic coherence, though its slightly bureaucratic tone and complex matrix structure might be a bit heavy for a small, two-person comms team to operationalise quickly.

  • Human judge (Michael MacLennan): ranked 4th of 8

    First decision up front, broken down well, like risks and response rules, also measurement and weekly control, some helpful research and URL at end