article Transform
How we build: app development ground rules
When we build AI tools for communications, we're not trying to create the next unicorn startup. We're solving real problems that comms professionals face every day.
When we build AI tools for communications, the aim is to solve problems comms professionals face every day (unicorn status is not on the cards).
Our Building Philosophy
Every app we create starts with a real pain point. Whether it’s drafting briefing documents, checking tone consistency, or generating media lists, we only build what we’d actually use ourselves.
We use simple, accessible tools – the sort of thing that you can use with minimal or no coding knowledge. The aim is to demonstrate concepts quickly and transparently. If something takes more than a weekend to prototype, we’re probably overcomplicating it.
We’re also looking to be handy with our budget: where possible, we’ll use cheaper – and ideally free – tools and platforms, also to make it something which is more accessible to replicate, remix, and be inspired by.
Staying Independent
We don’t take money from tool providers or platforms. When we use OpenAI’s API, Anthropic’s Claude, or any other service, we pay standard rates like everyone else. No special access, no affiliate kickbacks.
Any code we develop is available upon request; just get in touch. You can see exactly how we built something, fork it, improve it, or ignore it entirely. We’d rather share what works than create dependencies.
What You Can Expect
Honest failure reports alongside the successes. When an experiment flops (and we’re sure many will), we’ll tell you why. Even when something works brilliantly, we’ll endeavour to explain the limitations we discovered.
Real usage metrics. How long did it take to build? How much did it cost to run? What wasted hours of our time? We’ll share the unglamorous details.
No overselling. These are experiments and prototypes, not production-ready solutions. We’ll always flag when something is rough around the edges or suitable only for specific use cases.
The goal is to support human expertise. See our ethics policy for more on our stance around this. Every tool we build assumes you know your job – and need a better way to do parts of it.
More like this
articleTransform
Leaving Substack or Ghost without losing a reader: an AI-assisted migration checklistMoving a website off a platform like Substack or Ghost used to be real digital comms gruntwork. In our case, moving a publication with dozens of pages onto a new site took a few days rather than weeks – with Claude doing most of the building and me being the decision-maker for every switch. Here's what worked, what went wrong, and what you can put into practice today.
experimentGovern
No, Claude Fable 5.2 isn't AGI: five things it built for comms work, and the 15 errors it found in oursThe morning after Anthropic released Claude Fable 5.2, Michael asked it for its five most impressive pieces of comms work. It built a crisis simulation you can play and a 3D film made in Blender from code alone, then flagged 15 errors in our own archive.
interviewTransform
Three numbers: Stephen Waddington on making the case for comms in 20272027 planning rounds are asking comms teams to turn their AI use into savings. Stephen Waddington's answer: make the case in the three numbers management cares about, price the judgement work, and ask for budget for the new work AI creates.