Message Test
Test how your message lands.
Paste a draft, choose who is receiving it, and see what they actually hear.
You get a read on how it lands, the objections it will meet, what the draft leaves out, which trust signals are there, and a tightened rewrite that uses only your facts.
Already have a plan? Plan Comms With AI builds the brief. Message Test checks the draft that comes out of it.
New here? See it work in one click:
What this is, and what it is not. Message Test is an assessment of how an audience of the type you chose is likely to receive your draft, based on communications practice. It is not research, and it does not predict what any particular real person will think or do. Treat it as a second pair of eyes, not a verdict.
Where your words go. What you paste is sent to Claude, the Anthropic model that produces the assessment. We do not store it, and it is not used to train AI models under Anthropic's API terms. As with any AI tool, leave out names and anything confidential → Using AI safely
Guests get 1 test a day. A free account gets 2, no card needed. Members test without limits (fair use). Allowances reset each day.
Reading it as your audience…
The assessment starts appearing in a few seconds.
Assessed as
What they hear
The message received, which is often not the message written.
What they push back on
Each objection with an answer you could actually give.
What is missing
What this audience needs that the draft does not give them.
Trust signals
Evidence, named accountability, specificity, sourcing.
A tightened rewrite
Built only from the facts in your draft. Anything in [square brackets] is a gap for you to fill, never a figure to take on trust.
What next
Message Test is built and tested by Michael MacLennan. Tell us what worked and what didn't: send feedback.
That didn't work.