How to test a marketing message before you launch it
To test a marketing message before launch, show it to people like your audience and check three things: whether they understand it, whether they believe it, and whether it gives them a reason to act. You can do that with interviews, a message-testing survey, a preference test, a small paid experiment, or a rehearsal with a synthetic audience. The quick methods find problems; the slower ones measure.
Written by MiroFish.co.uk, which offers one of the methods described here, synthetic audience studies.
What a message test should tell you
- Clarity. Can people say back what you offer and who it is for?
- Credibility. Do they believe the claim, and what proof do they ask for?
- Relevance. Does it speak to a problem they have, in their words?
- Difference. Can they tell you apart from the alternatives?
- Risk. Could anyone read it in a way you did not intend?
Five ways to test a message
Interviews
Show the message to a few people from the audience and ask them to explain it back. Slow, but the best way to hear the words people use.
Message-testing surveys
Show each respondent one version (a monadic test) or several in turn (a sequential monadic test), then ask about clarity, believability and intent. You get numbers to compare, if the sample is right.
Preference and first-impression tests
Show a headline or a page briefly and ask what people remember, or which of two versions they prefer. Quick and cheap, but shallow.
Small paid experiments
Run two versions as ads or landing pages on a small budget and compare clicks or sign-ups. This measures behaviour, but it needs traffic, time and a clean comparison, and it tests in front of the market.
A synthetic audience rehearsal
Give the message to a simulated audience built around your segments and let it discuss it over several rounds. It shows quickly where people misread it, what they doubt and how segments differ, including how critics and journalists might react. It is directional, not a measurement.
Which method answers what
| Best for | Real people | Typical time | Main limit | |
|---|---|---|---|---|
| Interviews | Language and reasons | Yes | Days to weeks | Small numbers |
| Message-testing survey | Comparing versions with numbers | Yes | Days | Stated reactions; sample quality |
| Preference test | Quick first impressions | Yes | Hours | Shallow |
| Paid experiment | Real behaviour | Yes | Days to weeks | Needs budget and traffic |
| Synthetic audience | Objections, misreadings and segment differences | No | Minutes to hours | Directional only |
A worked example
In the campaign message example study, a message about rebuilding the middle class, built on a minimum wage rise, childcare subsidies and small-business tax credits, was read by suburban swing voters, urban progressives, rural conservatives, young first-time voters, journalists and opposition strategists over ten rounds. The childcare element carried the message for swing voters, young voters noticed what was missing, rural audiences split by policy, and the study listed the counter-messages to expect and the changes to make before launch.
A message-testing checklist
- Write the message as people will see it: the headline, the subhead and the first line of the body.
- Name the segments that matter, including the sceptical ones.
- Rehearse it and list the misreadings and the doubts.
- Rewrite, then test the two strongest versions with real people.
- Launch the stronger version, and keep what the other one taught you.
Questions people ask
What is message testing?
Checking how an audience understands and reacts to a message before you commit budget to it: whether it is clear, believable, relevant and different.
What is the difference between monadic and sequential monadic testing?
In a monadic test each respondent sees one version, so versions are compared across groups. In a sequential monadic test each respondent sees several versions in turn, which needs fewer people but lets earlier versions colour the later answers.
Can AI test a marketing message?
A synthetic audience can show likely misreadings, doubts and segment differences quickly. It cannot tell you how many real people will respond, so use it to sharpen a message before you test it with real people.
What is creative concept testing?
Testing an advertising idea, its message and its visual approach, before production, usually by showing a rough version to people from the target audience.