Skip to content

Latest US poll: 63% say gas prices are affecting their household budget a great deal or a moderate amount

See the poll

Message testing

Find the words that land with buyers

Test taglines, value propositions and campaign lines with the people you need to persuade. See which ones are clear, believable and different, and hear why in people's own words.

Talk to salesWatch a test run
3AI follow-up questions on each video or voice answer, at most
5audience groups compared side by side, at most
20question types, each with its own chart

Why test the line

What you mean is not always what people hear

A line that reads well in a meeting room can sound vague, hard to believe or just like a competitor. Message testing compares what you meant with what people actually take away, before the line goes on a website, a pack or a billboard. It shows where a message lands and where it falls short.

What can go on the bench

TaglinesValue propositionsCampaign linesWebsite headlinesLonger statements, as an imageEmail subject lines

Meant versus heard

Watch a line meet the people it must persuade

  1. 01The lines, written alike
  2. 02What did it tell you?
  3. 03Scores, next to what they heard
  4. 04Where a line is misread

Step 1 / 4

The lines, written alike

Similar in length and style, so the comparison is fair. We go through the wording with you first.

Study set-up

Message testing

  • StimulusFourteen lines about a smart thermostat
  • WhoUS homeowners
  • TaskSets of four: the most and the least convincing
  • ThenThe top three rated, and explained in people's words
  • Read byAge bands, to see where a line splits

Set up with our team, or yourself in CatalyseAI

Step 2 / 4

What did it tell you?

People pick the most and least convincing lines, then say in their own words what a line means to them.

Line 2 of 3

In your own words, what is this line telling you?

"A home that learns your routine"

That it watches when I come and go?

How believable is this line?

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5

Not at allCompletely

Next

"Lower bills without thinking: yes please."

"Learns my routine sounds a bit creepy."

"Set it once, save all year. Simple."

Voice answer

AI follow-up: How do you feel about that?

"A bit uneasy. I would want to know what it keeps."

  • Length
  • Words
  • Topic relevance 0.82

Transcribed, checked and searchable. Up to three follow-ups per answer.

Step 3 / 4

Scores, next to what they heard

Clarity, belief and appeal for each leading line, set next to what people say the line means.

Live tally, as answers arrive

  1. A"Lower bills without thinking about it"

    They heard"Saves money on its own."

    • Clarity82%
    • Believability61%
    • Appeal70%
  2. B"A home that learns your routine"

    You meant"It saves energy on its own."

    They heard"It watches when I come and go."

    • Clarity64%
    • Believability55%
    • Appeal66%
  3. C"Set it once. Save all year."

    They heard"Easy, and cheaper bills."

    • Clarity85%
    • Believability58%
    • Appeal62%

Top-two-box shares. "They heard" is the answer people gave most often, in their own words.

Step 4 / 4

Where a line is misread

Setting what people heard next to what you meant shows the line that needs a rewrite.

Verdict

"Lower bills" leads

"Learns your routine" is misread, so it gets a rewrite.

  • Lower bills without thinking about itLead
  • Set it once. Save all year.Back-up
  • A home that learns your routineRework

The scorecard

Every line scored, next to what people heard

Each row is one measure, named the way the report names it.

Base: US homeowners; the top three of fourteen lines, rated one at a time in a random order.

Top-two-box shares. Best-worst score: picked most convincing minus picked least, as a share of times shown.

Message testing · scorecardBrand A · message test report
Measure
ALower bills without thinking about it
BA home that learns your routine
CSet it once. Save all year.
ClarityIs the line easy to understand on first read?
82%B
64%
85%B
BelievabilityDo people believe it?
61%
55%
58%
RelevanceDoes it speak to something they care about?
77%BC
58%
69%B
AppealDo they like it?
70%
66%
62%
DifferenceDoes it sound different from what others in the category say?
44%
71%AC
49%
Brand fitDoes it sound like your brand?
66%
63%
70%
TakeawayWhat they think the line means, in their own words.
"Saves money on its own."
"It watches when I come and go."
"Easy, and cheaper bills."
Best-worst scoreFor long lists: how often a line is picked as most convincing compared with least convincing.
+38%BC
+21%
+27%

The lead line

Example: choosing lines for a smart thermostat

A home-energy company has 14 lines about its smart thermostat, from "Lower bills without thinking about it" to "A home that learns your routine". US homeowners see the lines in sets of four and pick the most and least convincing in each set. The top three then go to a short follow-up in which people rate each line and say what it means to them. The team sees which lines lead overall, which ones split by age, and which one people misread before choosing what goes on the website.

  • Lower bills without thinking about it

    Lead

    Clear, relevant and the most convincing in the best-worst sets.

    It goes on the website.

  • Set it once. Save all year.

    Back-up

    The clearest line, a little less believed.

    Keep it for ads.

  • A home that learns your routine

    Rework

    It sounds new, but over-55s read it as the thermostat watching them.

    Say what it learns, and why.

Ask the studyAnswers from your own data

Which line splits by age?

"A home that learns your routine" appeals to under-35s, but over-55s more often read it as the thermostat watching them. The gap is significant.

It checks its numbers against your data and draws the chart.

Step by step

How a message test runs

From a list of lines to the one that lands.

  1. 01

    List the messages

    Paste in your lines. They should be similar in length and style so the comparison is fair, and we go through the wording with you before fieldwork.

  2. 02

    A few lines: rate each one

    With a short list, each person rates the lines one at a time, in a random order, on clarity, believability, relevance and appeal.

  3. 03

    Many lines: pick best and worst

    With a long list, people see four lines at a time and pick the most and the least convincing. Every line ends up with a score and a rank, and you can see how far apart they are.

  4. 04

    Ask what it means

    People say in their own words what a line is telling them. Setting that next to what you meant shows where a line is misread.

  5. 05

    Read the results by audience

    Compare groups such as age bands or current and lapsed customers, then download the deck or the workbook.

  6. Then: decide

    "Learns your routine" is misread, so it gets a rewrite.

    Plan your study

What you get

What you get

  • Output 01

    Lines in order

    Scores for each line on clarity, believability, relevance and appeal, or a best-to-worst ranking for a long list.

  • Output 02

    What people heard

    Each line's meaning in people's own words, grouped into themes and shown next to the meaning you intended.

  • Output 03

    Word-level reactions

    Show a longer statement as an image and ask people to click the words they find most convincing, then the words they doubt. The heatmap shows which phrases carry the line.

  • Output 04

    Where audiences split

    Up to five groups compared, with markers where a gap is statistically significant, so you can see when one line will not suit everyone.

  • Output 05

    Ready-to-share files

    The report as a PowerPoint deck, plus an Excel workbook, CSV data and PNG charts for your own slides.

FAQ

Questions, answered

How many messages can I test at once?

A short list can be rated line by line. For a longer list, a best-worst design works better, because people only compare four lines at a time. We suggest the route that fits your list.

How is this different from claims testing?

Claims testing checks product statements, such as "no added sugar", for belief and appeal. Message testing covers wider lines: taglines, value propositions and campaign ideas.

Can we hear why a line does not work?

Yes. People explain in their own words, and you can add video or audio answers in which an AI interviewer asks up to three follow-up questions. For longer conversations, ask us about interviews with our researchers.

Should I test lines on their own or inside an ad?

Start with the lines on their own to pick the strongest few. Once a line sits inside a finished ad, the visuals and voice change how it lands, so test the ad as a whole with video, static or audio ad testing.

Related

Goes well with

SolutionClaims testingFind the claims people believe and care aboutSolutionNaming, logo & slogan testingChoose a name, logo or slogan with evidenceSolutionMaxDiffRank a long list by what matters mostOverviewAll solutionsEvery study we run, by the decision it supports.

Got a list of lines to choose from?

Send us the messages and tell us who needs to believe them. We will suggest the simplest test that gives you a clear answer.

Talk to salesAll solutions

Nearest office: Orlando, FL (corporate office)