> For the complete documentation index, see [llms.txt](https://docs.zigpoll.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zigpoll.com/guides/how-to-test-your-messaging-and-positioning.md). # How to Test Your Messaging and Positioning Your homepage headline was probably chosen by consensus. Someone wrote three options, the team voted, and the winner has been sitting there converting at whatever rate it converts at ever since — with no way to know whether option two would have done better. Messaging is testable. And unlike an A/B test on live traffic, a messaging survey tells you not just *which* line wins but *why*, which is what lets you write the next one. *** ## What to test There are three distinct questions here, and they need different surveys. Running them in the wrong order wastes the effort. 1. **Which benefit do customers care about most?** (positioning) 2. **Which way of saying it lands hardest?** (messaging) 3. **Do they even understand what we do?** (clarity) Start with clarity. A beautifully worded headline for a product nobody understands is a wasted headline. *** ## Test 1: Clarity — do they get it? Add a **Short Answer** question, shown to first-time visitors: > In your own words, what do you think we sell? Brutal, and worth it. If a meaningful share of visitors can't answer, no amount of clever copy will help — the problem is comprehension, not persuasion. Target it to new visitors on the homepage with [Targeting Settings](/polls/targeting-settings.md). Zigpoll's Z-GPT Insights will cluster the answers into themes automatically, so you can see at a glance how many people got it right. {% hint style="info" %} Run this before a redesign and again after. It's the clearest before/after measurement of whether a new homepage actually communicates better, and it's immune to the traffic-mix noise that muddies conversion-rate comparisons. {% endhint %} *** ## Test 2: Positioning — which benefit matters? Add a **Ranked Choice** question: > Rank these in order of what matters most to you. List your candidate value propositions — price, quality, speed, sustainability, convenience, service. Force a ranking rather than a set of ratings; if people can rate everything as important, they will. The result tells you what to **lead with**. Most brands lead with the benefit they're proudest of rather than the one customers rank first, and those are often different things. A brand that leads with sustainability while customers rank fit and price one and two is leaving money on the table — not because sustainability doesn't matter, but because it isn't the *opening argument*. Follow with a **Long Answer** question: **"What made you choose us over the alternatives?"** The repeated phrases here are your headline, pre-written by the people you're trying to persuade. *** ## Test 3: Messaging — which line lands? Now that you know the benefit, test the wording. 1. Create two or three variants of the same survey, each showing a different headline. 2. On each, add a **Range** question (1–10): **"How likely would this make you want to learn more?"** 3. Run them simultaneously as an [A/B test](/tutorials/a-b-testing-surveys.md) so each visitor sees exactly one headline. 4. Add a **Long Answer** follow-up: **"What does that line tell you about the product?"** Comparing top-box scores (9s and 10s) across variants tells you which headline wins. The open-ended follow-up tells you what each line actually communicated — which is how you find out that your favorite headline is testing badly because half of respondents read it as a discount offer. {% hint style="warning" %} Showing a respondent all the headlines side by side tests which one wins a beauty contest. Showing each person one headline tests how it performs in the wild, which is the only condition that matters. Use variants, not a multiple-choice question. {% endhint %} *** ## Test 4: The objection Whatever your messaging says, something in it is failing to answer a doubt. Add a **Long Answer** question to a post-purchase survey: > What almost stopped you from buying? Buyers navigated every objection and converted anyway, so they remember exactly what gave them pause. Each answer is a doubt your messaging didn't address — and for every buyer who overcame it, others didn't. See [How to Find Out Why Visitors Aren't Buying](/guides/how-to-find-out-why-visitors-arent-buying.md). *** ## Reading the results * **Segment before deciding.** Different customer types rank benefits differently. Run the positioning test against your [ICP segment](/guides/how-to-build-your-ideal-customer-profile.md) specifically — the blended average is a compromise that speaks to nobody. * **Prefer the customer's words to your own.** Copy lifted verbatim from open-ended responses consistently outperforms copy written in a meeting, because it's already been proven to make sense to a customer. * **Confirm on live traffic.** A messaging survey narrows five candidates to one cheaply. A live A/B test on the winner confirms it converts. Do both, in that order — the survey is fast and cheap, the live test is slow and definitive. [Synthetic Research](/synthetic-research.md) is useful upstream of all of this: generate responses from your target audience profile to sanity-check wording before you spend real traffic on a headline that turns out to be ambiguous. *** ## Related {% content-ref url="/pages/7ah7NOivpQ5byvqk41c2" %} [How to Build Your Ideal Customer Profile](/guides/how-to-build-your-ideal-customer-profile.md) {% endcontent-ref %} {% content-ref url="/pages/pXWRToES22ZJ7KatoKHF" %} [How to Test Product Names, Packaging, and Ad Creative](/guides/how-to-test-product-names-and-creative.md) {% endcontent-ref %} {% content-ref url="/pages/70rirBfJvG59hsGptkAE" %} [How to Find Out Why Visitors Aren't Buying](/guides/how-to-find-out-why-visitors-arent-buying.md) {% endcontent-ref %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.zigpoll.com/guides/how-to-test-your-messaging-and-positioning.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.