A Runbook for Testing Which Value Proposition Actually Resonates
Most so-called message testing programs test three phrasings of the same idea rather than three genuinely different value propositions, which explains why the results rarely change anything about how the company actually talks about itself. A real resonance test needs a wider spread of ideas and a cleaner way to isolate what actually moved the number.
Here is a runbook for running that test properly, from picking the candidates to declaring a winner you can actually trust and carry forward into ads, sales conversations, and everywhere else the message eventually needs to travel.
Vendors Covered in this Article
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
How do you write three different value propositions, not three headlines?
A test comparing 'Save time on X' against 'Save hours on X' is a wording test, not a message test, and it will not tell you anything about which underlying value your market actually responds to. Write candidates that lead with genuinely different core claims, speed, cost reduction, risk reduction, ease of use, so the test can actually distinguish which value matters most to your audience, not just which synonym reads slightly better. If two of your three candidates could both be true of the same sentence with a word swapped, they are not different enough yet to run as a real test.
Why test against a real decision, not just a click?
A headline that gets a high click-through rate but does not lead to an actual conversion is measuring curiosity, not resonance. Build each message variant into a full landing page with a real conversion action, a form, a demo request, a signup, so the test measures whether the message actually moves someone toward a decision rather than just earning an idle click from someone who was never going to convert either way, no matter how the headline was worded.
Isolate the message from the offer
Changing the headline and the discount at the same time in the same test makes it impossible to know which one actually moved the result, and that confound is one of the most common ways a message test produces a confident-sounding but meaningless answer. Hold the offer, price, and everything else on the page constant across variants, and change only the core value proposition being tested, so any difference in conversion can be attributed to the message itself and not to something else that quietly changed alongside it.
For example, a team wants to test a cost-saving message against a speed message. If the cost-saving variant also carries a discount banner and the speed variant does not, a win cannot be credited to the message. Keep the offer, price, form and page layout identical, and change only the core claim in the headline and the supporting copy that develops it. Write down the single variable before launch, and check every variant against that note before it goes live. The mistake to avoid is fixing an underperforming variant midway, since any edit turns a clean comparison into a new test that has to start from the beginning.
Pair the quantitative result with a qualitative signal
A conversion rate tells you which message won, but not why, and that gap matters when you go to apply the winning message somewhere else, an ad, a sales deck, an email subject line. Pull a handful of sales call notes or open-text survey responses from the same period and check whether they echo the winning message's actual language; that qualitative echo is what tells you the win reflects real resonance rather than a small-sample statistical fluke that happened to land in your favor this particular test.
Compare the winning variant against a real conversion benchmark
Before declaring the winning message a success in absolute terms, not just relative to the other candidates, check its landing page conversion rate against a current external benchmark1. A message that beat two weaker candidates but still converts well below a reasonable external reference has won the test without actually solving the underlying resonance problem, and is worth another round of candidates rather than an early victory lap.
Build variant pages fast, but keep the variable count under control
Once you know which value propositions you are testing, a tool like Gamma can help spin up the variant landing pages quickly, which matters when speed of test execution is the bottleneck. Resist the temptation to let a fast build tool tempt you into testing five variants at once on limited traffic; a controlled two or three-way test that finishes with real statistical confidence beats a five-way test that never quite gets there and leaves every candidate looking equally plausible.
Run the test in this order:
- Write three value propositions that lead with genuinely different core claims, not three phrasings of one idea.
- Build each into a full landing page with a real conversion action, such as a form, demo request or signup.
- Hold the offer constant so only the message changes between variants.
- Pair the conversion result with sales call notes or open-text survey responses to learn why the winner won.
- Compare the winner with an external benchmark, and keep the number of variables under control as you scale up variants.
What Good Looks Like
Good resonance testing means the winning message was tested against genuinely different alternatives, measured against a real conversion action, and checked against an external benchmark before being adopted more broadly.
Building The Capability (5-Stage Skill Ladder)
How to Get Started
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
Frequently Asked Questions
How long should a message resonance test run?
Until it reaches statistical confidence for your traffic volume, which varies a lot by site. Resist calling a winner early based on a promising few days; a message test decided too early is one of the more common ways teams end up standardizing on the wrong message.
Should different segments get different winning messages?
Often yes, especially if your segments have meaningfully different priorities, one buyer cares most about cost, another about risk. Test message resonance separately by segment when you have enough traffic to do so, rather than assuming one universal winner fits everyone.
What is the most common mistake in this kind of testing?
Testing wording variations of the same underlying idea and calling it a message test. A real test needs candidates that lead with genuinely different core value claims, not just different phrasing of the same one.
Sources
Where we quote a benchmark, we show its source. Other figures in this guide are estimates or general guidance, so check them against your own numbers.
- Landing Page Conversion Rate (Median, All Industries). Unbounce Conversion Benchmark Report.
Related Guides
Why Creative Testing Slows Down Exactly When You Need It Most
The bottlenecks that make ad creative testing slow down as a program grows, and how to keep shipping variants without cutting corners on what each test teaches.
Where Conversion Rate Benchmarking Actually Leaks Marketing Budget
The specific comparisons that reveal wasted marketing spend, and why comparing your conversion rate to the wrong benchmark hides the leak instead of finding it.
How to Run a Marketing Performance Audit Before a Board Meeting
A five-step way to check marketing spend, conversion rate and creative performance before a board meeting, so you present numbers you can defend.
Four Video Ad Hook Patterns Worth Testing This Quarter
A breakdown of what makes a short-form ad hook work, with worked examples of weak openings rewritten into stronger ones.
Testing Whether Your Value Proposition Actually Resonates
A question and answer walkthrough of how to actually test whether your value proposition resonates, instead of guessing from internal opinions alone.
An SEO Audit for Terms Your Competitors Already Rank For
A practical audit for winning search positions your competitors currently hold, covering content gaps, technical blockers and link signals.