Value Proposition A/B Tests and Examples

The value proposition examples that won A/B tests in our database all do one thing: they say plainly what the visitor gets, in the visitor's words, for the visitor's specific job.

CityCliq lifted clicks to pricing by 90% by swapping "Businesses grow faster online!" for "Create a webpage for your business", while vaguer, more "inspiring" lines lost.

Adding more value copy is not a safe bet, though: several tests below show it hurting when visitors were already convinced.

This hub covers SaaS and lead-gen tests tagged value proposition, headline or message match.

The full grid of results sits below the article.

This page is the analysis: what won, what lost, and how to test your own.

Value proposition examples from real A/B tests

The cleanest evidence comes from tests that changed only the value proposition line.

These are the strongest single-variable examples we have.

CityCliq's test is the most useful because it tested four lines at once.

"Get found faster!" reached 31.8% and "Online advertising that works!" fell to 20.2%, below the 25.3% control.

The winning line was the only one that named the thing you actually get.

Imsider.ru, a Russian service that helps people launch online stores, ran its test at 99.99% significance.

The winner kept the same offer but turned a process ("a step-by-step plan") into an outcome with a time frame ("in one evening").

The headlines are translations, so treat the wording as a pattern, not a template.

Shockbyte, a game server host, shows the focus pattern.

Most of its visitors came for Minecraft, so a hero built only around Minecraft beat a hero that listed every game, with a 95.67% probability to beat baseline.

The winning version paired the Minecraft-only copy with a Minecraft-themed image; a copy-only version also ran but was not the winner.

How to write a value proposition

Pulling the winners and losers together gives a short, testable method.

None of these steps is a guaranteed lift; each is a hypothesis the tests support.

1. Name the concrete outcome. "Create a webpage for your business" beat three abstract claims. If a visitor can't tell what they will have after signing up, the line is too vague.
2. Make the cost of effort small and specific. "In one evening" turned a plan into a promise. KlientBoost reports a similar idea for an insurance client, where "Get coverage within 24 hours" improved a page already converting 28.88% on step 1, though the variant number was only shown in an image, so we can't quote it.
3. Borrow the customer's words. Laura Roeder's winning headline was a direct customer quote. Review mining, surveys and sales calls are where these lines come from.
4. Lead with the segment that matters most. Shockbyte's Minecraft focus worked because that was the main audience. A generic line for everyone loses to a specific line for most people.
5. Give a reason to act, not just a label. Dutchento, a Dutch Magento community site, added a title and a bulleted list of reasons to subscribe to a box that had none. Newsletter link clicks rose 190.31%, though RSS clicks fell 44.46% and the metric was clicks, not completed sign-ups.
6. Put it where the decision happens. Tinkoff Bank added a "More details" link under its credit card application button across 60,000+ visitors. The version that expanded an inline box of card benefits lifted applications 15.5%; the version that sent people to a separate info page lost to the control.

For the line-level craft of headlines themselves, see headline and copy tests.

For value props shown as a strip of short benefits, see USP bars.

When value proposition copy lost

The losses matter as much as the wins, because they show where "add more value copy" breaks down.

• Visitors already convinced. On a seminary's donation page, NextAfter tested putting the value proposition and quotes from well-known figures before the ask. Donations fell 28%. People reaching that page were already highly motivated, and the extra copy got in the way.
• One value point instead of several. MECLABS ran four crowdsourced Consumer Reports donation emails, each focused on a single value point. All four lost to the control, which mixed several reasons. Narrowing the message is not always clarifying it.
• Too little traffic to judge. MarketingExperiments reworked a tech supplier's product page and saw Treatment C at +445.6%. A rerun with a cloned control then showed a 30% gap between identical pages, so the team did not trust the lift. The page only had 2 to 10 conversions a day.

The pattern: value proposition copy earns its space when visitors are still deciding and don't yet know what they get. When they already know, it is friction.

Bundled redesigns: big numbers, weak attribution

Many of the biggest numbers in this hub come from redesigns that changed the value proposition together with layout, proof and forms.

A MECLABS test for Aetna's HealthSpire found a longer Medicare page with value copy drove 638% more leads than a short page, but several copy and layout changes ran at once.

Agency-reported paid-media case studies, such as KlientBoost's LemonStand work (+47% conversion rate), changed ads, keywords and pages together.

These results suggest a clear value proposition is part of what a winning page looks like.

They can't tell you how much the value proposition alone was worth.

If you want that answer, test the line by itself first.

What is a value proposition?

A value proposition is the promise of what a customer gets from you, and at what cost.

The phrase is credited to McKinsey consultants Michael Lanning and Edward Michaels, whose 1988 staff paper "A business is a value delivery system" said a business "promises customers some value, a combination of benefit and price" (summary and citation).

On a website, the value proposition usually shows up as the headline, the subhead under it, and the first few benefit points in the hero.

It is not the same as a slogan.

"Businesses grow faster online!" is a slogan; "Create a webpage for your business" is a value proposition, and CityCliq's test shows the difference was worth 90% more clicks to pricing.

Three related ideas often get mixed up:

A good example of the overlap: Worship Digital added a bar of selling points like "no hidden fees" under the header of the home cover insurer Corgi HomePlan and reported +30.9% conversions.

That was a value proposition delivered as a USP bar, with no sample size or significance given.

Value proposition canvas: where it helps and where tests take over

The Value Proposition Canvas is a planning tool from Strategyzer, set out in the 2014 book Value Proposition Design by Alexander Osterwalder, Yves Pigneur, Greg Bernarda and Alan Smith (book details). It has two halves:

• Customer profile: the jobs customers are trying to get done, the pains they hit along the way, and the gains they want.
• Value map: what you offer, the pains it relieves, and the gains it creates.

"Fit" is when the right side answers the left.

Strategyzer's own page says fit "is a claim until customers confirm it", which is where A/B testing comes in.

Used for CRO, the canvas is a good source of variant ideas.

Each top-ranked pain or gain becomes a candidate headline.

Shockbyte's test is a canvas exercise in practice: the main customer job was "run a Minecraft server", so the winning hero spoke only to that job.

The canvas can't tell you which claim will win on the page.

The Consumer Reports test is a warning here: picking a single "most important" value point and leading with it lost four times out of four.

Fill in the canvas, write three or four lines from it, and let a test decide.

How to test a value proposition

• Test the line alone first. Change the headline and subhead only, as CityCliq and Imsider did. You'll know what caused the result.
• Run several candidates at once if you have the traffic. CityCliq's four-way test showed both the winner and which framings hurt.
• Measure past the click. Dutchento measured link clicks and Omnisend measured visits to a registration page, not sign-ups. Track the conversion you care about.
• Watch your conversion counts. VenueSphere reported +69% form submissions from a subheadline change on close to 1,200 visitors, and the tech supplier test above shows how small samples mislead.
• Check where visitors come from. Paid visitors expect the ad's promise; see message match tests before rewriting a paid landing page.

Browse every result for this element in the grid below, or search the full test database and the other page element hubs.