Before
TestedReebokProduct page
'Best For' section removed
What changed Removes the 'Best For' section from the product details list.
Seen 7 Aug to 4 Oct 2026
Shipped
ABSee / Experimentation intelligence
GuideExperimentation intelligence. Evidence before you test.
Experimentation intelligence is evidence about what the market has tested and kept, used before you run your own experiment, so the next test slot goes to a better-backed idea.
Updated October 2026 · Source: ABSee record · Methodology →
Definition
Fill the “because” clause. With the market, not only your analytics.
Every experiment starts with a hypothesis, and a good hypothesis starts with “because we saw…”. Most teams can only fill that clause with their own analytics: where shoppers drop, what they click, what they ignore. That tells you where the problem is. It says little about which fix is worth a test.
Experimentation intelligenceA record of real experiments other brands ran on their live sites: the page before, the version tested, when it ran, and what the brand did next.
Read before you test, it tells you whether an idea has been tried, how others built it, and whether they kept it. ABSee is experimentation intelligence for ecommerce: a record of real experiments run by 1,150+ brands, with the page before, the version tested, and whether it was kept.
How it differs
Four tools, four questions. None of them replaces the others.
Experimentation intelligence sits beside the tools you already use. Each answers a different question.
- Runs your tests
Experimentation platform
Splits your traffic, serves the versions and measures the result.
- Your data
Analytics
Analytics and session replay show where your own shoppers struggle.
- Expert review
UX benchmarks
Expert review and usability studies describe what tends to work.
- What the market tested and kept
Experimentation intelligence
What brands actually put in front of visitors as a test, and what they kept. It comes before the platform, when you are choosing what to put in it.
Competitive intelligence usually tracks prices, promotions and ads. Experimentation intelligence tracks experiments: two versions of the same page, served at the same time.
What it is used for
What teams use it for. Before the test slot is spent.
Stronger hypotheses
A hypothesis that cites real experiments on similar pages is easier to defend and easier to brief.
Prioritisation
Test slots and engineering time are scarce. Evidence that an idea has been tried, and kept, by brands like yours helps decide which idea goes first.
Avoiding repeats
If several brands tried an idea and rolled it back, that is worth knowing before you build it. It may still be right for you, but you go in with your eyes open.
Better variants
Seeing how others built the same idea, where they placed it and on which device, gives a design team more than one starting point.
What the evidence looks likeBeforeTested 
Before
Tested
Before
Tested
Three real experiments on product pages. Who kept the change, and who rolled it back.
Each record holds the page before, the version tested, the dates it was seen, and what the brand did next. The chip is that decision: Shipped means the tested version became the page; Rolled back means the original stayed.
BeforeTestedBeforeTestedAllSaintsProduct page
Wishlist button moved to the product name
What changed Moves the wishlist button from beside Add to bag to under the product name, as a bookmark.
Seen 19 Jun to 17 Jul 2026
Shipped
BeforeTestedKaren MillenProduct page
Green Add to Bag button
What changed Turns the Add to Bag button from black to green.
Seen 27 Aug to 27 Aug 2026
Rolled back
What it cannot do
What it cannot do. Your own test still decides.
It brings the evidence; your team owns the diagnosis and the decision. The methodology sets out how outcomes are read and the known limits.
It does not predict your result
Another brand's experiment was decided by its own visitors, prices and brand, and “kept” is that brand's decision, not a measured lift. ABSee never sees a brand's own numbers.
It does not replace your own test
And it does not tell you what to test. Treat each experiment as evidence for or against a hypothesis, not as a result to copy.
Where it sits in CRO
Where it sits in a CRO workflow. Step two, before you build.
With ABSee, step two happens inside Claude, ChatGPT or Gemini: connect it once and ask in your own words. See how to connect, or what is in the database.
Hypothesis
Find the problem in your own analytics, research and replays, and write the hypothesis.
Check the market's evidence
Ask what the market has tested against that problem, and what was kept or rolled back.
Prioritise
Put that evidence in the “because” clause, and rank the idea against your others.
Run your own test
In your testing platform. Read your result, and keep the evidence next to it for the next decision.
Questions
Questions
What is experimentation intelligence?
Evidence about what other brands have tested and what they kept, used before you run your own experiment. It sharpens the hypothesis; your own test still decides.
Is it the same as an experimentation platform?
No. An experimentation platform splits your traffic and measures your result. Experimentation intelligence comes before that step: it shows what the market already tried on a similar page.
Can it tell me which version will win?
No. Another brand's result was decided by its own visitors. Experimentation intelligence tells you whether an idea has been tried, how others built it, and what they did next. Only your own experiment tells you what works for your audience.
Who uses it?
Product, growth and CRO leads at ecommerce brands, and agency strategists building test roadmaps for clients. Anyone who has to choose which idea gets the next test slot.
Ask your assistant
What is the evidence for and against urgency messaging on product pages?
Works in Claude, ChatGPT or Gemini once ABSee is connected.
Connect your AI →