Find the mechanism, not the winning product
A winning product is an outcome. A mechanism is a testable claim about why it sold. Only one of the two is worth anything on the next launch.
Three platforms, three answers, one budget to defend. Your GA4 export already holds the answer, and the first finding is free and never leaves your browser. Get your first finding free
By Joris van Huët, Founder & CEOUpdated 3 min read
A product is a result. A mechanism is a claim about why the result happened, stated precisely enough to be wrong. The second one transfers to your next launch. The first one does not.
The difference, stated plainly
"This product sells" is a fact about one product in one season through one set of ads. It is not portable.
"Products that solve a visible problem in under a minute, shown in a fifteen second demo, convert well to cold traffic" is a mechanism. It is portable, and it is falsifiable, which is the more important property.
A mechanism you cannot be wrong about is not a mechanism. It is a slogan.
Why the distinction decides what you can learn
| Framing | What you can do with it | What it costs to be wrong |
|---|---|---|
| This product wins | Copy the product | A full launch, discovered late |
| This mechanism wins | Test the mechanism on a cheap candidate | One test window |
Product research done as list building produces a shortlist of things other people already sell. Mechanism research produces a hypothesis you can check against a product you can source quickly, which is a much smaller bet.
The trap inside product research tools
Trend data, marketplace unit counts, and ad libraries all show you things that are already working. That is their value and their limit: everything they show you survived. The ones that were tried and failed are not in the data.
That is selection bias, and it is not a reason to ignore the tools. It is a reason to treat their output as a source of hypotheses rather than a source of answers. The details are in why copying bestsellers is survivorship bias.
Turning a mechanism into a test
A mechanism becomes testable the moment you can name the one thing that would have to differ between two otherwise matched cases.
- Same product, two hero formats, one difference. Tests presentation.
- Two products sharing the mechanism, same format. Tests the mechanism.
- Same product, same format, two audiences. Tests the audience, not the mechanism.
Pick one. Running all three at once means the winner tells you nothing about why. That is the same discipline described in vary one element.
Where a causal read fits
Mechanism tests produce a difference between two groups, and a difference without an interval is a story. Causality Engine reads one file, the Attribution paths export from GA4, and shows what each channel caused next to what last-click gave it.
It is 99 euro once, refundable within 30 days, no questions asked.
Related answers
Frequently asked questions
What is the difference between a winning product and a winning mechanism?
A winning product is an observed outcome for one item in one season. A winning mechanism is a stated reason the outcome happened, precise enough to be tested and therefore precise enough to be wrong. Only the mechanism transfers to the next launch.Are product research tools useless then?
No. They are good hypothesis generators and poor answer sources, because everything they show you already survived. The products that were tried and failed never enter the dataset, so the pattern you read off it is shaped by survivorship.How do I test a mechanism rather than a product?
Hold everything fixed except the one property the mechanism names, across two products or two presentations. If three things differ between your test groups, a win tells you that something worked, which you already knew.
Go deeper: Causal attribution, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Have an audience of ecommerce operators? Earn 20% recurring commission by recommending the €99 read.
Keep reading
Terms in this article
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- CausalityCausality is the relationship where one event directly causes another, essential for identifying specific actions that drive desired outcomes in marketing.
- ClickClick is the action a user takes to interact with a digital advertisement, redirecting them to a website or landing page. Clicks are a fundamental metric for measuring ad engagement and a primary input for click-based attribution models.
- IncrementalityIncrementality is the part of an outcome, such as orders or revenue, that a marketing activity caused: the extra results compared with what would have happened without it. It is a causal effect, not the number of conversions credited to a channel.
- Selection BiasSelection Bias occurs when data points selected for analysis do not represent the target population. This leads to distorted findings about marketing campaign impact.
- Survivorship BiasSurvivorship bias is the logical error of focusing on successful outcomes while ignoring failures. This leads to false conclusions by overlooking unseen data.