Skip to content

For GA4 usersFrustrated with GA4 attribution? Upload your GA4 export, see causal insights in 5–10 minutes for €99 pay-per-use.

Causal Inference

3 min read

A bestseller is not evidence of why

Sales prove a product sold. They do not isolate which of its properties caused it, and copying the wrong property is the most common mistake in the category.

Share
Quick Answer·3 min read

A bestseller is not evidence of why: Sales prove a product sold. They do not isolate which of its properties caused it, and copying the wrong property is the most common mistake in the category.

Read the full article below for detailed insights and actionable strategies.

The attribution problem

One sale. Four channels. 400% credit claimed.

100
1 sale
Meta
100%
claimed
Google
100%
claimed
TikTok
100%
claimed
Klaviyo
100%
claimed

Reported revenue: 400 · Actual revenue: 100 · Gap: €300

A bestseller tells you that something worked. It does not tell you which part. The distinction decides whether your next product repeats the win or the coincidence.

What a sales figure contains

Every unit sold is the joint result of the concept, the execution, the listing copy, the price, the images, the traffic that reached it, the season it launched into, and the position it happened to occupy in a feed.

The sales number is the sum. Nothing in it is labelled.

Why this matters more than it sounds

The obvious move after a winner is to make more like it. "Like it" requires a theory of which property mattered, and that theory is usually formed by looking at the product and deciding which feature is most visible.

Most visible is not most causal. A design might sell because of its subject rather than its style, or because it launched the week a related topic was everywhere, or because it was the only item in its category with a clear photograph.

The decomposition problem

Candidate causeWhy it is hard to isolate
ConceptChanged together with execution, almost always
StyleRarely varied while holding concept fixed
PriceCorrelated with category and with margin decisions
Launch timingConfounded with everything seasonal
Traffic sent to itYou chose where to send traffic, based on early signs

The last row is the quietest trap. Products that look promising early get more traffic, and then sell more, which confirms they were promising. That loop can manufacture a bestseller out of an ordinary product.

What would actually settle it

Varying one property while holding the others fixed, across enough items that noise averages out. That is a designed test rather than a look at history, and it is the only thing that converts a hunch into a repeatable rule. The general form is causal inference.

The read

A causal read on a Google Analytics export returns per channel an estimate, a confidence interval, a coverage share, and a label for what could not be resolved. It is 99 euro once, refunded if it does not move a budget decision.

It answers the channel question rather than the product question, and it is the same discipline: an estimate with an interval, or an honest blank. The interactive demo shows it with no signup.

Key Terms in This Article

Related Articles

Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.

Ready to see your real numbers?

Own the budget? Upload your GA4 export and see which channels drive incremental sales, with confidence intervals, in minutes. Have to defend it? Start with the live demo and take the read to your CFO.

Full refund if you don't see value.

Stay ahead of the attribution curve

Weekly insights on marketing attribution, incrementality testing, and data-driven growth. Written for the person who owns the budget and the person who has to defend it.

Which one are you? Optional.

No spam. Unsubscribe anytime. We respect your data.

Frequently Asked Questions

Why is a bestseller not proof of what works?

Its sales are the joint result of concept, execution, copy, price, images, traffic, timing and placement. The number is the sum and none of it is labelled, so it cannot tell you which property to repeat.

What is the traffic loop problem?

Products that look promising early get more traffic and therefore sell more, which appears to confirm they were promising. That feedback loop can manufacture a bestseller out of an ordinary product.

What would actually identify the cause?

Varying one property while holding the others fixed across enough items for noise to average out. That is a designed test, not a review of history.

Related reports

Real reports on this topic.

Anonymised reports from the Attribution Report Library tagged with causal inference.

Browse all related reports

Find your wasted ad spend in 5–10 minutes.

Watch the model work on a sample store first, no signup. Then upload your last 40–90 days of GA4 sessions and get incremental ROAS with confidence intervals. No pixel, no SDK. €99 per read.

Prefer to talk it through? Book a 20-min call, or read how it works.

Last-click guesses.We run the math.

Causal attribution for ecommerce brands. Watch the model work on a sample store first, then upload your GA4 export and see which channels really drove revenue in 5–10 minutes. €99, pay-per-use. Pro at €299/mo when you want it continuous.

No signup for the demo. Book a 20-min call or compare plans.