The lift surprises hiding in your own Shopify data: Four things that keep surprising brands the first time they read their channels causally, and the unglamorous explanation behind each one.
Read the full article below for detailed insights and actionable strategies.
Customer journey
The customer journey last-click attribution misses
One conversion. Five touchpoints. Last-click credits the final touch with 100%.
Last-click attribution
Every other channel gets zero credit, even though they created the demand.
Causal inference
Four patterns show up again and again the first time a brand reads its channels causally, and none of them are exotic. They surprise people because platform reporting is structured to hide exactly these four.
The four
| Pattern | What it looks like | The ordinary explanation |
|---|---|---|
| Branded search overstated | Huge reported ROAS, thin causal estimate | It captures demand other channels created |
| Retargeting overstated | Excellent reported ROAS, wide causal interval | It reaches people already intending to buy |
| Email under-credited in platforms, real in the read | Small platform number, solid causal estimate | Platforms do not see it, so they cannot claim it |
| A channel that cannot be measured at all | No estimate returned | Spend below the level any method can resolve |
Branded search
Branded search converts extremely well because the person already typed your name. Something made them type it, and that something is usually a channel further up. Reported ROAS credits the last click; a causal read asks what would have happened without the branded campaign, and the answer is often "most of those people would have arrived anyway".
That does not mean cut it to zero. Defensive reasons to bid on your own brand exist. It means the reported figure is not the argument for the spend level.
Retargeting
The same structure with a different mechanism. Retargeting reaches people who already visited, so it is selecting on intent rather than creating it. High reported ROAS is a description of the audience, not of the channel's effect. The holdout on retargeting is the test that settles it for a specific brand.
Email doing better than the dashboards say
The mirror-image surprise. Email rarely appears in ad platform reporting at all, so it looks small in any view assembled from platform claims. In a read anchored on your own orders it frequently holds up, particularly for repeat purchases.
The channel with no estimate
A well-built method declines to score a channel whose spend is too small for its effect to be separated from noise. The first reaction is usually that the tool is broken. It is the opposite: a tool that returns a confident number for a channel spending very little is filling in rather than measuring. The reasoning is in the measurability floor.
What to do with a surprise
Not act on it immediately. A single observational read is one piece of evidence, and the correct response to a surprising result is to check whether the window contained something unusual, then design a proper test on the channel with the most at stake. That sequence is in cut the channel you would holdout first.
Causality Engine produces the read from a Google Analytics export at €99 once, with the confidence interval, coverage and design label on every channel, refundable if it does not move a decision. The interactive demo shows the same output on a sample store first.
Why these four recur
Because platform reporting credits proximity to the conversion, and proximity is not causation. Any method that asks the counterfactual question will disagree with proximity-based credit in the same predictable places.
Related answers
Get attribution insights in your inbox
One email per week. No spam. Unsubscribe anytime.
Key Terms in This Article
Attribution
Attribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
Confidence Interval
Confidence Interval is a statistical range of values that likely contains the true value of a metric. In marketing analytics, it quantifies uncertainty around estimates, indicating the precision of an outcome or causal effect.
Conversion
Conversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
Counterfactual
Counterfactual is a hypothetical outcome that would have occurred if a subject had received a different treatment.
Google Analytics
Google Analytics is a web analytics service that tracks and reports website traffic.
Incrementality
Incrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
Multi-Touch Attribution
Multi-Touch Attribution assigns credit to multiple marketing touchpoints across the customer journey. It provides a comprehensive view of channel impact on conversions.
Retargeting
Retargeting is online advertising that targets users who have previously interacted with your website or content. Attribution analysis shows the causal role of retargeting in driving conversions and improving ad spend.
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.
No spam. Unsubscribe anytime. We respect your data.