A Shopify measurement stack: what each tool answers
Shopify answers what sold, GA4 which visits led to key events, ad platforms what they claim, a holdout whether it would have happened anyway, and a causal read what each channel caused. Drop any tool with no question.
€99 once, excl. VAT. Full refund within 30 days, no questions asked. You keep the read.
By Joris van Huët, Founder & CEOPublished 4 min read
Buy tools by the question they answer: Shopify answers what sold and what it was worth, GA4 answers which visits led to key events and looks back 90 days by default, ad platforms answer what they say they drove inside their own windows, a holdout answers whether the sale would have happened anyway, and a causal read estimates what each channel caused from history. A tool with no question attached is a cost, not a measurement.
Which question does each tool answer?
| Question | Tool | What its documentation says it measures | Blind to |
|---|---|---|---|
| What sold, and what was it worth? | Shopify reports | The value of goods in each sale, not the money received | Whether marketing caused the sale |
| Which visits led to a purchase? | GA4 attribution reports | Credit assigned to touchpoints along a user's path, within a lookback window | Touches outside that window |
| What does each platform claim? | Ads Manager, Google Ads | Conversions recorded after an ad interaction, inside that platform's window | Other platforms' claims, and the overlap between them |
| Would it have happened anyway? | Holdout or lift test | The difference in conversions between people who saw ads and people who didn't | Channels and periods you didn't test |
| What did each channel cause in your history? | A causal read | Each channel's share of the credit, estimated from GA4 history, next to last-click | Campaigns, and anything not in the export |
Which tools can you stop paying for?
Write the question each tool answers, then the last decision its number changed. A tool with no question, or no decision in the last quarter, is a candidate to drop. Where two tools answer the same question with different numbers, reconcile both to Shopify orders for one week before you trust either; the steps are in the cross-channel reconciliation.
Pass: every tool maps to one row above and to one decision you made. Fail: a tool answers a question you never ask, or two tools disagree and you can't say which one reconciles to your orders.
Where does a holdout fit, and what does it cost in data?
It is the row that tests "would it have happened anyway" directly, by withholding the ads, and Google says its Conversion Lift isn't available for all Google Ads accounts. It also demands volume. Lewis and Rao's 25 field experiments with major US retailers and brokerages put the median confidence interval on return on investment at over 100 percentage points wide, and informative experiments can easily require more than 10 million person-weeks (Quarterly Journal of Economics, 2015). Expect wide ranges from a short test, and read a regional holdout as a range, not a point.
Where does a causal read fit?
A causal read estimates what each channel caused from the history you already have, so there is no test to design. Later, once the stack above reconciles to your orders, a causal attribution read like Causality Engine's can give a second opinion on the channel numbers from your GA4 export.
Sources, 30 September 2026: Marketing reports (Shopify Help Center); Select attribution settings (Google, Analytics Help); About conversion windows and About Conversion Lift (Google Ads Help); The unfavorable economics of measuring the returns to advertising (Lewis and Rao, Quarterly Journal of Economics, 2015, peer-reviewed).
Related answers
Frequently asked questions
What do I need to measure marketing on Shopify?
Start with Shopify's reports for orders, GA4 for visits and key events, and each ad platform's manager for what it claims. Add a holdout or lift test when you need to know whether sales would have happened anyway. Reconcile each tool to your orders before trusting it.Do I need a paid attribution tool for Shopify?
Not to start. GA4 is free and Shopify's marketing reports are included with Shopify. A paid tool earns its place when it answers a question those can't, and you can name the decision its number would change.How do I know which measurement tools I can drop?
Write the question each tool answers and the last decision its number changed. A tool with no question, or no decision in the last quarter, is a candidate to drop. When two tools answer the same question differently, reconcile both to Shopify orders first.
Go deeper: Incrementality testing, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
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.
- Attribution ReportAttribution Report shows which touchpoints or channels receive credit for a conversion. It identifies which campaigns drive desired actions.
- Causal AttributionCausal Attribution uses causal inference to determine which marketing touchpoints genuinely cause conversions, not just correlate with them.
- Confidence IntervalConfidence 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.
- ConversionConversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
- ExperimentsExperiments are scientific procedures that test hypotheses or demonstrate facts. In marketing, experiments like A/B tests determine the causal effect of campaign changes, enabling data-driven decisions.
- Google AdsGoogle Ads is an online advertising platform where advertisers bid to display ads, service offerings, and product listings.
- TouchpointTouchpoint is any interaction a customer has with a brand throughout their journey. In marketing attribution, each touchpoint is a data signal to understand marketing impact.