Skip to content
Get in touch

Attribution: why ad reports don't match

| Article: 3 min read | Analytics

Attribution is the rule that "assigns" a purchase to advertising touchpoints: a click, an impression, an email. Different systems (Google Ads, GA4, Meta, your CRM) apply different rules, so they show a different number of sales for the same period. This isn't a bug. They are different ways of measuring.

In plain words

A shopper saw an ad on Instagram, clicked a Google ad a week later, and came straight to the site that evening to buy. Who gets the sale? Each platform answers in its own way, and each number in the reports can be "right" by its own rule. The sum of the platforms' reports can be larger than the real number of orders.

Why the numbers differ

  1. Different models. GA4 offers data-driven attribution, "paid and organic last click" and "Google paid channels last click". The "first click", "linear", "time decay" and "position-based" models have not been available in GA4 since November 2023 (Google Analytics Help). Google Ads has only "last click" and data-driven; the other models were removed and their conversions moved to data-driven automatically (Google Ads Help).
  2. Different windows. Meta has click and view windows. From January 12, 2026, the "7-day view" and "28-day view" windows were removed from the Ads Insights API; what remains is 1-day view, 1-day click, 7-day click and 28-day click (Meta for Developers). In data-driven mode, GA4 redistributes conversions up to 7 days after the purchase (same help page).
  3. Direct traffic. In GA4, no model gives credit to direct visits if there was at least one other channel before the purchase (Google Analytics Help).
  4. Each platform grades itself. A platform sees only its own touchpoints and counts the purchase by its own rule.
  5. Technical losses. Not every purchase makes it into the reports: cookie refusals, blockers, tag errors. Google reminds you that conversion tags must be on all the needed pages, and the Conversion Linker must be active if you use Tag Manager (Google Ads Help). See Consent Mode and Server-side tracking.

What to do about it

A hypothetical example (the numbers are made up). In a month the store had 1,000 orders in the CRM. Google Ads reports 500 purchases, Meta 420, GA4 850 from all channels. Google and Meta add up to 920, which is more than you sold from ads at all. That's normal: the same buyers are credited to two platforms.

Steps: 1. The source of truth is money and orders in the CRM or accounting. The platforms are a hint about direction. 2. Write down which models and windows you use in each account, and don't change them without a note. Changing the model in Google Ads affects only future conversions; the "current model" columns let you see how past data would have looked (Google Ads Help). 3. Add the question "How did you hear about us?" to the order form. It's a simple correction that catches what tags can't see. 4. For budget decisions, use a test as well as attribution: see incrementality.

How different sources see it

In its help, Google describes data-driven attribution as a model that compares the paths of shoppers who converted and those who didn't (Google Analytics Help). Haus, which sells lift measurement, shows with an example that a platform can claim buyers who would have purchased without the ad (Haus). Both positions are reasonable: attribution splits the credit among touchpoints, but it doesn't answer the question "what would have happened without the ad".

Common mistakes

  • Adding up the platforms' reports and comparing the total with revenue.
  • Comparing ROAS in Google and Meta as "apples to apples".
  • Changing the model in the middle of a campaign.
  • Forgetting about windows and judging a campaign after a day or two.

FAQ

Why do Google Ads and GA4 show a different number of purchases? Different models and rules (for example, GA4 can credit free channels), different windows and different counted events.

Which attribution model is best? There is no universal answer. Google advises trying a model that isn't tied to the last click first and assessing the effect on profitability before choosing.

Why does Meta show more conversions than GA4? Meta counts views (up to 1 day) and clicks, while GA4 gives no credit to direct visits. The result depends on the chosen window.

Sources

Want to check this on your store?

Audit

We review the customer journey, catalog, technical state and data. A prioritized plan of changes in three days.

About the service →

Share

Related

  • Session recordings and heatmaps

    | Article: 3 min read

    A session recording is a replay of how one visitor used a site: where they moved the cursor, where they clicked, how they scrolled. A heatmap is a summary of such actions by many visitors on one page: where they click and how far they scroll. They show what…

  • LTV (customer lifetime value)

    | Article: 5 min read

    LTV (lifetime value, also CLV or CLTV) is the amount a customer brings to the store over the whole time of the relationship. For management decisions it is better to calculate it not in revenue but in margin: how much money the store keeps from all of the…

  • Incrementality and media mix models

    | Article: 3 min read

    Incrementality is the share of sales that advertising delivered on top of what the store would have received without it. If a person would have bought from you anyway, the ad added nothing with them, even though the report records them as a conversion. A…

  • CAC (customer acquisition cost)

    | Article: 5 min read

    CAC (customer acquisition cost) is the amount a store spends to get one new customer: all marketing and sales costs for a period divided by the number of customers who bought for the first time in that period.