How does Google Ads attribution work?

Google Ads usually credits a sale to its own ads when the buyer clicked one in the 30 days before buying, the default window. Data-driven attribution, the default for most conversion actions, then shares that credit among the Google ads on the path. Your other channels never enter the sum.

How Google Ads hands out the credit

Four things do the work: the tag on your site, then the window, the model and the count set on each conversion action.

The tag records the sale. A click on your ad leaves a temporary cookie on the shopper's device. When they buy, the Google tag on your site recognises it and Google records a conversion.

The window sets how far back a click can reach. Google's default click-through window is 30 days, and you can set anything from 1 to 90 days, depending on the source. Engaged video views get 3 days by default, and view-through gets 1 day. A conversion with no ad interaction inside its window is not counted at all.

The model splits the credit. Data-driven attribution compares the paths of people who convert with the paths of people who don't. It then gives more credit to the ad interactions that seem to make the difference. It only looks at Google's own inventory: Search, Shopping, YouTube, Display and Demand Gen. Last click is the only other option left, since Google retired first click, linear, time decay and position-based.

The count decides repeats. Every conversion is the default for website actions, so two purchases after one click count twice. For a shop, that is what you want.

One quirk catches everybody once. The main Conversions column files each conversion on the date of the click, not the purchase. A sale today can raise last week's numbers.

What one store's data shows

Store A is one store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue, not spend, clicks or order counts.

Start with timing. The slowest group on the Journeys sheet is four to nine touches, at 16.9 days to buy. Ten or more touches take 16.0 days, also on the Journeys sheet. Both sit inside Google's default 30-day click window. If a Google click opened one of those journeys, the window would usually still be open at checkout. In this store, the window is unlikely to be what cuts Google out.

Then the 1-touch row: 79.5% of revenue on the Journeys sheet, at 0.5 days to buy. A journey with one touch gives a model nothing to split. If that touch was a Google ad, data-driven and last click both hand it the full sale. So for most of this store's revenue the model setting changes nothing, and the argument is about the other fifth.

Last, the touched view on the Channels sheet sums to 110.4%, by design. A journey that touched two channels counts in full for both. Ad platforms do the same thing one level up. Shopify's help pages say an email and a Google Shopping ad clicked by the same buyer can each record a separate conversion.

None of this shows what Google Ads cost or caused for this store. The export holds shares of revenue, not Google's click or cost data.

Why does "data-driven" mislead people?

Because it sounds like a verdict on cause, and it is a split of credit. Google's model only sees Google ad interactions. If a buyer found you through Instagram and came back through a brand search ad, the model sees one Google click and gives it everything. Google's own help says the Path metrics report only reflects the keywords and ads in your Google Ads account.

GA4 can widen the lens. Its own reports credit paid and organic channels by default, while conversions shared with Google Ads default to Google paid channels. Same tag, two rulebooks.

The click date misleads too. Last week's total can keep climbing after the week ends, as late purchases file back to their clicks.

What can the Google number not tell you?

Whether the click caused the sale. A shopper who types your brand name and clicks your ad was already on the way to you. Attribution, data-driven or not, divides credit for sales that happened. Whether they would have happened anyway is a question for a holdout test.

It also cannot show how Google overlaps with Meta or email. Each platform grades its own homework, and the touched view above shows what happens when everyone gets full marks.

What to do this week

1. Read the settings on your purchase conversion action. In Google Ads, open Goals, then Conversions, then Summary, pick the action and select Edit settings. Pass: the model, windows and count were chosen on purpose, with Every for purchases. Fail: settings nobody remembers choosing, or two purchase actions counting the same sale.
2. Add Conversions (by conv. time) next to Conversions. Pass: the conversion-time totals sit closer to Shopify for the same dates. Fail: they are still far apart, which points at tagging rather than timing.
3. Open Path metrics and measure from the first ad interaction. Go to Attribution in the Goals menu, open the Path metrics tab and change Measure from. Pass: most conversions land well inside your window. Fail: a pile-up near the window's end, which means the window is cutting off sales.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: About conversion measurement (Google Ads Help); About conversion windows (Google Ads Help). About attribution models (Google Ads Help); About data-driven attribution (Google Ads Help). About conversion counting options (Google Ads Help); Understand your conversion tracking data (Google Ads Help). About attribution reports (Google Ads Help); Select attribution settings (Google Analytics Help). Measuring marketing performance (Shopify Help Center).

Related answers

• All Google Ads questions
• How to check Google Ads attribution, step by step
• How does Google Ads attribution work for electronics?
• Data-driven attribution, defined
• Causal attribution, explained