What is Google Ads data-driven attribution?
Data-driven attribution is the default model in Google Ads. It splits each conversion's credit across the Google ad interactions before it, based on how converting and non-converting paths differ in your account. It cannot see Meta, email or organic visits.
By Joris van Huët, Founder & CEOUpdated 7 min read
Run the numbers for your store: the free break-even ROAS calculator.
Data-driven attribution is the Google Ads model that splits the credit for each conversion across the Google ad clicks and engagements before it. It learns from your own account, comparing the paths of people who convert with those who don't. It is usually the default, and it never sees channels outside Google Ads.
Google's help page sums up the method in two moves. The model compares the paths of customers who convert with the paths of customers who don't. It then gives more credit to the ad interactions that make a conversion more likely.
Three details decide what that means for you. Each model is built for one advertiser, from that account's own conversion data. It only scores Google ads: Search including Shopping, YouTube, Display and Demand Gen. And its only rival is last click, because Google retired first click, linear, time decay and position-based.
How much data does it need? Google's main page says every conversion action is eligible, whatever its volume. It still recommends at least 200 conversions and 2,000 ad interactions within a 30-day period. Its Switch to DDA page is stricter: some action types need 300 conversions and 3,000 ad interactions over 30 days to qualify. When the pages disagree, the eligibility column in your own account settles it.
What one store's data shows
The usual answer stops at "a clever model shares the credit fairly". One store's export shows what any credit-sharing model has to work with.
One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts. Its paths are GA4's cross-channel paths, not Google Ads' own.
| What the export shows | Share of revenue | Source cell |
|---|---|---|
| Journeys with 1 touch (14 distinct paths, 0.5 days to buy) | 79.5% | Journeys sheet, 1 touch row |
| Journeys with 2 to 3 touches (296 distinct paths, 12.5 days to buy) | 12.2% | Journeys sheet, 2 to 3 touches row |
| Journeys with 2 or more touches (3,656 of 3,670 distinct paths) | 20.6% | Journeys sheet, 2 to 3, 4 to 9 and 10+ touches rows added |
| Direct, in last click, first click and touched views | 57.7% | Channels sheet, Direct row |
First, the shape. On the Journeys sheet, 14 distinct one-touch paths carry 79.5% of revenue. On the same Journeys sheet, the other 3,656 distinct paths carry 20.6% between them.
A model learns from patterns that repeat. Here the variety lives in thousands of thin paths, while the money sits in a handful of short ones. Those thin paths are where data-driven and last click part ways, and also where patterns are scarcest.
Second, the missing half. Google's model learns from people who never converted as well as from buyers. A GA4 paths export holds buyers only: Google's help says that without a key event, there is no path to show. So no paths export, this one included, can rebuild Google's model from the inside.
Third, the biggest channel. On the Channels sheet, Direct holds 57.7% of revenue in all three views. Google Ads' model never sees a direct visit, because it only scores Google ad interactions. GA4's own models skip direct visits too, unless the whole path is direct.
What the export cannot show: Google ad clicks, ad costs, or how Google split any sale. It is one store, not a benchmark.
Is it the same as data-driven attribution in GA4?
Same name, different job. GA4's version shares credit across the channels GA4 records. Google's own GA4 example spreads one conversion over paid search, social, an affiliate and search.
The Google Ads version only shares credit among Google ads. Meta, email and organic visits never enter its sums. So one sale can be all Google's in Google Ads and partly Meta's in GA4, with no tag broken anywhere.
History behaves differently too. A model change in Google Ads only affects conversions counted from then on, and the current model columns restate the past. In GA4, changing the reporting model rewrites history as well as the future.
GA4's version also keeps moving after the fact: it can reattribute a conversion for up to 7 days afterwards. Last week's split is not final yet.
Does data-driven attribution show what your ads caused?
Not on its own. Google describes GA4's version as counterfactual: it contrasts what happened with what could have happened. That is a sharper question than last click ever asks.
But the credit still has to add up. GA4's help says the fractions for one key event sum to 1.0, so every conversion gets handed out in full. A model built that way can rank the touches, but it cannot say a sale would have happened without any ad.
Google's own example shows the effect. In it, data-driven takes a brand campaign from 200 to 150 conversions and a generic campaign from 50 to 100. Add them up and the total is the same both ways: credit moved, and no sale was added.
That matters most for bidding. Google says the model you pick affects how Target CPA and Target ROAS bids are optimised. Leave the old targets in place after a switch, and Google's example warns of underbidding on brand and overbidding on generic.
Whether the generic campaign really sells more is a separate question. It takes a holdout: one group sees the ads, and a comparable group does not. Incrementality testing walks through how that works.
What to do this week
- Count your multi-click conversions. In Google Ads, go to Attribution in the Goals menu, open Path metrics and pick the Avg. interactions to conversion tab. Pass: a clear share of conversions follows two or more ad interactions, so the model has something to split. Fail: nearly all follow one, so data-driven and last click will report almost the same thing.
- Open Assisted conversions with campaigns as the dimension. It sits in the same Attribution menu. Pass: you can name campaigns with many click and view assists but few last-click conversions, the ones data-driven is most likely to lift. Fail: almost nothing assists, so a model change will move little.
- Check whether a switch is coming. Open the Switch to DDA tab under Attribution, then search your inbox for Google's notice. Google emails admins 30 days before an automatic switch. Pass: you know which purchase actions will move, and when. Fail: a switch lands unannounced and your bid targets go stale overnight.
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 data-driven attribution (Google Ads Help); About attribution models (Google Ads Help); About "Switch to DDA" (data-driven attribution) (Google Ads Help); Best practices for managing attribution model changes (Google Ads Help); About attribution reports (Google Ads Help); Get started with attribution (Google Analytics Help); Select attribution settings (Google Analytics Help); Key events attribution paths report (Google Analytics Help).
Related answers
Frequently asked questions
Do I need a minimum number of conversions for data-driven attribution?
Not according to Google's main help page, which says every conversion action is eligible regardless of volume. Google still recommends at least 200 conversions and 2,000 ad interactions in 30 days. Its Switch to DDA page lists higher bars for some action types, so read your eligibility column.Does data-driven attribution change my Smart Bidding?
Yes, if your bid strategy optimises for conversions. Google says the attribution model affects how bids are optimised for strategies such as Target CPA and Target ROAS. After a switch, Google advises moving each target by the same percentage that cost per conversion or ROAS moved.Why did Google switch my conversion actions to data-driven?
Actions on the retired rule-based models were upgraded to data-driven, and Google may pick others for an automatic switch. Admins get an email 30 days before one. You can let it happen, opt out, or switch earlier on the Switch to DDA tab.
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 ModelAn Attribution Model defines how credit for conversions is assigned to marketing touchpoints. It dictates how marketing channels receive credit for sales.
- 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.
- CounterfactualCounterfactual is a hypothetical outcome that would have occurred if a subject had received a different treatment.
- Google AnalyticsGoogle Analytics is a web analytics service that tracks and reports website traffic.
- IncrementalityIncrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
- Incrementality TestingIncrementality Testing measures the additional impact of a marketing campaign. It compares exposed and control groups to determine causal effect.