# Google Data-Driven Attribution

> Learn how Google Ads data-driven attribution assigns credit, what data the model reads, and how AnyTrack feeds it through the Conversion API.

Data-driven attribution (DDA) is the default attribution model in Google Ads. The model decides how much credit each ad interaction receives for a conversion. It reads the conversion data that your Google Ads account received.

AnyTrack has no data-driven model. AnyTrack sends the conversions that the Google model reads.

<Callout icon="📘" theme="info">
<p><strong>Why This Matters</strong></p>

<p>Google states that "the performance of data-driven attribution improves with more data". The model reads the conversions that Google Ads received. A conversion that never reaches Google Ads is absent from the model.</p>

<p>AnyTrack sends every conversion that you map to <a href='/docs/google-ads'>Google Ads</a> through the Conversion API, with the Google click ID and the customer data that <a href='/docs/google-ads-enhanced-conversions'>Enhanced Conversions</a> uses. Conversion Mapping decides which events reach Google Ads.</p>
</Callout>

## What Data-Driven Attribution Is

Data-driven attribution is a Google Ads attribution model. Google calculates the contribution of each ad interaction on the conversion path. To do this, Google compares the paths of customers who convert with the paths of customers who do not convert. The model then assigns fractional credit to each interaction.

A rule-based model applies a fixed rule instead. Last click gives all credit to the final ad interaction. Data-driven attribution applies no fixed rule, and the credit split changes when the underlying data changes.

Google applies the model to website conversions, store visits, and Google Analytics conversions. The supported networks are Search (including Shopping), YouTube, Display, and Demand Gen.

## Eligibility and Data Volume

Every conversion action is eligible. Google states: "All conversion actions are eligible for data-driven attribution (DDA), regardless of conversion or interaction volume."

Volume changes the accuracy of the model, not the eligibility of the account.

| Question | Google position |
|---|---|
| Minimum volume to use the model | None. Every conversion action is eligible |
| Recommended volume | At least 200 conversions and 2,000 ad interactions in supported networks within a 30-day period |
| Below the recommendation | The model still functions. Google assigns credit with less precision |

<Callout icon="📘" theme="info">
<p>At 200 conversions and 2,000 ad interactions per 30 days, the model identifies patterns and assigns credit more precisely. Source: <a href='https://support.google.com/google-ads/answer/6394265' target='_blank' rel='noopener'>About data-driven attribution</a> in Google Ads Help.</p>
</Callout>

## What AnyTrack Sends to the Model

AnyTrack does not model attribution for Google Ads. AnyTrack sends the conversion data that the model reads. Conversion Mapping decides which events AnyTrack sends to Google Ads. For each mapped conversion, AnyTrack sends:

- The conversion event and its value.
- The Google click ID (`gclid`, `gbraid`, or `wbraid`) that AnyTrack read from the landing page URL.
- The hashed customer data that [Enhanced Conversions](/docs/google-ads-enhanced-conversions) uses.

Before AnyTrack sends a conversion, AnyTrack [deduplicates](/docs/deduplication) it. One sender per platform prevents a double count on both the converting and the non-converting paths.

## Why the Feed Changes What the Model Sees

The model reads the conversions that Google Ads received. Three gaps keep conversions out of that set. AnyTrack closes each one.

- **Browser limits.** If a browser blocks the Google Tag, the browser records no conversion. AnyTrack records the conversion on the server and sends it through the Conversion API.
- **Conversions that happen outside the browser.** A purchase confirmed in Shopify, a lead status change in HubSpot, and an affiliate network postback all happen on another server. AnyTrack receives each one as [server-side tracking](/docs/server-side-tracking) and sends it to Google Ads.
- **Conversions with no active session.** AnyTrack stores the click ID of the visit in the Data Store. When the conversion arrives days later, AnyTrack matches it to the original ad click and sends the conversion with its Google click ID.

Each closed gap adds a real conversion path to the data that the model compares. The model assigns credit on a more complete set of paths.

## How AnyTrack Attribution Differs

AnyTrack reports with two click-based attribution models: First Ad Click and Last Ad Click. Switch between them with the Attribution dropdown in the [Campaign Report](/docs/campaign-report). AnyTrack has no data-driven model, no multi-touch model, and no post-view model.

The two systems answer different questions. Read them together, not as one number.

| | Google Ads | AnyTrack |
|---|---|---|
| **Attribution model** | Data-driven attribution by default. Rule-based models are also available | First Ad Click or Last Ad Click |
| **Credit split** | Fractional credit across the ad interactions on the path | Full credit to one ad click |
| **Post-view conversions** | Counted | Not counted |
| **Conversion modeling** | Google estimates conversions that it could not observe | None. AnyTrack reports the conversions it received |
| **Where you read the result** | The Conversions column in Google Ads, and Attributed Results in the Campaign Report | The conversion counts in the Campaign Report |

The two numbers differ for these reasons. Expect a gap between them. For the full list of causes, read [Google Ads Conversion Goals](/docs/google-ads-primary-conversions-account-goals).

## Next Steps

<Cards columns={2}>
<Card title="Enhanced Conversions" href="/docs/google-ads-enhanced-conversions">
Send hashed customer data with each mapped conversion so Google matches more of them.
</Card>
<Card title="Google Ads Integration" href="/docs/google-ads">
Connect Google Ads and select the conversion actions AnyTrack sends.
</Card>
<Card title="Conversion Goals" href="/docs/google-ads-primary-conversions-account-goals">
Set primary actions and account-default goals so the right conversions count.
</Card>
<Card title="Troubleshooting" href="/docs/google-ads-troubleshooting">
Find why conversions do not appear in your Google Ads reports.
</Card>
</Cards>

<FaqAccordion
  title="Data-Driven Attribution FAQ"
  icon="fa-duotone fa-circle-question"
  items={[
    {
      question: "Does AnyTrack use data-driven attribution?",
      answer: "No. AnyTrack reports with two click-based models, First Ad Click and Last Ad Click, switchable in the <a href='/docs/campaign-report'>Campaign Report</a>. Data-driven attribution is a Google Ads model. AnyTrack sends every conversion that you map to Google Ads through the Conversion API, with the Google click ID and the customer data that Enhanced Conversions uses. The Google model runs on the data that Google Ads received, and the AnyTrack feed is that data."
    },
    {
      question: "How many conversions do I need for data-driven attribution?",
      answer: "None to turn it on. Google states that all conversion actions are eligible for data-driven attribution, regardless of conversion or interaction volume. Google recommends at least 200 conversions and 2,000 ad interactions in supported networks within a 30-day period. Below that volume the model still functions, and Google assigns credit with less precision."
    },
    {
      question: "Why do Google Ads and AnyTrack report different conversion numbers?",
      answer: "The two systems apply different attribution. Google Ads splits fractional credit across the path, counts post-view conversions, and models conversions that it could not observe. AnyTrack gives full credit to one ad click and reports only the conversions it received. Read <a href='/docs/google-ads-primary-conversions-account-goals'>Google Ads Conversion Goals</a> for the full list of causes."
    },
    {
      question: "Which campaign types does data-driven attribution cover?",
      answer: "Google applies data-driven attribution to website conversions, store visits, and Google Analytics conversions. The supported networks are Search (including Shopping), YouTube, Display, and Demand Gen. Check the attribution model on each conversion action in your Google Ads account. A conversion action can use a rule-based model instead."
    },
    {
      question: "Does Enhanced Conversions affect data-driven attribution?",
      answer: "Yes, indirectly. Enhanced Conversions sends hashed customer data with each conversion, so Google matches more conversions to an ad interaction. A matched conversion enters the data that the model compares. An unmatched conversion does not. Read <a href='/docs/google-ads-enhanced-conversions'>Google Ads Enhanced Conversions</a> to set it up."
    }
  ]}
/>
