GA4 Attribution Models on Shopify: Data-Driven vs Last Click Explained

TL;DR

GA4 switched its default attribution model to data-driven in November 2023, replacing last-click as the out-of-the-box setting. Data-driven uses machine learning to split conversion credit across every touchpoint that contributed to a sale, while last-click assigns all credit to the final channel. For Shopify merchants, the practical effect is that your channel performance reports in GA4 may show fractional conversions (0.3 from email, 0.7 from paid search for a single order). This is not a bug. This article explains what each model does, when to use which, and how server-side tracking feeds cleaner data into whichever model you choose.

Key Takeaways

• Data-driven attribution distributes fractional conversion credit across touchpoints using Google's ML model, so you might see 0.4 conversions from organic and 0.6 from paid search for the same order.
• Last-click attribution gives 100% credit to the last non-direct channel before conversion. Simpler, but penalizes top-of-funnel channels like organic social and display.
• GA4 deprecated five models in 2023: first-click, linear, time-decay, position-based, and last Google Ads click are no longer available. Only data-driven and last-click remain.
• Data-driven needs volume to work well. Google recommends at least 400 conversions per month for reliable model training. Below that threshold, last-click may be more stable.
• Server-side tracking improves attribution quality for either model by ensuring purchase events and campaign data reach GA4 even when browser pixels are blocked by ad blockers or Safari ITP.

What is data-driven attribution in GA4?

Data-driven attribution (DDA) is Google's machine learning model that analyzes all the touchpoints in a customer's path to conversion and assigns fractional credit to each one based on how much it actually contributed to the outcome. Rather than applying a fixed rule ("last click gets everything" or "split evenly"), DDA looks at patterns across your real conversion data to determine which channels, campaigns, and interactions are genuinely driving results.

Here is what that looks like in practice. A customer visits your Shopify store through a Google Ads click on Monday, comes back via an email link on Wednesday, and finally converts through a direct visit on Friday. Under last-click, the email gets 100% credit (direct visits are excluded). Under data-driven, GA4 might assign 0.4 conversions to the Google Ads click and 0.6 to the email, based on how those channels perform across all your conversions.

The model trains on your specific data. A store where email consistently precedes high-value purchases will see email receive more credit than a store where email clicks rarely lead to conversions. This is the core advantage: the model adapts to your actual customer behavior rather than applying a one-size-fits-all rule.

Google made DDA the default for all GA4 properties in November 2023. Every new property starts with it, and existing properties that had not changed their settings were automatically switched.

One important caveat: DDA requires conversion volume to function properly. Google recommends at least 400 conversions per month for reliable credit distribution. Below that, the model does not have enough data points to distinguish between channels, and the fractional assignments become noisy. For a Shopify store doing 50 orders per month, data-driven attribution is technically active but practically unreliable.

What is last-click attribution?

Last-click attribution assigns 100% of the conversion credit to the last non-direct channel the customer interacted with before purchasing. Direct visits (where someone types your URL or uses a bookmark) are excluded from consideration. If the only touchpoint was a direct visit, that session still receives credit.

This model is straightforward: one conversion, one channel gets the credit. There are no fractional numbers. Your reports show whole numbers.

For Shopify merchants with a simple marketing stack (one paid channel, email, and organic search), last-click often tells a clear enough story. You can see which channel closed the sale. The tradeoff is that channels responsible for discovery, like organic social posts or display ads, receive zero credit even when they introduced the customer to your brand.

Last-click was GA4's default before November 2023, and it remains available as an option. Many merchants still prefer it because the reports are easier to interpret and act on.

What happened to first-click and other models?

In late 2023, Google removed five attribution models from GA4:

• First-click (100% credit to the first touchpoint)
• Linear (equal credit to every touchpoint)
• Time-decay (more credit to touchpoints closer to conversion)
• Position-based (40% first, 40% last, 20% distributed to middle)
• Last Google Ads click (100% credit to the last Google Ads interaction)

These models are no longer available in GA4 properties. Google's reasoning, stated in their announcement, was that data-driven attribution is strictly better than fixed-rule models because it adapts to actual conversion paths rather than applying arbitrary rules.

For merchants who relied on first-click to evaluate top-of-funnel campaigns, data-driven attribution is the replacement. DDA will credit discovery channels when the data supports it, but not by a fixed 100%. The credit is proportional to how often that channel actually initiates conversions.

If you need first-click-style analysis, the closest alternative is GA4's Path Exploration report, which lets you see the first touchpoint in a conversion path without changing your reporting attribution model.

How does attribution affect what Shopify merchants see in GA4?

Attribution model choice changes the numbers in GA4's acquisition and campaign reports. The same underlying orders produce different channel breakdowns depending on which model is active.

Here is a concrete example with a single order:

Under data-driven, all three channels share credit. Under last-click, email gets everything. Multiply this across hundreds of orders, and the channel performance picture can shift significantly.

This matters for budget allocation. If your GA4 reports use data-driven and show Google Ads contributing 35% of conversions (fractional), turning off Google Ads would not necessarily drop conversions by 35%, because some of those fractional conversions would have happened anyway through other channels. But it does tell you that Google Ads is part of the path more often than last-click would suggest.

Shopify merchants commonly notice these differences in the Acquisition > Traffic acquisition and Advertising > Attribution paths reports. Revenue numbers will also differ between models because GA4 distributes both conversion count and revenue proportionally.

One thing that does not change: the raw event data. Your purchase events in GA4's DebugView and event-level exports always show the full order value. Attribution only affects the aggregated reports.

How do I change my GA4 attribution model?

The setting lives in your GA4 property's admin panel:

1. Open Google Analytics and navigate to Admin (gear icon, bottom left).
2. Under the property column, click Attribution settings.
3. Under Reporting attribution model, choose either "Data-driven" or "Last click."
4. Click Save.

The change applies retroactively to your reports. GA4 recalculates past data using the new model, so your historical reports will shift when you switch. This is not reprocessing raw data. GA4 stores the touchpoint paths and reapplies the attribution logic on the fly.

There is also a lookback window setting on the same page. This controls how far back GA4 looks for touchpoints. The default is 30 days for acquisition events and 90 days for all other events. For most Shopify stores, the defaults work well. Stores with longer purchase cycles (B2B, high-ticket items) may want to extend the window.

You can switch between models at any time without losing data. Try both, compare the reports, and use whichever gives you more actionable insight for your specific marketing mix.

Does server-side tracking affect attribution quality?

Yes, and this is where attribution models and tracking infrastructure intersect.

GA4's attribution models can only work with the data they receive. If a purchase event never reaches GA4 because an ad blocker suppressed the browser pixel, that conversion is invisible to the attribution model. It cannot assign credit to any channel for a sale it does not know about.

Server-side tracking closes this gap. When a Shopify app sends purchase events through GA4's Measurement Protocol (server-to-server), those events arrive regardless of what is happening in the buyer's browser. Ad blockers, Safari ITP, and Shop Pay's cross-domain redirect all become irrelevant for the server-side copy.

But attribution quality depends on more than just the purchase event arriving. GA4 needs to know which campaign, source, and medium brought the customer in. Without that context, the purchase event lands in GA4 as "(direct) / (none)" and the attribution model has nothing to distribute credit to.

This is where campaign_details events matter. WeltPixel Conversion Tracking sends a campaign_details event server-side via Measurement Protocol before every purchase event, carrying the source, medium, and campaign parameters from the customer's session. This means GA4's attribution model has clean session data to work with, even when the browser-side campaign tracking was interrupted.

The practical effect: data-driven attribution becomes more reliable because the model is trained on a more complete dataset. Last-click also benefits because fewer orders fall into the "(direct) / (none)" bucket.

For a full walkthrough of how server-side tracking works across GA4 and other channels, see the Shopify server-side tracking guide.

Which model should you use?

The right choice depends on your conversion volume and marketing complexity.

Use data-driven attribution if:

• Your store processes 400+ orders per month (giving the ML model enough data to train on)
• You run multiple paid channels (Google Ads, Meta, email, organic) and need to understand how they interact
• You are making budget allocation decisions between channels and need to see top-of-funnel contribution
• You are comfortable with fractional conversion numbers in your reports

Use last-click attribution if:

• Your store processes fewer than 400 orders per month
• You primarily use one or two marketing channels
• You want simpler reports with whole-number conversions
• Your team is accustomed to last-click and uses it for benchmarking against prior periods

There is no wrong answer. Data-driven is generally more accurate for stores with sufficient volume, but accuracy matters less than consistency when you are tracking performance over time. If you switch models mid-quarter, your period-over-period comparisons break because the same underlying data produces different numbers under each model.

A practical approach: start with data-driven (the default). Check your GA4 conversion count in Admin > Attribution settings. If you are above the 400 monthly threshold, keep it. If you are well below, switch to last-click for cleaner reporting and revisit when your volume grows.

For a deeper look at how GA4 connects to your Shopify store, including server-side setup and event configuration, see the complete Shopify GA4 setup guide.

FAQ

Does changing my attribution model affect my raw GA4 data?

No. Your raw event data (purchase events, page views, etc.) stays exactly the same. Attribution models only change how GA4 distributes credit in aggregated reports. You can switch between models at any time without losing or altering any underlying data.

Why do I see decimal conversion numbers like 0.3 in my GA4 reports?

That is data-driven attribution at work. When a customer interacts with multiple channels before purchasing, DDA splits the conversion credit proportionally. A single order might show as 0.3 conversions from email and 0.7 from paid search. The total across all channels always adds up to the actual number of conversions.

Can I use data-driven attribution with fewer than 400 monthly conversions?

Technically, yes. GA4 will still apply the data-driven model. But the model does not have enough conversion data to reliably distinguish between channels, so the credit distribution may be inconsistent or dominated by noise. Google set the 400-conversion recommendation as the threshold where DDA starts producing stable, actionable results.

Does attribution model choice affect Google Ads bidding?

Google Ads has its own attribution settings, separate from GA4's reporting attribution model. When you import GA4 conversions into Google Ads, the attribution model applied in GA4 determines how conversion value is distributed. If GA4 uses data-driven and assigns 0.4 conversions to a Google Ads click, that is the value Google Ads Smart Bidding sees. This can change bid behavior compared to last-click, where the same click would receive either 1.0 or 0.0 conversions.

What is the difference between GA4 attribution and Shopify's order attribution?

Shopify's order attribution (visible in your Shopify admin under Orders) uses its own last-click model based on UTM parameters and referrer data. This is completely separate from GA4. You will often see discrepancies between what Shopify reports as the order source and what GA4 reports, especially for orders that involved multiple touchpoints. Neither is wrong. They are measuring different things with different methodologies.

WeltPixel Conversion Tracking sends campaign_details events server-side before every purchase event, ensuring GA4's attribution model has clean source/medium/campaign data to work with. Install on the Shopify App Store.