AI Search and Marketing Attribution: What Every Shopify Store Needs to Know

AI tools like ChatGPT, Perplexity, and Google Gemini are fundamentally changing how consumers discover and purchase products online. Instead of searching Google or scrolling Instagram, shoppers are now turning to AI agents for personalised recommendations. This evolution presents exciting opportunities, but it also breaks the traditional rules of marketing attribution.

In this post, we’ll explore how AI search is disrupting product discovery, why traffic from AI tools often disappears from analytics platforms like Google Analytics 4, and how Shopify brands can fix this with smarter attribution tools like Littledata.

This blog draws insights from a recent YouTube walkthrough by Edward Upton, CEO of Littledata, where he tested real-world AI search journeys using ChatGPT, Perplexity, and Google Gemini. The video highlights how inconsistent linking, poor UTM usage, and redirect issues can undermine Shopify store attribution.

The Rise of Conversational Commerce

AI-powered search is becoming a core part of the customer journey. And shoppers are more than ready:

• 70% of consumers are open to using AI to book flights
• 65% would use AI to book hotels
• 50–60% would use AI to buy products like clothing, beauty, electronics, and medicine
  (Source: SellersCommerce)
  

This marks a shift from traditional keyword-based discovery to intent-based, conversational journeys. Instead of typing “best slingback flats under £150,” customers are now asking, “What are some stylish flats with no heel under £150 I can wear to work?”

As a result, Shopify brands are seeing more traffic from these AI tools, but struggling to track where it came from.

Why AI Traffic Is Invisible in Google Analytics

Traditional attribution models rely on:

• Clean, one-click referral links
• UTM parameters (like utm_source and utm_medium)
• Predictable user journeys from discovery to conversion

AI search breaks all of these assumptions.

What’s going wrong:

• Incomplete UTM parameters: ChatGPT might pass a utm_source=chatgpt.com, but without utm_medium, GA4 won’t categorise it correctly.
• Redirects destroy referrer info: If an AI tool links to an old or out-of-stock product page, redirects to your homepage can strip away the attribution.
• Multi-step journeys: Perplexity might link to an internal result page first, then the product page. Gemini might not link at all.
• Session loss: When users browse across devices or tabs, the initial source often gets lost.

All of this results in AI-driven traffic showing up as “direct” or “unassigned” in GA4, leaving you blind to which channels are driving valuable sessions.

Testing the Tools: How Each AI Platform Handles Attribution

In the video experiment, Edward Upton searched for a Rothy’s Almond Slingback in hibiscus pink across ChatGPT, Perplexity, and Google Gemini to simulate how customers are discovering products via AI.

ChatGPT: Strong Discovery, Weak Consistency

• Starts with generic results; needs refinement to get to specifics
• Displays product images and sometimes links directly to the correct product page
• Occasionally uses utm_source=chatgpt.com, but rarely includes a full set of UTM tags
• Redirects and outdated links often result in broken attribution

Read: How Session Enrichment improves attribution for Google Analytics

Perplexity: Smarter Queries, but Attribution Falls Short

• Excellent at interpreting intent (e.g., “elegant shoe under £150”)
• Uses multi-step links, which damages referral tracking
• Rarely includes any UTM parameters
• May recommend physical stores, making digital attribution impossible
• Accurate product results, but limited traceability

Google Gemini: Helpful Guide, Lacks Purchase Intent

• Offers helpful commentary, but not always focused on products
• Requires explicit prompts to surface shopping links
• Links usually go to homepages or marketplaces like Amazon
• No UTM tagging; attribution is almost always lost

How to Fix It: Attribution Solutions for AI Traffic

Shopify merchants can’t afford to lose track of high-intent traffic coming from AI platforms. That’s where Littledata comes in.

Littledata solves this problem by:

• Automatically enriching traffic data with missing UTM parameters
• Preserving source and medium data even through redirects
• Streaming server-side events to Google Analytics 4 and Meta Ads Manager
• Capturing pre-checkout activity like product views and add-to-cart. Even if the journey started with AI

This gives you a clearer picture of where traffic is coming from, how customers discovered your products, and what’s actually driving conversions.

What Shopify Brands Need to Do Next

AI search is here to stay, and growing fast. To adapt, your team needs to:

1. Ensure your store and products are discoverable via AI platforms
2. Upgrade your analytics stack to account for new traffic sources
3. Track AI traffic accurately to inform budget and content decisions

Simply appearing in AI search results isn’t enough. You need to know what’s working, and why.

Read: Winning on ChatGPT: How to Drive Traffic to Your Shopify Store

Start Tracking Smarter with Littledata

Don’t let valuable traffic slip through the cracks.

Littledata helps Shopify brands track traffic from ChatGPT, Perplexity, Google Gemini, and other modern discovery tools, so you can make informed decisions, protect your ad spend, and grow confidently in an AI-powered world.

Want to see these AI product searches in action?

Watch the full breakdown by Edward Upton on YouTube to see how AI tools are changing discovery and what Shopify brands can do about it.