The New Rules of Multi-Touch Attribution for Ecommerce Brands

You already know last-click isn’t the full story. Tracking breaks. UTMs get overwritten. Third-party pages kill session integrity. All of it makes it harder to trust what’s actually working. So, what can you do?

In this episode of The Conversion Tracking Playbook, Brad Redding (Elevar) talks with Jacob Rokeach (Fluency Firm) and Fei Wang (Source Medium) about how top eCommerce brands are moving beyond surface-level metrics to build attribution systems they actually trust.

They dig into what clean data really looks like, how to evaluate performance across complex journeys, and why the best insights don’t come from your dashboard. Here’s what you’ll find below:

• Why Multi-Touch Attribution Still Matters

• Why GA4 Isn’t Enough—And What Is

• Evaluating What’s Really Working

• A Lean Stack for Clear Attribution

• Attribution in the AI Era

Why Multi-Touch Attribution Still Matters

Customers don’t follow straight lines anymore. One might see a creator ad, sign up for SMS, click an email flow, and convert after a branded Google search. Another might hear about you on a podcast, check out your Instagram, and buy through an affiliate. These are real journeys, and last-click won’t capture any of it.

Jacob Rokeach put it bluntly, “A brand that’s only looking at last click – I think that’s a mistake.”

Multi-touch attribution (MTA) isn’t about chasing every impression. It’s about understanding influence across channels—especially now that Meta, CTV, TikTok Shop, and creators can all engage the same customer within days.

Relying on last-click doesn’t show you what’s working. It only tells you what happened last.

Why GA4 Isn’t Enough – And What Is

Let’s be honest: you probably have GA4 installed. Maybe it’s helpful for basic reporting. But is it helping you make strategic decisions? Probably not.

Default client-side setups limit visibility to last-touch, obscure assist pages, and often can’t track across subdomains or logins. Even when configured correctly, GA4 lacks flexibility for multi-touch analysis unless your underlying tracking is solid.

Fei pointed to the root cause, “Most of the GA4 installations are client-side, and then you’re gonna be limited to Google’s last touch attribution, right?”

If your UTMs are inconsistent or your campaign naming is sloppy, it doesn’t matter what reporting platform you use. Attribution breaks long before your data hits a dashboard.

Evaluating What’s Really Working

Sometimes attribution can’t answer the question directly but behavior change can.

If you pause Pinterest and see no change in new users or sessions, that’s a signal. If removing affiliate spend drops branded search performance, that’s another signal. Attribution is less about perfect precision and more about directional insight that drives smarter action.

Fei shared that Source Medium often uses a complexity score to guide decisions factoring in number of channels, funnel variants, and landing pages. In other words: the messier your system, the more important it is to build clarity into your framework.

A Lean Stack for Clear Attribution

Massive budgets aren’t required for attribution to work. What matters is the order of operations.

Start with server-side tracking. Elevar captures enriched, consistent clickstream data across every platform, campaign, and funnel. Then, bring that data together in Source Medium to connect campaign results with customer behavior. GA4 still plays a role—but more as a directional reference, not your single source of truth.

“You don’t need a 200-person team anymore,” said Jacob. “If your tracking is consistent and your data is clean, you can answer the questions that matter.”

This streamlined stack gives you flexibility without losing visibility. It lets you test, iterate, and improve without relying on modeled ROAS or stitched dashboards.

Attribution in the AI Era

Touchpoints are evolving and attribution has to evolve with them.

Customers now ask AI tools like Perplexity, ChatGPT, or Gemini to recommend products and compare prices—before ever landing on a brand’s website. Those tools pull data directly from your catalog, skipping your ads and landing pages entirely.

As Fei explained, “The product discovery behavior is you’re going into Perplexity… it’s hitting the API and getting the price and availability of a bunch of different brands.”

Attribution strategies have to adapt because the buyer’s journey already has.

Attribution You Can Trust Starts With Data You Control

There’s no perfect model. But there is a smarter one – built on clean, consistent data and structured for real world decisions.

“You don’t need to build the perfect system,” said Jacob. “You just need to start asking better questions of your data.”

That’s what Elevar helps performance teams do. From server-side tracking to enriched events and real-time pass-through, Elevar gives your team the attribution infrastructure it needs.

Book a demo to learn how Elevar helps high-performance eCommerce brands get attribution right.