The Attribution Crisis: How Can Marketers Navigate a Post-Cookie World?

I’ve been hearing this asked more and more from my network, and honestly, it’s no surprise with the changes we’re facing:

• "One attribution model shows 2x ROAS, another shows 5x, and another shows 15x... where's the truth?"
• "GA4 tracking feels inconsistent and unreliable."
• "Our digital agency can’t prove why it’s spending in each category—it all feels like guesswork."

These are common frustrations being felt in our industry today.

The reality is, traditional attribution is failing to give marketers the clarity they need.?

GA4, despite positive intentions, has left many scrambling to make sense of incomplete data.?

The impact of privacy laws and the nearing end of third-party cookies, has meant tracking performance feels more and more like a guessing game.

This comes at a time where 77% of CMOs are feeling increased pressure to prove short-term ROI. It’s hard to make informed decisions when the data itself is unreliable.

Many of us have been there—consistently spending tens of thousands on a particular channel without solid evidence that it’s actually contributing to anything. But with budgets tightening, justifying our marketing spend has never been more critical. Without accurate measurement, it’s almost impossible to provide that justification.

A recent Gartner study showed that 60% of CMOs are now planning to cut their marketing analytics teams, citing "failed promised improvements." There’s a critical gap between the promise of marketing analytics and the reality of what many teams are experiencing. Bridging this gap requires rethinking measurement altogether.

Google’s recently published Modern Measurement Playbook is an important read. They recommend combining Marketing Mix Modelling (MMM) with Attribution and Incrementality Testing for a more comprehensive measurement strategy.

MMM provides a more holistic view than attribution by analysing how all marketing channels—both online and offline—drive overall business results. It offers a clear understanding of the exact contribution per channel and where to focus your efforts for maximum impact.

Unlike attribution, which focuses on individual user tracking, MMM doesn’t rely on granular user data. This makes it a crucial tool in today’s privacy-first environment, where user-level tracking is becoming increasingly restricted.

When combined with Incrementality Testing, MMM becomes even more powerful, offering businesses a way to understand which channels are truly driving incremental growth.

One of the major challenges with adopting MMM is the complexity involved. Open-source tools like Facebook’s Robyn haven’t seen widespread adoption because, let’s be honest, implementing a Bayesian Marketing Mix Model isn’t something most of us are wired to handle.

That’s why many turn to experts. I personally recommend a company like Seeda , Specialists in this space, they provide a precise understanding of how each marketing channel contributes to overall success. What’s exciting is that they don’t stop there. Seeda also provides a clear plan for how to spend your budget across channels moving forward.

As Michael Kingston summarises it: “This is exactly where you should be spending, this is exactly how much and here's exactly why.”

Here’s how they do it:

1. Data Activation: Integrates all your data for a holistic view.
2. Channel Contribution: Breaks down how each channel impacts your outcomes.
3. Saturation Modelling: Identifies diminishing returns and underutilised channels.
4. Optimal Budgeting: Provides the best allocation mix for ROI.
5. Incrementality: Ongoing testing to prove which campaigns drive growth.

If you’re trying to navigate the complexity of modern marketing or the challenges of today’s privacy-first world, I’d highly recommend a conversation with them. Michael Kingston is an expert in this field, and I’ve learned a lot through our talks.

Have you explored MMM or other measurement strategies? I would love to hear your thoughts.