Marketing attribution: why perfect measurement is impossible

Marketing attribution is often treated as a solvable problem. If we collect enough data, apply the right model, and use increasingly sophisticated technology, we should be able to pinpoint exactly which activities drive performance. But much like trying to perfect a chocolate cake recipe, attribution quickly becomes complex. Every input affects every other input, variables interact in unpredictable ways, and what worked yesterday doesn’t necessarily work tomorrow. This article explores why accurate marketing attribution is so difficult, where attribution models fall short, and how marketers can still make better decisions despite these limitations.  

The core problem 

To see why, let’s return to the chocolate cake analogy. Even with countless chocolate cake options on supermarket shelves, there’s no single “perfect” recipe. Beyond the subjectivity of taste, it quickly becomes clear how complex the process is: every ingredient affects every other ingredient. Even in a perfectly controlled kitchen, where you could measure and experiment with every variable, understanding exactly how each input impacts the final cake is challenging. Marketing attribution works the same way. Accurately assigning business performance to a specific activity, when variables are uncontrolled or unmeasurable, is nearly impossible. Even with perfect data and sophisticated, real-time AI models, past results can’t fully predict future performance. Changes in culture, competition, or external events (like a pandemic) constantly shift the equation, making perfect attribution an unattainable goal.  

Why marketing attribution models struggle in the real world

Even the most advanced models suffer from bias, noise, and structural limitations.  

The ‘brand building’ challenge

Attribution models can devalue longer-term marketing channels. ‘Brand building’ typically drives small changes in behaviour across a mass audience. In a market where you are competing for a share of voice, your ‘brand building’ activity may be protecting sales, rather than driving additional sales. These nuances are hard to pick up in a model, particularly in low-frequency industries. We often heavily rely on econometric models, without giving sufficient thought to what we are trying to measure and why.  

The ‘interaction term’ challenge

The combined impact of two factors; let’s say discounting + PPC spend, is almost impossible to measure. Firstly, relationships like these are almost always non-linear which presents its challenges. But technology like smart bidding means that many factors change all at the same time. For example, a 20% discount with heavy PPC spend may not have been effective, but a 30% discount with heavy PPC spend might have pushed people over a threshold. The point is, relationships between channels, customer targeting, creative and content, and products and pricing are more complicated than ever, so are harder to measure.  

The ‘offline-online’ challenge

If you operate both online and offline, two opposing forces impact attribution. Firstly, people purchase in-store but if no shop existed, they would have purchased online. This means you’re likely to undervalue the impact of digital activity, as people compete in-store. Conversely, a halo effect is caused by the shop acting as an out-of-home placement. This means you are likely to falsely attribute performance to digital activity rather than a shop. It is almost impossible to measure these two effects on attribution.  

The ‘temporal’ challenge

If I hadn’t prompted a purchase today, would they have just purchased tomorrow? This is particularly true when trying to attribute discounting, and also any call-to-action. Bringing forward a purchase at a discounted rate represents a net loss to the business, but a net gain when reporting on that campaign alone.  

Should we stop using marketing attribution models altogether?

It’s not all doom and gloom. 

• Try to find short-term proxies for long-term value. For example, can you prove that reach and short-term unprompted awareness have a measurable effect on ROI by looking at patterns across the industry?
• Can you run smarter experiments? Tracking specific cohorts with specific marketing elements vs. a holdout can give you some understanding of the impact of some combination of factors.
• Can you see what cake worked well for other bakers? We have historic insight from publications such as The Institute of Practitioners in Advertising (IPA) showing aggregate ad-spend and results. We don’t need to learn from our own mistakes when we can learn from others.
• Most importantly, start by getting the data in one place! You can notice and test correlations between variables once you start plotting them out, without the need for a complex econometric model.

It can be tempting to simply deploy an off-the-shelf attribution model that fixes all our problems. But we must go back to the basics of understanding the data, and the variables we are trying to optimise.  

Using marketing attribution to guide smarter decisions

Just like baking the perfect chocolate cake, perfect attribution may never exist. Each ingredient interacts in complex, unpredictable ways, and external factors constantly shift the outcome. The goal isn’t to hit a perfect bullseye with every model or data point, it’s to understand the limitations of your models and make better, evidence-based decisions. By focusing on insight rather than perfection, marketers can treat attribution as a guide rather than a guarantee.  

Connect with the team

Speak to our team to explore how we help brands approach attribution, measurement and experimentation more intelligently.