Marketing Attribution Models Explained

So you’ve nailed your last month of SEM, web-traffic is up and your CEO thinks you’re a CMO rockstar. Boom! Go pour yourself a well-earned Gin & Tonic.

But then Monday comes and your CEO wants to chat about attribution. You see, she attended a terribly informative event last week which delved into attribution and what leading companies are doing in this area. She’s fully-loaded with buzz words and is newly-curios about how you’re handling this.


What do you mean last click?!


Gulp. Attribution was on your plan to address sometime this year… but it slipped down that ever growing list of priorities (Ahhh GDPR!). You have a decent high-level understanding of attribution, but the details are vague.

Fear not, our handy guide will help you grasp the finer points.


1. Last Click

Regardless of what happens, 100% of the credit goes to the last click. The hard work done to get the user on-site that first time, or coax them back for a 2nd or 3rd visit before conversion is ignored. It’s all about the last action taken. This is Harry Kane taking 100% of the credit for all his goals, regardless of his teammates’ efforts.


2. Last Click non-Direct

A modification of the previous model, this assumes that a direct visit which leads to a conversion was a mere formality. The decision was made, the conversion was inevitable. The user just needed to get back to the site to wrap things up, so 100% of the credit goes to the previous channel.


3. First Click

The inverse of last-click – this gives 100% of the credit to the first-click. This ignores any work done by other channels to nurture and ultimately convert a lead – it’s all about that first interaction. This is David De Gea claiming 100% credit on all goals scored that originated from his goal kicks, regardless of how many other passes were involved.


4. Linear

Getting a little more sophisticated now, Linear gives some credit to all platforms involved in a conversion. Downsides are that it is still assumption based, but it’s an improvement in that it doesn’t depend on just one channel. This is the West Brom midfield + striker sharing credit equally on goals they were all involved in… which isn’t very many these days.


5. Time Decay

Another step up in nuance, time decay gives to credit to multiple channels, but weights the credit based on recency. So a paid search ad that leads to an immediate conversion will generate a large amount of credit for the Paid Search channel, but other channels involved will receive some credit, which reduces the further back the interaction was.


6. U-shaped / Position-based

U-shaped is seen by many as the most sophisticated model of attribution outside of Algorithmic. Most commonly, 40% of credit it given to the first and last channels, with the remaining 20% split between all other channels, although some variation on these numbers is possible. Intuitively, it’s a model that seems reasonable – getting leads into a funnel and ultimately converting those leads are two critical stages. However, reasonable as it sounds, ultimately it is a model that relies on assumptions. The 40-20-40 split is a completely arbitrary weighting. Why not 42-16-42, or 38-16-38?


7. Algorithmic

Enter algorithmic. The prince of attribution models, algorithmic is the only of these 7 models that is rooted in real data, maths and statistics. Algorithmic attribution investigate conversion (and non-conversion) paths statistically to determine the true weighting of each channel. There are a few different models of algorithmic attribution – we use Markov. The result is a completely custom model which weights the impact of each channel. These models require maintenance (as the data changes, so does the model), but are the only way to get truly scientific about attribution.