How incremental is your attribution?

Every marketer wants to understand cause and effect – by understanding the impact that our actions have we can try to do more of what works and less of what doesn’t, and so become more effective over time. But measurement is tricky, and some of the jargon used doesn’t help. The terms incrementality and attribution are a case in point – too often they are used interchangeably, but they can mean very different things and it’s crucial to understand which one you are dealing with to avoid making costly mistakes. It is worth being pedantic about this topic – so you can be certain that the results you are presented with are actually telling you what you think they’re telling you. When large sums of money are being spent, the implications of misunderstanding can be huge.  

Attribution, in its most general sense, is defined as ascribing or regarding something as the effect of a stated cause. The confusion stems from the fact that there are general and specific definitions of attribution, and in a marketing context it may or may not be a true incremental effect. 

So how should we think about the terms? 

When some action is taken, incrementality relates to the true impact of that action above some baseline that would have occurred otherwise, i.e. isolating what impact would not have happened had the action not been taken. As a concept, this is required to understand the true payback from any marketing investments. Attribution may or may not measure a true incremental effect. Often, we measure what can be easily measured – for example how many people purchased while a product was on a particular promotion, or how many people clicked on a particular link and then went on to purchase. But attributing those sales to the promotion or the digital ad is a mistake – these things are useful to measure, but crucially they do not tell us the true impact of the promotion or the ad because we don’t know how many of those sales would have happened anyway (i.e. the baseline). In these examples, some people may have bought the product anyway at full price or navigated directly to the website if they had not seen the ad, so we need some other method to measure the true incremental impact that they had. 

 Marketing Mix Modeling (MMM) is perhaps the best known tool to measure incrementality. The gold-standard approach is to use scientific experimental design but, in the messy and complex real world, this is often not possible to scale effectively. Consequently, MMM is the next best solution. Confusingly MMM is sometimes generally defined as a type of attribution, but more often the word ‘attribution’ is shorthand for ‘digital attribution’, which is an alternative measurement solution focused on assigning credit to different touchpoints (i.e. impressions or clicks) along a digital customer journey. 

Data-driven attribution (DDA) or multi-touch attribution (MTA) models vary in their level of sophistication, but the core issue is that digital attribution shares out the relative impact of digital channels on a specific conversion, without accounting for the impact of any other factors that might influence sales, and without quantifying what would happen if spending in one or more channels stopped entirely. 

So if an attribution method does not measure incrementality, does that mean it isn’t useful? 

No, it doesn’t, as long as you understand the roles for the different measurement approaches in your toolkit. Incrementality is important for big strategic decisions – how much should I spend on media, what should the overall media mix look like – but digital attribution offers more granular results (down to the campaign and ad group level) that can help with more tactical and high-frequency decisions. 

There is no “one-size-fits-all" approach to marketing measurement and no single solution is perfect, so successful modern measurement frameworks tend to incorporate multiple complementary methodologies to evaluate performance together. 

If you’d like to discuss which marketing measurement approach might suit your organization, please reach out and talk with one of our expert consultants.