Ian Harris is the founder and CEO of Search Laboratory, an award-winning digital marketing agency with 150 staff based in Leeds, London and New York. He tells us how Marketing Attribution has come of Age.
The old adage of “only 50% of my marketing spend works, I just don’t know which 50%’ has been defunct for some time. Tracking of digital marketing activity has been part of the landscape since the start of digital and each channel, be it Facebook, Google AdWords, or online display advertising, have built-in tracking to determine who has seen the ad and then gone on to convert into a paying customer.
Marketers then faced a different problem. Each of these channels claims credit for a conversion, but in most cases, the customer touched multiple channels before they converted. They may see a Facebook ad, then later perform a Google search and click on a PPC ad, then later search for the company directly and click on an organic search link. Each channel may claim that conversion. Alternatively, we can apply some kind of logic, such as ‘last click gets it’ to make sure we are not double counting.
However, this puts us back to the bad old days of not really knowing which parts of that user-journey had what level of influence on the final purchase decision. Would the person have ordered anyway if they hadn’t seen the Facebook ad?
This conundrum is called attribution modelling and it is the science of assigning credit to the marketing touchpoints that led to your customer making a purchase.
It has baffled digital marketers for years, and all sorts of models have been developed to try to give credit to each touchpoint such as ‘linear’, where each touchpoint gets an equal share of the credit, to ‘U shaped’ models that overweight the first and last touchpoints, and many more besides. All of these models are flawed because they are essentially guesswork.
The reason attribution modelling is so critical though is that, if you get it right and you know the exact effect of every touchpoint, you can spend your marketing pounds optimally. You will minimise waste and beat your competitors.
No More Excuses
The great news is that the world of attribution modelling has come of age so there is no reason to continue applying these arbitrary models, or indeed operating on a ‘last click gets it’ basis. This change is largely due to two significant advances in technology:
1. Analytics tracking has got much more advanced. It is now possible to track all touchpoints within a single analytics platform. It is possible to track when a user views an online ad but doesn’t click through to the website. It is even possible to track across different devices, so if a user views an ad on a phone, performs a search on a tablet, then completes the purchase on a laptop, these touchpoints can be joined together to form a single journey.
2. Advanced data driven attribution modelling systems are becoming accessible to everyone. Data driven attribution is, in short, complex statistical modelling that analyses all user journeys and then calculates the contribution of each touchpoint. It does this by analysing thousands of journeys with and without certain marketing touchpoints and comparing them statistically to determine the probability of a conversion with and without that touchpoint. It is complex maths, but it is now built in to Google Analytics 360 (the paid version of Google Analytics), Google AdWords and other media purchasing platforms. Google are also releasing purpose-built platforms to do this modelling called Google Attribution (a free tool) and Google Attribution 360 (the paid version, with more features and forecasting capabilities).
Data driven attribution requires a lot of data, so if you have a small website with low visitor numbers and only a few conversions it is probably not right for you. If you do have the volume though, these two technological advances mean that you must make sure your marketing teams are implementing data driven attribution or laying plans to move that way very shortly.
There is no longer any excuse for guesswork in digital marketing. Data driven attribution will enable you to spend optimally and give you the best chance of beating your competitors.