Attribution Last Click: Is the Tail Wagging the Dog?

Back to White Papers

Are Current Attribution Models leading Businesses Astray?

Attribution is the buzzword for 2016, just as Big Data was for 2015, and used properly it can help organisations optimise channel usage and spend whilst maximising either sales, new customers or other goals.

Attribution in its truest sense should consider all channels online (e.g. SEO email and PPC), offline channels including DM and even PR. The model can also incorporate external macro events such as competitor activity.

However, for many people attribution is all about the online environment, this is partly because most open source attribution tools solely consider the online journey with many measuring just ‘final click’. Certain organisations get pulled down the route of just measuring final click and there is a growing trend towards ignoring channels that can’t be measured in such a way. In other words, there are severe limitations of using Google Analytics (GA) type attribution modelling techniques.

The type of channels being dropped or reduced are often TV, Press and to a certain extent DM, whereas money is poured into SEO and PPC, with PPC being dominant as it is both measurable and cost effective. In a multi channel, multi touch-point world there are numerous avenues for consumers to interact with a brand. These channels should not be seen in isolation and measurement must include all traditional and digital channels as well as activity of competitor brands. 

Online Retailer Example

To illustrate this consider one of our clients, an online retailer who has adopted our new state of the art attribution model. It is very accurate in mapping customer journeys, whereas all most online models do is push PPC as the champion channel. Previously, based on the suggestions of one dimensional GA type attribution models, our client focused most of their spend on online channels, cut radio and direct mail completely and reduced TV spend. This was because sales generated from the latter three’s activity could not be attributed using the model. Initially there was little change in sales, but subsequently sales did fall and fall quite substantially in volume and interestingly in value. After using OmniattTM, which gives a fuller picture of the different ways customers interact with their brand, they had to change strategy.

Based on its findings, the decision was taken to reallocate spend on online channels, diverting resources in favour of radio and direct mail. As a result the company witnessed a turnaround in fortunes experiencing an increase in both sales volume and value.

Traditional Media: Not an afterthought

Wider industry trends would seem to suggest that companies ignore offline channels at their peril. Businesses from all sectors, in particular financial services, are reverting back to neglected yet affective traditional media. DM door drops in particular are enjoying a resurgence. Many firms assume the increasing sophistication of analytics and refinement of data quality lends power only to digital channels, but this is not the case. DM produces great results in conjunction with precise targeting and customer profiling. Capturing information online about consumer buying habits can be followed up with DM door drops in that area, dramatically increasing the likelihood of conversion.

Indeed to get the best results research indicates that integrating online and offline channels is key. DM with an email follow up can generate up to 50% uplift. The complementary nature of traditional and digital media channels reveals the necessity for an attribution model that accounts for both.

Companies should step back and take stock of that fact we are now in the most complex online and offline multi channel environment ever and they should not let the final click attribution tail wag the multi-channel marketing strategy dog.