We need to talk about Measurement
Online advertising has been around for 22 years and circ. 15 of these years we accepted the notion of a ‘last click’ methodology. Why? We knew it was flawed and skewed the data. 30 day post impression. Last click wins. Laughable now that you think about it.
More than a decade later this was still the methodology of choice. Thankfully now, the industry has grown up. It had to.
By attributing the digital data, we have the ability to look at all the triggers, signals and engagement data and to identify the combination of activities and touchpoints contributing to a sale. Assigning value and therefore credit to the different touchpoints.
The key advantage here is speed of analysis and individuality over more traditional econometric modelling. The analysis typically traces the actions of single individuals (unique users) rather than aggregated level data in traditional modelling which evaluates the offline sales impact of offline marketing investments. Here’s the challenge…
The obvious flaw is that online attribution modelling fails to account for the impact offline media has in driving people online. This means it cannot measure true sales effect if sales happen offline. Whilst some offline channels can be ‘plugged’ into the model, not all channels can be (door drops and DM examples here) the methodology therefore is unreliable.
The journey through a modern purchase funnel is complex, non linear and fraught with cross device tracking issues. Both marketing mix modelling (top down) and path analysis (bottom up) does need to be considered for a more realistic measurement framework though. The two are intimately interconnected.
The combined use of econometrics and attribution modelling can unearth valuable and actionable insight to inform the entire marketing strategy and ROI across channels.
Where do you draw the line between the two techniques?
With the right tracking; retention initiatives, CRM, and loyalty programmes can help link offline sales with online marketing. More traditional offline media is becoming addressable.
As Dr Peter Cain from Marketscience Consulting observes, ‘With the advent of multichannel marketing, comes the need to measure the sales impact of inherently micro-focused digital media alongside more macro-oriented traditional advertising. Consequently, any analytical approach that aims to incorporate both elements inevitably involves a degree of data aggregation or disaggregation depending on whether we adopt a macro or a micro route’.
An accurate model of measurement for the current ‘drive to digital’ initiative by many brands today is challenging. Without question, offline marketing does drive digital sales but, to what extent? Traffic increases can be observed, brand searches may rise, shop to buy conversion metrics might surge and quite possibly sales improve but because there is no consistent or systematic flow of directly attributable data, some assumptions have to be made. These are modelled.
Understanding customer behaviour is essential for business success so it’s critical to uncover all the steps within the purchase funnel, discover how they interact and what relationship they have with each other. Only then can we connect the online and offline worlds and truly understand performance.
