Without trying to sound condescending, marketing attribution is simply the process of measuring the true effectiveness of your marketing. Unfortunately, it’s one of those things which isn’t as simple as it sounds as there’s more than one (core) model you can (and should) apply. Why? To truly understand the effectiveness of your marketing, it’s important to see what’s working and what’s not – along with the role of each channel.
An example; someone comes to your website via paid search but doesn’t convert. They come back later via organic search and sign up to your (email) newsletters – before eventually converting from one. A post click (or last touch) model would say it’s down to the newsletter. What about the first two visits – which got them to the website to sign up to your newsletters. Using a first click (or first touch) model, paid search would be responsible for the sale (as it introduced them to the website – but it didn’t convert them).
Many (if not, most) use post click as it’s easier, but what you’ll find with post click, is that there are one or two channels that tend to ‘clean up’ (usually paid search being one of them) – whereas other channels don’t do so well.
In the example above, if you simply use post click (where email is the ‘winner’) and stop doing paid search, then you might stop getting (new) sales.
To further complicate things, in the example below, these are all clicks – what about if someone saw (e.g. impression as opposed to click) one of your display ads first (before searching and click on a paid search ad) – and then you stop doing display advertising as ‘nobody ever clicks on banners’.
Google Analytics, for example, easily lets you compare models – of course, user journeys are based on cookies (unless logged in and setup correctly) so generally only account for journeys that have taken based on the same device, e.g. using the same cookies:
Ideally (as time consuming as it is), you’d use a combination of the two – along with ‘any touch’ attribution, taking into account any interaction(s), e.g. impressions and clicks, in order to see the true effectiveness of your marketing campaigns. That way you can see how channels contribute to user journeys – or if they don’t. Just remember when using any touch to ensure one sale is only counted once, e.g. it’s not ‘awarded’ to two channels.
In order to ensure you’re not double counting and you’re correctly measuring the effectiveness of your marketing spend, it’s imperative to ensure your data is as accurate is possible. Try to use one core system, e.g. Google Analytics and ensure your data is ‘tagged’ correctly – although a lot of systems support integrations. If you’d like to talk through marketing attribution, simply get in touch.