Home / Google Analytics  / Multi Touch Attribution models and how it will change the way you market online.

Up until now, a conversion is a conversion is a conversion. Regardless of whether a conversion is a phone call, a lead, a download, an email submit or a product purchase, it’s an event that we (should) track.
Why do we track? Well in its simplest form, it’s a measure of performance. As marketers, we need to know what channels and traffic sources lead to conversions then measure these against how much we spent to buy that traffic (ROI). Understanding this, we then get to work to find out what channels we should continue to spend money on and what channels we should leave out of our next media plan.
This is where the problem lies. The question most marketers don’t ask is how to much weight should be attributed to each channel. The majority of tracking software as a default (actually all of them!) use ‘last touch’ attribution to tell us where a conversion came from. This can be misleading especially when using multiple channels to drive traffic for a campaign.
Below is how last touch attribution works:
last touch attribution
This pretty much means that if a user came in to your site 5 times (from varying channels) and their final interaction was via search (quite common actually), then that conversion will simply say it came from Search. However do you place any weight for the contribution all the other channels such as made? Why? Why not?
On the flip side there is ‘First Touch’ attribution modelling:
 First Touch Attribution Model
First touch looks at it the other way and gives weighting completely on the first source that traffic came in on, ignoring all the other channels that eventually lead to the conversion.
You can always argue for a linear attribution model which gives every channel equal weighting:
linear attribution
Which one of these is right? Different schools of thought exist supporting each different attribution model. The truth is there is no one right answer, however it’s important to understand how your channels fit in using different types of attribution models. This is super important when you have ‘branding’ based traffic such as display running concurrently with Search and other transactional mediums.
If your data is telling your that Search is the best ‘last touch’ based converter, then dismissing all other channels can be detrimental. Those other channels could effectively we building up the brand awareness and bringing the user back indirectly leading them to convert on your website.
It’s not just digital that can be measured this way. Radio, TV, Print are all channels which can be measured if executed with attribution and tracking in mind. Vanity URLs and redirects can all pulled into multi-channel attribution models and measured within a free tool such as Google Analytics.
Attribution Model Example
The above is an example of us comparing First touch, last touch and linear conversion models. You can see the numbers noticeably alter as we change modelling and conversion weighting.
Feel free to get in touch for a current analysis of your traffic or if you want to learn more about how multi-channel attribution modelling works. You’ll be surprised at how quickly you can start tracking.
No Comments
Post a Comment