Here we have a very typical challenge for every Digital Agency, Head of Digital Marketing or performance team: Your CFO, CMO or any responsible of your investments without understanding of ‘Attribution’ sees a report of cost/lead or cost/sale and challenges the team ‘Why do we keep investing on Channel A or Channel B if they are not profitable?”
Well, this article will try to bring some help!
Overall this is a hard explanation for every digital media buyer as most of the times it is not possible to give all the credit to one single channel or activity along your marketing (or sales) funnel…. and was treated extensively on the article “Rethinking today’s attribution problem in digital marketing”
Going to the point, the only way to answer that question is by explaining the concept of “Attribution”, the science behind assigning values to individual digital touch points across your online customer journey since the first time interaction with your brand/product till complete finalize the purchase.
A proper attribution model is not only essential to optimize marketing campaigns but also to understand the real ROI of each channel and should be considered at each budget allocation review.
If Attribution is totally new to you, please read this article of TehcCrunch, it comes with very detailed examples of the concept and how to map the credit among different channels.
On this article though we mostly focus on understanding:
- How many attribution models we have.
- Which model is more suitable for our funnel and strategy.
How many attribution models are there?
There are different models that distribute the credit of a conversion in different ways. Below I drop a list of the most popular attribution models ones
These are the most popular models listed by the Square Online Digital Marketing courses
As a minor clarification of each of them:
- First click: Will provide the credit to the first click channels. In performance this will probably your Facebook posts, GDN banners, first-page of an organic visit or any E-mail marketing activities. Comparing these results with the ‘Last click’ model is very powerful as allows you to optimize your content and visuals.
- Last click: The most common one. This is an important but dangerous model, important because obviously allows you to track who score the conversion but dangerous as usually make people forget the importance of the first interaction. Bear in mind that most of the users that will convert in your site will interact with your brand several times before converting, in other words, Last click misses probably the most important part of the conversation.
- Last non-direct click: This comes with two requisites:
- Ignores direct traffic: Considers only Organic, Referral, Paid, Social and Others. Direct traffic is missed as it is usually traffic from current customers, employees and very common users that have probably been won through a different channel and is usually filtered to let you focus on the last marketing activity before conversion.
- Ignores the previous before the last to provide more credit to the referring channel.
- Linear model: Same credit to every single channel that takes part of the conversion.
Some more info about the default models can be found on this link of Google Support
Why is important to understand attribution models different to ‘First’ and ‘Last’ click interaction?
Using the last-click model means you are either ignoring early, top-of-funnel activity or instead focusing on bottom-of-funnel elements like branded search and remarketing (which tend to drive the final conversion). Without giving value to the top-of-funnel channels, sooner or later, your remarketing efforts will dry up.
Other attribution models, such as time-decay and metric-driven, offer more sophisticated modeling but still end up assigning arbitrary values and leaving gaps when customers switch between channels, especially when going from online to offline.
Further Articles to read:
- What’s the best attribution model For PPC? – Link MUST READ
- Attribution modeling overview – Link
- Guide to Google Analytics Attribution Models – Link
- Understanding View-Through Attribution – Link