“First click attribution is akin to giving my first girlfriend 100% of the credit for me marrying my wife.” – Avinash Kaushik
Digital marketing has always been a core marketing function at Freshworks. This means we had numerous digital channels contributing to website traffic.
When we began, there was only one channel we invested in – Google Adwords. In mid-2012, we started expanding our scope to more digital channels such as social media platforms.
As good marketers, we all know through painful experiences that the Google Analytics numbers never match with other dashboards. The digital channels’ interface did not consider other touchpoints with the website while Google Analytics worked with the last click attribution. This was why they were a mismatch.
To solve this data discrepancy, we had to consider first click/interaction of the user with the website. The marketing automation tool we were using, at that point of time, gave us the first click/interaction channel data.
Did this solve the problem?
No! We still observed data mismatch. That’s when we realized that on an average, a user has at least three touchpoints with the website before signing up for the product.
The timing couldn’t have been better as we found Avinash Kaushik’s post on multi-channel attribution and how companies should be utilizing it.
This post opened our eyes just as we were trying to solve the problem of first click/interaction and last click/interaction. We realized that we had completely neglected the interactions in between.
By this time, a year and half had passed, and that’s when we decided to take the multi-touch attribution route to get closer to solving marketing attribution.
Before answering the question of what multi-touch attribution in marketing is, it is imperative to understand what marketing attribution is and why it is necessary for every business on the digital spectrum.
Gone are the days when marketing teams used to work towards a big billboard campaign, a prime time TV ad or sponsor an event, hoping to reach the right audience. Gone are the days of attributing the entire marketing spend to company’s revenue growth. The digital world provides the right tools to identify the real marketing drivers towards sales and revenue, the marketing channels that actually contribute to the sales funnel from the channels that don’t.
Marketing attribution is weighing the marketing channel(s) that contribute to the sales funnel, and thus, revenue.
So what is Multi-touch Attribution?
Multi-touch or multi-channel attribution is a process of determining the role of marketing channels from start to end in a user buying cycle.
A good metric to start with would be with the last channel that helped a conversion. As you slowly evolve, better understanding of the channels that have an impact on customer conversion will help you optimize your campaigns better.
For most SaaS products, it takes the user at least three attempts before they decide to sign up for the free trial. If we go with last channel attribution we completely eliminate the first two channels that contributed to the sign up.
This is an example of a the basic multi-touch attribution report, available on Google Analytics.
From this example, we can see the different paths users take before converting. Whereas, the traditional Google Analytics Source/Medium report would have given all the credits to referrals.
Here comes Attribution -> Model Comparison Tool report to the rescue!
GA offers you the choice to select between various attribution models under its belt so that you don’t have to worry about attributing the right points to each channel.
You can choose the apt model for your business from the drop down and GA would throw in the right report. Every time we decide the lifetime of a channel, we ensure that the two models are compared thoroughly to decide on the performance of the channel.
Here is a quick visual example:
Comparing last interaction vs position based (40% first click, 40% last click, and 20% to all middle interactions) gives a ~15% increase in conversions for paid search when compared to the traditional last click model.
At Freshworks, we use this tool often, and well. Each year as we keep growing, various complexities allowed us to build an attribution tool with our very own machine learning centric model.
Since we solved the first hurdle in the area of attribution, it was relatively easy for us to translate our learnings to build a new technology and devise our own multi-touch attribution model. For all you digital marketing and artificial intelligence (AI) enthusiasts, here is our detailed research paper on multi-touch attribution.
It is imperative that we attribute credit to digital marketing channels (multi-touch) that have been a part of the process pipeline for conversions rather than going with first click and last click conversions. Marketers should frequently consider looking at the “Top Conversion Path” option in Google Analytics to get more productive results.
Looking back in time, it’s great to see how we as a team have evolved from using last click attribution and then first click and now have accepted the multichannel path. However, automating the attribution model and following the path of the Data Driven Attribution model would take us one step closer to solving this problem.
Google has announced a free tool Attribution 360 to help bridge the gap between Google Adwords and Google Analytics reporting. In the future, this will not only help us extract the best value from our marketing channels but also let us monitor the performance of individual channels.