The consumer journey is not clear cut. Changes in how consumers search presents several challenges in understanding attribution. Not only are people using multiple devices (mobile, desktop, and tablet) devices to search for something, they spend more time online before decision time which includes multiple visits to the same website. People rarely see an ad online and immediately make a purchase as a result.
For most consumers, they first learn about a brand through a paid campaign or other marketing effors, they compare it to other similar brands and do all this across multiple channels before deciding to purchase.
However, even though we know the path is not straightforward, some marketers still view their data with a direct cause-effect mindset expecting a click on a paid ad to immediately result in a sale. This approach is a last click attribution model where all the credit for a conversion is given to the channel a visitor clicked immediately before converting. When advertisers look only at that last click channel, it undervalues the other actions taken before converting.
For example, if advertisers look at paid traffic and immediate conversions, it may appear the paid traffic did not influence the decision to purchase. However, people may be very interested but decide to visit other channels and sites between the time they clicked on a paid ad and the time they converted on a company’s website. A last-click model does not account for the multiple interactions that happen in the funnel.
When marketers consider the full funnel and explore attribution models other than the last click model, they are looking at all the touchpoints that drive conversions on a website. This is why marketers want to consider a multi-touch attribution model where they can view the full consumer journey to account for all actions before the conversion. They no longer wonder which marketing touchpoint drove the sale or conversion.
Think about your own online activity before making a purchase. Have you booked a vacation on the first visit to a site and did it on that single device? Probably not. Just a few years ago, the average shopper was visiting a retailer site nine times before deciding to buy. If credit was given only to the first visit or last visit, that is lost information about behaviour during visits two through eight.
So how do you determine the best attribution model? First, view the different options in AdWords and ensure you understand the difference between each of them.
First-Click and Last-Click: We do not recommend using only these as your primary tool for understanding attribution. They give 100% of the credit to the First-Click or Last-Click and looks for a direct cause-effect in the decision to buy.
Position-Based: This views the first and last clicks as key touches with 40% of the credit going to each and distributes the remaining 20% evenly along the middle. With position-based modeling, this can undervalue the middle touches since the 20% is spread out. However, since this gives more credit to the first (and last) touch points, it can be helpful to better see the impact of Display Network behaviour. These channels would be overlooked in a last click model since there is a role for education and branding that will not lead to direct conversions.
Time-Decay: It allocates more credit to touch points closer to the time of the conversion but may overvalue those touchpoints.
Linear: Every channel is given equal credit. With a moderate growth strategy, the linear attribution model is a good option since it distributes credit equally for all clicks that led to the final conversion.
Data-driven: Google describes this model as giving “credit for conversions based on how people search for your business and decide to become your customers. It uses data from your account to determine which ads, keywords, and campaigns have the greatest impact on your business goals. You can use data-driven attribution for website and Google Analytics conversions from Search Network campaigns.”
Data-driven attribution does take out the guesswork and since it is dynamic, it is always learning by leveraging your account data and utilizing machine learning to see which touchpoints have the biggest influence. Credit is given based on the contribution of each keyword across the consumer journey. However, not all advertisers will have this option in their AdWords account. To use a data-driven attribution model, an account must have at least 15,000 clicks and a conversion action must have at least 600 conversions within 30 days.
In addition to reviewing these Attribution Models in Google AdWords, you also want to view this data in Google Analytics to see how your paid traffic performed compared to other channels. To see this data, go to Assisted Conversions in Google Analytics under Conversions > Multi-Channel Funnels > Assisted Conversions. This provides insight on how these channels work together.
Also this section of Google Analytics, you can also drill into the time lag and path length reports specifically for your paid traffic. Time lag report shows how long it takes from the time a visitor clicks on your ad to when they convert. With low dollar items, such as a book, the purchase could happen on the first day depending on the investment requested. However, with bigger purchases, such as planning a vacation online, you will likely see a bigger date range, perhaps over several weeks.
And when there are a large number of days (time lag) before a significant purchase is made, there are also more likely to be a number of touchpoints happening from that point of introduction to when the purchase is made (path length). In other words, you will likely see more channels in your path. If this data is new to you, this short clip will walk you through view funnels in Google Analytics.
As you explore attribution models through the entire funnel, you want to rethink the default attribution model you use for analysis, especially if you have been using the default of Last-Click Attribution. With a multi-touch model, you are including earlier clicks in the journey prior to conversion and giving them credit for the decision to purchase. This provides a complete view of the consumer journey.
You also want to revisit your timeline for analysis based on the time lag in your Google Analytics account. To give credit to earlier clicks, you will want to analyse data over a minimum of 30 days to account for these earlier visits. A smaller conversion window may miss conversions for people who convert at a later date.
Tracking your touchpoints from start to finish allows you to see the value of your efforts and make decisions that are data-driven. It helps you understand which channels are the most valuable in bringing you new consumers and how they work with other channels in the funnel. Remember that attribution data, like any other metrics you use in marketing, is only powerful if you act on it. Understanding these touchpoints in a consumer journey is valuable only if you will implement your findings in future campaigns.