I think it’s safe to say that last click attribution is out. Not only because the majority of industries have reached online maturity and thus conversions are driven by strong brands (read “The decline of generic search and how to save your brand”), but also because consumers have evolved and are now using numerous digital channels, which are in fact influencing their purchasing decisions.
Understanding how different digital channels are used through the decision journey has never been more important. Cross-channel consumer decision journey reports offer a simple way of gaining a better understanding of the role of each digital channel.
Although Google Analytics offers some predefined reports on multi-channel funnels and attribution, it’s hard to make any sense out of them. But actionable insights are just a few clicks away.
The whole point is to look at each channel separately and understand how users acquired by one channel are using other channels down the road.
Let’s take Display advertising as an example. Display advertising doesn’t commonly convert well (actually making just a few conversions and even these with really high CPA). But we don’t usually panic, as we know or presume anyway that Display is actually acquiring new users who are then converting over other channels. But how many of these newly acquired users are actually converting? What’s the real ratio between last-click conversions and first-click acquisitions? This is where our Cross-channel report comes in handy.
How to create a Cross-channel report
To create a Cross-channel report (we’ll continue to use Display as an example) you will need to look at the percentage of Assisted and Last interactions, which starts with Display as the First interaction for every channel.
To do that you should open the Assisted Conversion report in Google Analytics and under Conversion Segments create a new User-defined segment for “First interaction is Display”. Click Apply and make sure you have only E-commerce transactions selected under Conversion. Also adjust Interaction Type and Look-back Window settings to match your customer behaviour. The Look-back window should match the length of your buying cycle to make sure you are actually looking at the whole consumer decision journey. Set Interaction Type to “Clicks” to include only actual visits to your website. Include “Impressions” if staying top of mind has a crucial effect on your business.
Select the First Interaction Analysis to see which channels those users acquired by Display are actually using to convert.
In our case above we see that 38.91% of users acquired by Display are actually using Display again as the last channel before the conversion. In the majority of cases you will see that users acquired by one channel are seldom converting over the same channel. Additionally, we see that the First/Last Click ratio for Display is 2.57. To get the numbers for assisted conversion, open Assist Interaction Analysis and check the Assisted Conversions.
You can do the same for other channels. Get the numbers out and visualize your Cross-channel consumer decision journey to understand it better. A simple Stacked Column chart will do the trick.
Voilá! You can clearly see how users acquired by Display are slowly switching to other channels in the latter stages of the decision journey. 49.72 % of them are using Display in the middle of the funnel and 38.91 % are actually converting over Display (which is quite a high rate). We also notice that almost one-third of users acquired by Display are converting over Organic Search.
But don’t take the example above for absolute fact – your numbers could be completely different. Open your Google Analytics and start exploring.
Here’s the Cross-channel report for Display as First interaction for another client of ours. As you see, the numbers are quite different.
Before you start creating your Cross-channel reports make sure your traffic is attributed correctly to Default Channel Grouping. Common issues arise with Facebook advertising, which is attributed to Paid Search. Be consistent with your UTMs – if needed, create a Custom Channel Grouping.
So how can these reports help me?
You should use Cross-channel reports to optimize your cross-channel and intra-channel strategy.
Analyse at which stages of the consumer decision journey users are using specific channels and adapt targeting, copy and bidding to that. In our two cases above, the strategy for Display advertising would be totally different. In the first case, Display plays a significant role in the cross-decision journey, as almost 40% of users are using Display to complete the purchase. In the second case Display is highly important in the first steps of the decision journey, but not in the last step. This could either mean you are not using remarketing in the last step (or your remarketing is really bad) or that Display is really not suited to your users’ behaviour in this step. Instead, remarketing in the last step of the decision journey in our first case is really important, and you should spend more time optimizing it.
Channels used at the beginning of the consumer decision journey (First interaction) should help build brand awareness. Cross-channel reports show how effectively your brand building is for each channel. If you use Display to build up awareness and you don’t see any “Pull” channels (Direct, Organic Search, Paid Search) in the Assisted interactions section, you’re not doing your job very effectively. I would say that at least 50% of sessions in the Assisted interactions section and the vast majority of sessions in the Last interaction section should come from “Pull” channels. Otherwise, you’re just pushing your product or service down people’s throats, and this is not sustainable.
For me, the most important insights we gain from Cross-channel reports are related to user behaviour. These reports really help you to understand how users are using different channels, and are especially helpful in understanding where users came from (in terms of previously used channels). If I know that I acquired a user through a Display campaign and then successfully build interest and intent with him, which resulted in the use of Organic Search in the Assisted interactions section, I would treat that user differently in the Last interaction section (using separate remarketing lists and adapting the copy and bidding). Insights into user behaviour are priceless when it comes to building an advanced remarketing strategy.