Last week I outlined the current challenges of attribution in digital paid media. Before moving on to the recommended solution, let’s examine another way attribution plays out and how you can educate yourself on what’s really happening in your analytics. In the scenario outlined last week, we mentioned Display and Search, with a brand search eventually getting credit in your CRM system. That individual conversion in your CRM system is registered as a single conversion. However, in the eyes of your marketing channels, it’s two. How is this possible and does that mean we are duplicating conversions? It is possible, but also important that it is, and actually a good thing. Your CRM and Google Analytics, if you are using that, will always attempt to de-duplicate conversion activity so if you are using those platforms to quantify how your business is performing that’s fine, not ideal, but fine.
Notice I said “how your business is performing”- not your advertising. Those are two separate things and as marketers we’ve got to get comfortable with that. Here’s what I mean. It’s vital that your marketing, platforms and partners are getting a real-time feedback loop on what is converting. That means in our previous crib purchase example, the display impressions and their associated targeting tactics get credit for influencing a user’s purchase decision. We have to be aware of the difference between “influencing” a conversion and “driving” a conversion action.
Display models also default to the last touch impression (if there is view based credit allowed), so it will be the last interaction with a display ad that will get the credit, same as search. The problem is both of these platforms, display and search, think they had the last interaction with that user before they converted and so are both claiming credit for being the singular driving force that resulted in the conversion. This will result in pulling a report from both platforms that, when combined, will inflate your conversions beyond what they really were in your CRM. A problem? Not if you remember the core tenet of marketing which should be to continue to do the things that are driving the desired, monetize-able user action that results in your business being increasingly successful.
High level offline and online attribution
Importance of measuring the right targeting tactics against the right metrics
The widely adopted default last touch methodology
The challenges to analysis that last touch methodology represents
Having duplicate conversions register for the same user action and why that’s not a bad thing
Whew. So how do we wrap this up? How about with my preferred solution for solving attribution for marketers? Multi-touch attribution.
Multi-touch attribution is a third-party platform solution that sits above all of the digital marketing channels, which is important as it is completely agnostic to all the channels running on the plan. While it is technically complicated to explain, the important thing is that it sees and measures everything.
By tagging your marketing at the most granular level, keywords and ads, with a unified tracking code that can be used in both a click and view-based capacity, no impression’s impact is missed. This gives you a chronological order of every single interaction your digital marketing had with a converting user. Once the model has “learned” it starts developing a fractional weighting system to give credit to EVERY single ad interaction, not just last touch credit. And it keeps learning as it gets more data, it’s constantly tweaking the weighting to be as accurate as possible.
With a fractional approach, many contributing elements of our marketing plan finally start showing their contribution in a meaningful and quantifiable way that allows you to smartly optimize around those elements that are positively contributing to your KPIs (Key Performance Indicators). The cherry on top is that the fractional credit will always add up to one per each converting user’s path.
We’ve evolved beyond duplicate lead credit and can align to the way advertisers see the performance within their CRM. So exports of data out of this system will be much closer to what brands are seeing on their side as well, giving everyone a good feeling that both marketing and the brand results are headed in the right direction.