Marketing Entrepreneur. Day Trader. Business Contrarian.

5 Reasons You Should be Using an Attribution Model

Posted on Jan 22, 2016 in Analytics, Marketing Estimated reading time: 5 minutes


When I talk to other marketers about the concept of measuring channel performance with full-funnel visibility, the initial reaction I see most often is “Yes, why haven’t I been doing this sooner?” It makes sense, because we’ve been talking about leveraging attribution as a more effective way of measuring the performance across marketing channels for years now. Still, it remains a topic people want to be doing, but aren’t quite sure how.

I’m not talking about  a traditional “last touch” model, either. I’m referring to a number of more sophisticated options ranging from linear, weighted, regression, etc. Fortunately, I’ve compiled the five tips below to hopefully push you over the edge and get you thinking more seriously about how you can adopt and implement this concept in your own company.


1. Attribution helps set more realistic expectations on individual channel performance

In a traditional mindset, we tend to measure the performance of all of our marketing channels the same way, even though they were designed to complete different objectives. At the end of the day, you should be holding all of your marketing programs accountable for increasing customer acquisition, but the way that we hold marketing programs accountable for that end goal need to be different. For example, if we simply measured a lookalike modeling program through a programmatic display channel the same way we measured a paid search channel, the the results would likely look bad. The job of a lookalike model is to introduce new prospects to the top of the funnel, not convert people based on last touch. If they do, it’s just icing on the cake. The use of an attribution platform allows you to credit each of your programs for doing the actual job they were intended to do.


2. Your larger competitors are most like using an attribution solution

Nobody likes to be falling behind their competition. If your competition is doing something that you aren’t, it helps make a stronger case for you to be doing it too. In the case of attribution modeling, any competitor that’s adopting this technology can use it to help them acquire new customers and lower their customer acquisition costs. If you’re in a niche industry where the market size is a zero-sum game, it’s important to grab each incremental percentage of the market you can. When you’re able to identify the common trends in the way prospects are interacting with your messages and content online, you’ll gain a huge leg up if you can turn it into a scalable, repeatable process.


3. Attribution shows you more depth in the conversion path than what marketing automation can provide

Most sophisticated marketers rely on marketing automation behavior/activity reporting to show the touch points a prospect goes through leading up to the point of purchase. However, one shortcoming of these solutions is that the touch points can only be measured once the prospect is already a known record in your database. Prior to being known, they’re simply an anonymous IP address hitting your site. Attribution platforms go the extra mile by identifying the touch points and behaviors these same prospects go through prior to being known. In large companies, think of all the paid media you’re running. Wouldn’t it be nice to know who’s seeing your messages across multiple channels? I’d sure like to know!


4. An attribution model helps you measure engagement and acquisition more accurately

Your management teams cares about where they can invest more money to get more customers. I hate to be the bearer of the “bottom line is all that matters” bad news, but that’s actually the reality from their perspective. Communicate that attribution modeling platforms serve this goal directly, down to being able to provide spend recommendations at a per advertisement or creative level. Aside from having the insights into knowing where you can invest more money, current attribution modeling platforms also show you where you’re spending that isn’t leading to a conversion. It’s best to use an algorithmic model for these types of insights over a weighted model, because they take future forecasts into account.


5. A proper attribution model helps you validate the audience you’re targeting

Most targeting options through online marketing channels rely on the use of behavioral data, whether it’s collected via first-party on your website or via third party through a vendor (build and learn how to create your own website). Attribution platforms help you understand additional behavioral characteristics about your audience. How many days does it take a prospect to convert? How many touch points are they exposed to between the time they first hear about you and convert? These questions and more can all be answered. To a further extent, the answers to these questions become actionable. For example, if the median number of days to conversion is 15, you can take steps to reduce the path length by optimizing the messages they were exposed to and which content they saw when they searched for you. Through these optimizations, you can see how your rate of spend and CPA changes.



The five tips above should be a great start to getting your executive team on-board with how you want to measure marketing channel performance. This is a big area that we’re focusing on internally as well. Ideally, we’d like to see marketing attribution applications as ubiquitous as a CRM or marketing automation application, and we’re very quickly moving in that direction.