Confused when people talk about attribution modeling? Let me simplify it -- using SPORTS!

"Attribution? Yeah, yeah, yeah, I toootally know what that means."

LIAR!

It's cool though, not everyone loves to nerd out on this topic, so let me remove the confusion and break it down for you.

Attribution doesn't have to be complicated, but it can definitely be made more complicated (than it has to be). Even though I love to nerd out on math and modeling, I also love to keep things simple. Attribution is essentially getting a better understanding of what digital sources are contributing to your sales and crediting each of them with an accurate percentage of the sale.

If you want to get all technical about things, Wikipedia defines attribution as:

In marketing, attribution, also known as multi-touch attribution, is the identification of a set of user actions ("events" or "touchpoints") that contribute in some manner to a desired outcome, and then the assignment of a value to each of these events.

1 Player Can't Do It All & Neither does 1 Source

Did your eyes glaze over?

Your head hit the desk?

Breathe, it's all gonna be OK, let me break this down and say that in a simpler way that we can ALL understand --- by USING SPORTS, BABY!! So, in sports term (before I go further, if you also don't like sports, seriously, I really don't want to know you!), let's say that your football team won a big game last night. Well, upon looking at the stats, you find that your quarterback had an AMAZING game! As usual, ESPN and all the casual, lazy fans (and owners) credit HIM with the victory, but to the smarter analyst, you and I, we both know that there is a deeper story.

Different Ways to Give Credit

So what do you do first, you review all of the sources that lead to the win (the sale, the email, the ebook download, etc). In football, the quarterback had a great game, but maybe his receivers had lots of yards after catch (YAC), the running back made a key block that let the QB launch that bomb that broke things open, or the competition had major injuries. In this scenario, would you give the QB 100% of the credit for the win?? Unless you're just trying to irritate me, and every other sports fan, OF COURSE NOT!! It's the same with attribution models. For this reason, I am NOT a fan of the first, or last, touch models which give 100% of the credit for the win to your customers first or last interaction.

Personally, I prefer the last approach best, the data driven approach, because every company is different, but the position-based and linear approach I find to be best for general application in your Google Analytics multi-channel funnel (MCF) attribution reports. The time decay approach is also good, but I wish Google had another model that was the reverse and credited the majority of the sale to the FIRST touch, instead of the last, because if you're a company that is investing a lot into brand awareness, I personally credit the initial contact as the primary reason that they eventually lead to a conversion.

Sooo, annnnywaaaays, I hope that you enjoyed this introduction to attribution, and my sports analogy (even if you didn't, say you did!). This is not only a very interesting topic, but one that is very important to you figuring out which of your marketing channels are helping you reach your goals and where you should keep investing (or stop investing) your time, money and efforts.

Peace.

~LCThr33