And now for something more complex…
Calculating your customer acquisition cost would be much easier if all users came from the same channel! Unfortunately, as we saw yesterday, that rarely happens and you need to have an attribution strategy in place to decide which channels are responsible for a customer.
Few companies use the “Last Touch” model, which means you are likely to spread credit out across a few different channels. That is great, we love spreading credit around! But how do you calculate your CAC when you have assigned credit to multiple channels? It’s easier than you might think.
Step 1: Determine how much each channel costs you.
For Ads this is easy; it’s just your average cost per click or impression. For PR it’s the time you spend courting the media, as well as the cost of your PR firm (amortized). For sending e-mails it’s more complex, as it includes the time you spend writing them as well as what it might have cost to get the customer’s e-mail address in the first place. In all cases, you should be able to determine an average cost for that channel action.
Step 2: Weighted distribution.
With your channel costs in hand, you can use the credit distribution from your model to calculate a weighted cost! Let’s say a customer touches three channels (Ads, E-mail, Social) and we use a linear attribution model to assign credit evenly, our CAC would be calculated as follows:
There you go! You can calculate this for each customer (hard) or segment your customers into common cohorts (easier) depending on how accurate you need your CAC estimate to be. In businesses where margins are measured in pennies, the CAC is usually measured at the individual customer level (e.g. mobile gaming) and in high margin businesses (e.g. enterprise software) by cohort.
As I mentioned yesterday, there are tools and services that will do all of this for you, but it’s up to you if you want to use them. Even if you do, it’s important to understand the fundamentals of what they do!
Quote of the Day: “A major indication that there are problems in a field is when there is no consensus as to correct methodology or technique.” – The State of Authorship Attribution Studies by Rudman, Holmes, Tweedie, and Baayen