This week we kicked off a multi-part series focused on best practices for driving maximum value out of your customer acquisition strategy. The first point of order was to differentiate between the cost of acquiring an email address or lead versus acquiring a paying customer; moving forward we’ll call these metrics CPAR (cost per acquired registrant) and CPAC (cost per acquired customer).

With that distinction clear, next let’s answer an important question underpinning those metrics: how can you possibly assess which of your acquisition campaigns will be most effective today based on “true” acquisition cost? Let’s be realistic here: it often takes new users months, if not years, to buy for the first time.

The simple answer is forecasting. More specifically, you can rely on customer activation curves to help make quick projections today regarding what customer behavior might look like years from now.

An activation curve like this one explains how quickly a buyer purchases for the first time:

As companies expand their marketing mixes, they should build distinct activation curves for each channel (for instance, paid search cohorts will perform drastically differently than display, etc.).

In this example, we can see:

• The total universe of anyone who has ever purchased.

• Of all buyers who have ever purchased, ~52% of them purchased within 24 hours of signing up.

We can thereby deduce that on any given day, the number of day 1 buyers that come in is representative of only 52% of the total people who will ultimately go on to buy. Indeed, activation curves are powerful because they allow you to forecast CPAC fairly immediately.

Now, consider the math below:

• You spent $1,000 on a Google campaign yesterday, and it resulted in 100 signups (meaning a $10 CPAR).

• You don’t know precisely how many of those 100 people will ultimately go on to purchase something, but you do know how many of them purchased something yesterday (day 1) – let’s say it was 6 people.

• From your historical activation curve (above), you know that 6 buyers on day 1 represent only 52% of the total purchasers to come.

Stay with me:

• You can then extrapolate that 11 (6 ÷ 52% = 11) of the 100 will become buyers.

• That works out to a forecasted CPAC of $90.91 – very different from that $10 CPAR!

If you are a Sailthru client, you can also leverage our Query Builder to build your activation curves as well as to assess downstream behavior and revenue insights for any given cohort; feel free to reach out to your account manager if you have any questions around how to do this.

So, the next question on your mind should be, is a $91 CPAC any good? Those optimization decisions are a function of your customer lifetime value (CLV) metric. In this example, if CLV is more than $90.91, the unit economics for this campaign make sense. If not, we’d suggest reconsidering this particular ad spend.

In our next post, we’ll explore some effective tactics for converting more browsers into buyers as well as for accelerating time to first purchase (or to whatever your relevant engagement threshold might be) – stay tuned!

Important: this forecasting methodology is an estimate. Accordingly, it’s important for clients to revisit older acquisition campaigns every few months to analyze whether the cohort behavior played out differently than anticipated. If it did (in a positive fashion), the marketer can always turn that campaign back on – but we prefer turning something back on – even if some volume is lost – to spending hefty sums to only later find out that the cohort is severely unprofitable!