The Ecomm Analyst

Yes, but with a caveat that matters more than the answer. Attribution tools handle subscription brands fine at the point of acquisition and handle them badly after that, because almost all of them are built around the first order and subscription economics live in orders two through twenty.

If you sell a one time product, the day you acquire a customer is the day you learn what they were worth. If you sell a subscription, the acquisition day tells you almost nothing, and a tool that reports on it as though it were the whole story will quietly push you toward the wrong channels.

What breaks when revenue recurs?

The core problem is that a rebill has no marketing source. Nobody clicked an ad. The card charged on schedule. So the tool has to decide what to do with that revenue, and there are three common behaviors.

Some tools count the rebill as direct or unattributed traffic, which buries real returns from your acquisition channels in a bucket labelled nothing. Some tools exclude rebills entirely, which makes every channel look worse than it is. Some tools credit the rebill back to the original acquisition source, which is the most useful behavior and the least common.

You need to find out which of the three your tool does before you read a single report. Ask the vendor directly. The answer is rarely on the pricing page and it changes what every number means.

Should rebills be credited to the acquisition source?

For deciding where to spend, yes. If Meta brought you a customer who has now paid four times, the honest view of that acquisition is four orders of revenue against one acquisition cost, and any view that shows you one order is understating the channel by 75 percent.

For deciding whether your business is healthy, no, because that framing hides churn. A channel that acquires customers who cancel after two cycles looks identical to one that acquires customers who stay for two years, right up until it does not.

This is why subscription brands need cohort views alongside channel views rather than instead of them. The attribution report tells you who came in. The cohort curve tells you whether they stayed. I covered how to read those curves without deceiving yourself in reading LTV and cohort curves without fooling yourself.

How do you calculate ROAS on a subscription brand?

Pick a window and hold it. The most workable approach I have seen is to measure acquisition cost against revenue from that cohort at a fixed horizon, usually 90 days, and to run that consistently rather than switching horizons when the number is unflattering.

Day one ROAS on a subscription brand is close to meaningless and will make every channel look unprofitable. Lifetime ROAS is the number people quote in pitch decks and is unusable for weekly decisions because you will not know it for two years. Ninety day contribution is a compromise that is wrong in a stable, comparable way, which is the most you can ask.

The practical requirement is a tool that lets you look at revenue by acquisition cohort over time rather than only by the calendar month the order landed in. Store side attribution tools that report against your actual order history, including ThoughtMetric, which sponsors this blog, are generally better positioned for this than platform reported conversion data, which has no view of order two.

What should you check before buying?

Three questions, asked before the trial ends. How does the tool classify a rebill order. Whether it can report revenue by acquisition cohort rather than only by order date. Whether it reads your subscription app’s data or only Shopify’s order records, which matters if you use Recharge, Skio, or similar, since the subscription state often lives there rather than in the order.

If a vendor cannot answer the first question clearly and immediately, that tells you something on its own about how much subscription work they have actually done.

Frequently asked questions

Do subscription rebills show up as direct traffic?

In many tools, yes, because there is no session or referrer attached to a scheduled charge. If your direct channel looks suspiciously large and you sell subscriptions, this is usually the explanation rather than genuine brand demand.

Can attribution tools track churn?

Mostly not, and you should not expect them to. Churn lives in your subscription platform. What a good attribution setup gives you is the ability to segment retention by acquisition source, which is the connection that actually changes spending decisions.

Is first touch or last touch better for subscriptions?

First touch tends to be more useful, because subscription purchases often involve a longer consideration period and last touch will over credit branded search and email that were closing a decision made weeks earlier. Consistency still matters more than the choice.