Attribution for Subscriptions: First Orders Are Not Reorders
Subscription businesses break standard attribution by design: the first order is marketing, the tenth is habit. How to measure acquisition without crediting ads for renewals they never caused.
A subscription business has two economies sharing one dashboard. The first order is bought with advertising, offers, and persuasion. The reorder is bought months earlier, by a product that did its job. Attribution that cannot tell them apart will credit this months ads for revenue that last years product quality created.
This guide covers a cleaner way to read subscription attribution: separating acquisition from retention economics, and judging channels on the only order they actually influenced.
Split the dashboard at the first order
Every subscription report should answer two questions separately: what did it cost to win this subscriber, and what is a subscriber worth over time. Blending the two produces a return on ad spend figure that flatters acquisition channels with retention revenue they did not earn.
The practical split is simple: first orders attributed to marketing with the usual windows and methods, reorders attributed to the subscription itself. Channels are then judged on cost per new subscriber, never on total subscriber revenue.
Reorders still carry information, just not credit
Reorder behaviour is where cohort quality shows. A channel whose subscribers cancel after the discount box and a channel whose subscribers stay for a year are not equal sources, even at identical first order cost. Lifetime value by acquisition cohort is the correction.
Read reorder patterns as a quality signal on acquisition, fed back into channel budgets quarterly. That is reorders informing attribution without contaminating it: they grade the channels, they do not pad the channels numbers.
Discount led acquisition distorts both sides
Introductory offers pull forward demand and select for price sensitive subscribers. Their first order economics look efficient and their retention is often the worst in the cohort, a combination blended dashboards report as average and therefore miss entirely.
Judge offer led channels on retained subscribers at ninety days, not on first orders at seven. The channel that wins on the honest measure is frequently not the one the blended report has been funding.
Report the two economies to different owners
Acquisition attribution belongs to the marketing review: channels, costs, cohort quality. Retention belongs to product and lifecycle: reorder rates, pause and skip behaviour, cancellation reasons. One meeting can cover both, but the numbers should never merge into a single blended return figure.
The discipline pays for itself the first time a budget argument uses it. When everyone can see which revenue ads bought and which revenue the product earned, the argument is about evidence instead of whose dashboard is bigger.
There is a planning dividend too. Acquisition targets set on first orders alone can be forecast honestly, because they no longer borrow confidence from reorder revenue that arrives regardless. Forecasts built on the split survive contact with the quarter. Blended ones rarely do.
Related reading
If this topic is useful, these earlier guides go deeper on connected questions: Reading Attribution During Short Flash Sales Without Fooling Yourself, Attribution When Shoppers Switch Devices Before Buying, and Planning Incrementality Tests You Can Actually Run After Peak.
Frequently asked questions
Why should first orders and reorders be attributed differently?
Because different things caused them. Marketing causes first orders. Product experience causes reorders. Crediting ads for reorders overstates channel performance and hides the retention problems that actually decide a subscription business.
How should subscription channels be judged?
On cost per new subscriber, corrected by cohort quality: how long subscribers from that channel stay and what they spend over time. A cheap channel with fast churn is expensive on the honest measure.
Do reorders have any role in attribution reporting?
Yes, as a quality grade on acquisition sources, reviewed through cohort lifetime value. They should inform budgets without ever being credited to the ads running when the reorder happened to land.
What is the most common subscription attribution mistake?
Reporting blended return on ad spend that mixes first order and reorder revenue. It flatters acquisition, disguises churn, and funds channels for revenue the product created.