A customer buys, receives the product, and cancels five days later with no reason given. If that event lands in the same counter as a cancellation caused by bad service, your retention dashboard is adding up two different things as if they were one.
What Disposición 954/2025 requires
Since September 4, 2025, the Subsecretaría de Defensa del Consumidor y Lealtad Comercial unified the cooling-off right and the service cancellation link into a single rule, replacing the earlier resolutions on the topic. Every provider selling online or through digital channels in Argentina has to show a visible link, the cancellation button, that lets a consumer cancel the purchase within 10 calendar days of receiving the product, with no reason required and no extra paperwork.
The rule carves out two cases: a product or service already used or consumed (Article 1116 of the Código Civil y Comercial) and purchases made for resale. Outside those two cases, the right is unconditional: a customer can return something in perfect condition for no reason other than changing their mind.
Why folding it into churn breaks the metric
Churn measures a decision to stop being a customer, usually tied to price, a bad experience, or a product that failed to deliver. We already covered how a one-time price increase produces a cancellation pattern distinct from baseline abandonment: that same distinction disappears once every cancellation lands in one bucket. A cooling-off cancellation needs none of those three causes: it can happen with a flawless product, a competitive price, after a smooth buying experience. If your churn report does not separate that event out, you will chase a cause that does not exist every time the number climbs.
The problem is not only definitional, it is also about where the data lives. The cancellation link is usually processed in logistics or payments, while churn gets calculated in the CRM or the lifecycle platform. If those systems do not share the cancellation reason, the information stays split across tools that do not talk to each other, the same gap a customer data platform exists to close, and rebuilding it by hand does not scale.
How to separate the two signals
- Tag the reason at the source: if the cancellation came through the cooling-off link, mark it in that same event, do not infer it later from the date.
- Pull cancellations inside the 10-day window with no other stated reason out of your churn calculation.
- Measure cooling-off cancellations as their own rate: a high number there says something too, but about how clear the offer was before purchase, not about the relationship with the customer afterward.
- Check the pattern by product category: a rate concentrated in a few products usually points to a description or photo that does not match what arrives, a pre-sale problem, not a retention one.
If your churn metric rises after a successful campaign, before looking for what went wrong in the relationship with those customers, check how many of those cancellations happened in the first 10 days. You may not have lost a customer. Someone may have used exactly the right the law gives them.