Nobody Stays on the Plan They Signed Up For

Operators tend to think about upgrades and dread downgrades, which is the wrong way round. Somebody moving from your top tier to your cheapest one is still a subscriber, and the alternative was almost never that they stayed where they were. It was that they left. A platform which makes downgrading awkward does not protect revenue, it converts a smaller payment into no payment at all, usually in January when everybody reviews what they are paying for. Making both directions easy is the cheapest retention work available, and it costs you nothing except the discipline to stop treating a downgrade as a failure.

When somebody does leave, the fair behaviour and the commercially sensible one happen to be identical. A cancellation stops the next renewal rather than cutting access off the moment it is clicked, so a subscriber who paid on the third and cancels on the tenth keeps what they already bought. Cutting them off immediately saves you nothing, since the money is already taken, and it guarantees the last thing they remember about your service is being shut out of something they paid for. It also makes them significantly harder to win back later, which matters because the people most likely to subscribe to you are the ones who already did. Cancellations keep their own view rather than being deleted, which is what makes a win back campaign possible at all, and that is covered under the subscribers manager.

Movement between plans only happens if the difference between them is obvious, and the differences that work are the ones a viewer already understands about their own household. How many screens can watch at once is the clearest of them, because everybody knows how many people live in their house. Video quality is the next, particularly where somebody has bought a large television. Slices of the catalogue work least well, because a viewer cannot evaluate what they are missing. Whichever you choose, the number lives on the plan rather than in code, so adding a tier or changing what one includes is an afternoon. The plans themselves are defined under packages and subscriptions, the screen limits under device management, and everything that happens between renewals under the subscriptions system.

A plan change is an event against the account, with the adjusted charge recorded alongside it, which matters twice. It matters to support, because a question about why a bill looks different has a factual answer rather than a reconstruction. And it matters to you, because the pattern of movement between tiers is one of the more useful things you can know about your pricing. A tier nobody ever upgrades to is priced wrong or described wrong, and a tier everybody downgrades from in month two was probably oversold. Those readings come out of the sales dashboard and reports and analytics rather than out of anybody counting by hand.