Subscription and billing ops

Subscription and billing ops is an automation that runs the recurring billing lifecycle for a subscription business. It generates invoices when they are due, retries failed payments, chases cards that are about to expire, applies upgrades, downgrades, and proration when a customer changes plan, and flags revenue that should have been billed but was not. The routine, rule-based parts run on their own, including the dunning follow-ups that recover a failed charge. Disputes, refunds, and anything the system is not certain about go to a person, so real customer charges always have a human on the judgment calls.

Stripe already retries failed cards and can send basic dunning emails, and if that covered it you would not be reading this. This sits across Stripe, your billing tool, and your CRM and handles the parts Stripe does not: checking what was actually billed against what should have been, applying proration on mid-cycle plan changes, catching a subscription that was set up wrong and never charged correctly, and updating the customer record in Attio so billing and CRM agree. It uses Stripe's retry and card-updater tools rather than replacing them, and adds the judgment layer and the exception queue on top.

That is the risk it is built around. An automation that touches real customer charges can do real damage if it double-charges an account or chases someone who already paid, so it is set up to act only where the rule is unambiguous and to stop everywhere else. Before any charge runs it checks against the safeguards: has this already been paid, is this the amount the plan says, is this a duplicate of a charge from this week. Anything that does not pass cleanly is held for a person instead of pushed through. Disputes and refunds never run automatically. The safe direction is to flag and wait, and that is the direction it errs.

The routine cases are easy; the edge cases are where the care goes. Proration on a mid-cycle plan change, a partial refund that has to net against the next invoice, an account with an annual and a monthly plan running at once, a payment that succeeded but posted late, a discount that should have expired: these are exactly the situations where a wrong automatic move is worse than no move. So the automation handles the clean, rule-based cases end to end and routes the ambiguous ones to a person with the full context attached. It does not close disputes, decide refunds, or invent a billing rule that was never defined. Those stay human by design.

On billing, Stripe, Chargebee, Recurly, and Paddle. On the accounting side it can post to QuickBooks or Xero so your ledger matches what was collected. Customer records and plan status sync to your CRM, usually Attio, so billing and sales see the same picture. Dunning emails and exception alerts route through your email tool and Slack, so the right person sees a held charge fast. Most tools with an API can be added. The audit maps your exact stack and confirms where subscriptions, payments, and customer records actually live before anything is built.

Usually two to three weeks. The first days go to mapping your plans, pricing, proration rules, and where subscriptions and customer records live across Stripe, your billing tool, and your CRM. Then the retry logic, dunning sequences, and billing checks run against your real charges for a week or so, with a person reviewing everything the system flags, so the rules get tuned to your business before it acts on its own. You are recovering failed payments quickly, and that tuning window is what makes it safe to let the routine cases run without a person watching each one.

Two parts. Tooling runs as a monthly cost for the retry, dunning, and processing layer, usually a range that scales with how many subscriptions and charges you run through it. Implementation is a fixed scope, quoted once the audit maps your plans, pricing rules, and systems, so you are pricing a defined build rather than an open-ended retainer. The audit itself is where the scope and price get set against every other opportunity in your business, so you are not guessing at the effort up front. In most cases the revenue recovered from failed payments covers the cost well inside the first months.