One honest number, without hiring an analyst to produce it

The problem is not that you lack charts

Stripe already gives you charts. So does whatever you built in a spreadsheet, and so does the analytics tool you trialled and did not keep. The problem a founder actually has is that the numbers disagree, nobody owns the definitions, and the person who has to reconcile them is you — usually the night before you need them.

That is a definitions problem wearing a dashboard costume. Stripe reports what it billed: trialing users count as customers, your own test account is in there, and delinquent subscriptions are handled inconsistently. Your spreadsheet reports whatever you decided last time. Neither records the decision, so next quarter you make it again, differently.

What changes when the rules are visible

In Bastle the rules are workspace policy rather than a vendor constant. Whether a delinquent subscription is still revenue after thirty days or sixty, whether trials are excluded, whether a full-discount coupon is churn, whether a same-day start and cancel is a wash, how far back annual churn looks — you set them, and changing one recalculates your entire history against the new rule. A versioned record of who changed which rule and when is on the roadmap, so for now write the change down somewhere yourself.

The practical effect is small and enormous at once: you can answer the question “why is this different from Stripe?” without opening a spreadsheet, and you can answer it the same way twice. How metrics are calculated covers the mechanics.

Doing this without an analyst

The reason founders end up with a spreadsheet is that the interesting questions cross boundaries. Which cancellation reason costs the most MRR. What share of last month's churn was a failed card rather than a decision. Whether the cohort you acquired after the pricing change retains better than the one before it. Each of those is a join between two datasets that usually live in two products.

Bastle runs the measurement, the dunning and the cancellation flow off one normalised billing history, so those questions have one answer instead of three exports. And Ask will do the assembly for you — in plain English, with the records it used attached, so you can check it in about ten seconds rather than trusting it.

The parts that are worth switching on later

Connect Stripe and the measurement half works immediately: metrics, cohorts, segmentation, Ask. Recover and Cancellation Insights stay off until you review them, because they send messages to your customers and that should be a decision rather than a default. Forecasting is in beta and labelled that way everywhere it appears — useful as a strong second opinion, not as the only input to a hiring decision.

What this does not do

Bastle does not do revenue recognition, and if your accountant needs ASC 606 schedules you need a tool that does. It does not do attribution, so it will tell you what you earned and not which campaign caused it. It has no warehouse destinations — no Snowflake or BigQuery sync — though the REST API and webhooks mean you can move the data yourself. And there are no customer stories on this site, because Bastle has none yet.