About
What we are building, and what is not built yet
Bastle is new. This page is the version of an about page that a founder who has been lied to by vendors can actually use: the argument, the status of every part of it, and the things Bastle will never do.
What is Bastle, and who is using it?
What it is
Answers, not a wall of charts
Bastle connects to your billing system, normalises the history into a single revenue model, and gives you a way to interrogate it. You ask why MRR moved and get a sentence, the drivers behind it, and the exact customers and invoices the answer was built from. The metric definitions are workspace policy rather than vendor defaults, so churn means what your business means by churn. Dunning, cancellation insights, segmentation, cohorts, forecasting and investor data rooms sit in the same product rather than in four separate subscriptions. The whole thing is reachable over a REST API, from a CLI, and by agents through an MCP server, because a web dashboard should be one client of your revenue data, not the only one.
Why
The category's default option stopped moving
This is not a claim that the incumbent is bad software. It is widely liked, and it created this category. The argument is narrower: the things it has not shipped are structural, and they are the things a revenue tool most needs now.
Baremetrics was acquired by Xenon Partners, a private equity firm, in November 2020, for a figure publicly reported at around $4M — roughly one times revenue. Its current chief executive has said publicly that it has become harder for the company to be agile. The years since have gone mostly into repackaging existing capability into new tiers rather than into new capability.
Four gaps follow from that, and they are corroborated by independent reviews rather than by us:
- No shipped AI. Natural-language querying has been discussed publicly by the company and, as of this research, remains unshipped. Third-party comparisons still list the absence of AI-powered insight as a gap. Meanwhile Stripe’s own Sigma assistant translates questions into SQL over Stripe data today.
- Black-box metric definitions. Churn windows, trial handling and refund treatment are fixed, with no configuration and no audit trail, which is why the reported MRR frequently fails to reconcile against Stripe and there is no supported way to find out why. Competitors including ChartMogul expose these rules directly.
- A thin data surface. No warehouse export and no webhooks, against a market where warehouse destinations and custom import APIs are table stakes.
- No free tier. This is the most common complaint recorded against the product. Add-ons stack, revenue-based pricing charges you more as you grow, and ChartMogul and ProfitWell both offer a genuinely free entry point.
Bastle is built directly against those four. Answers before dashboards. Definitions you set and can inspect. A warehouse and API surface treated as a first-class client. And free, so that the pricing objection is not an objection at all.
One concession worth making plainly: none of this is a moat. Stripe could extend Sigma into opinionated SaaS metrics, and ChartMogul could add natural-language querying. The bet is on shipping the answer layer and the trust layer faster than the incumbents get to them, and that bet is currently unproven.
Competitor facts above are drawn from public coverage of the acquisition, the companies’ own documentation and changelogs, and independent review sites. Last reviewed July 2026. Competitors ship continuously, so treat anything here as a starting point and check it against the vendor. If something has gone out of date, it gets corrected and re-dated.
Status
What is true today
Bastle is in beta, free, and has no customers. Below is the state of every substantial part of the product, including the parts that do not exist. It is checked against the source, not against the rest of this website.
- Shipped
Subscription metrics
28 metrics computed from your billing history in one engine, backfilled to your first customer. Every figure opens onto the customers, subscriptions and invoices behind it.
- Shipped
MRR movement ledger
Each day decomposed into new, expansion, reactivation, contraction and churn, per customer, so a fall in revenue tells you which of those it was.
- Shipped
Configurable metric definitions
Delinquency window, annual churn lookback, trial exclusion, full-coupon handling and same-day start-and-cancel washes are workspace policy, not engineering constants.
- Shipped
Ask
Plain-English questions over your connected data, answered with the records the answer was derived from. Available in the app, over the API and through MCP.
- Shipped
Anomaly detection in the daily brief
A daily pass flags unusual churn, failed-payment and expansion movement and names the drivers. It arrives as a brief; it is not yet a routed alerting system.
- Shipped
Recover and Cancellation Insights
Dunning sequences with a hosted card-update page, and cancellation reasons captured in the moment with reason-specific offers. Included, not a separate product.
- Shipped
API, import API and webhooks
A versioned REST API over metrics, customers, segments and events; an import endpoint for sources with no native connector; outbound webhooks with rotatable secrets.
- Shipped
CLI and MCP server
The product is drivable from a terminal and by agents. Metrics, customers, segments, cohorts, goals, forecasts, webhooks and Ask are all reachable without opening the web app.
- Beta
Forecasting
Projections run from live subscription history, with scenarios, runway and accounting actuals from QuickBooks Online or Xero. Marked beta because the assumptions layer is still moving.
- Beta
Connector coverage beyond Stripe
Stripe is the path we have exercised hardest. Coverage for the other billing and accounting sources is uneven, and each one carries its own status on the integrations page rather than a blanket claim here.
- Not built
Versioned audit log for definition changes
Definitions are yours to change today, and history recalculates when you do. What does not exist yet is the versioned, attributed record of who changed which rule and when. Until it does, treat a definition change as untracked.
- Not built
Reconciliation view against the billing provider
Every figure opens onto the formula, the period boundaries and the exact customers, subscriptions and invoices behind it, so a number can be traced to its records. What does not exist is the view that itemises, line by line, why a Bastle figure differs from your provider's — that bridge is still assembled by hand.
- Not built
Warehouse sync
CSV export works today. Managed sync into Snowflake, BigQuery or Redshift is a waitlist, not a feature. If a warehouse destination is a requirement, Bastle does not meet it yet.
- Not built
Alerts and Benchmarks
Both are on the roadmap and neither is usable. Anywhere they appear on this site they are labelled roadmap for that reason.
- Not built
Revenue recognition
No ASC 606 or IFRS 15 schedules, no GL journals, no locked periods, no audit-grade subledger. Bastle is not the system of record your auditor should be reading.
- Not built
Customers
Zero. No case studies, no testimonials, no logos, no adoption numbers. Every placeholder on this site is an obviously fake company name for that reason, and none will be replaced until a real company has agreed to be named.
The honest summary: the metrics, the movement ledger and the answer layer are real and running against live billing data. The provenance tooling around them is half-built. Everything else on this list is either beta or a promise.
Non-goals
What Bastle will not become
A roadmap says what a company hopes to do, which costs nothing. Naming what it refuses to do is the part you can hold it to.
We will not become an ERP
Bastle reads your billing system. It does not bill, invoice, dun your ledger or hold your chart of accounts. Products that try to be both end up mediocre at each.
No audit-grade revenue recognition
Deferred revenue schedules, GL journals and locked periods are a different discipline with different liability. If you need ASC 606 automation, buy a tool built for it — several are named on the comparison pages.
Not a CRM
Bastle will sync to the CRM you already have rather than asking you to move deals into it. Owning the pipeline is not the job.
No SQL tax on basic questions
Answering 'why did MRR move' should not require writing a query against a billing schema. Bastle exposes SQL-grade depth through the API, but never as the price of admission.
Your data is not the product
Revenue data is not sold, brokered or used to train anything for anyone else, and it does not enter a benchmark set you did not explicitly opt into.
No sales gate
Nothing on this site says contact us for pricing, and no capability is held behind a demo call. Bastle is free while in beta, and you can see the whole product without talking to anyone.
Principles
The commitments the product is held to
These are not values in the poster sense. Each one is a constraint that has shaped how the product is built, and each one is checkable against the software.
Every number expands to show its work
A figure you cannot decompose is a figure you cannot defend. Click any metric and you get the formula, the policy in force, the period boundaries and the underlying records.
The definitions are yours
Whether a delinquent subscription is still revenue is a business decision, not a vendor default. Bastle ships opinionated defaults and then gets out of the way when you disagree with them.
Reconciliation, not assertion
When a Bastle figure differs from your billing provider's, the product owes you the line-by-line reason — trials excluded, test accounts removed, refunds applied — rather than asking you to pick a number to believe. This one is a commitment rather than a description: today Bastle gives you the formula and the source records, and the automated variance view is on the roadmap.
Your data leaves as easily as it arrives
Export is a feature, not a retention lever. There is no export fee, no throttle and no support ticket required to take your history somewhere else.
Roadmap items are labelled roadmap
Anything not built says so on the page where it appears, in the same type size as the things that are. That rule is why the status list above includes the empty rows.
Corrections are bugs
A wrong fact about a competitor, a stale research date or an overstated capability gets fixed the way a crash gets fixed. That applies when the correction makes Bastle look worse.
Check it yourself
Nothing above needs to be taken on trust
The live demo runs the real product against a seeded dataset, so every claim on this page is inspectable before you connect anything of your own.
Questions a sceptic should ask
Does Bastle have any customers?
Is Bastle really free, or is this a trial?
Is Bastle safe to run a business on today?
Why build another subscription analytics tool?
What happens to my data if Bastle shuts down?
Will my revenue data be used for benchmarks?
The only useful test is your own data.
Connect Stripe and Bastle backfills your billing history, so you can put its numbers next to the tool you already trust and find out where they disagree. Free while in beta, no card required.