ObrussaLabs
For founders raising, and the people writing the checks

Does your reported ARR follow from your billing data?

Paste your subscription export and your reported ARR. This recomputes the number from your own events — setup fees, trials, churn, and annual plans all counted correctly — and shows you every place the headline doesn't follow. It forms no opinion about your company.

● Runs entirely in your browser. Your billing data never leaves your machine — it's processed in your browser and never uploaded.

checking engine…

Your billing export sample: padded deck sample: honest
Your subscription events — export from Stripe, Chargebee, or your billing table
or paste / edit the rows
Result
— load a sample or paste your export, then Reconcile
What it checks, and what it honestly doesn't

"ARR" is easy to say and easy to inflate. This computes it the honest way — 12 × the monthly recurring revenue of your active, recurring subscriptions — and compares to the headline you entered. Either the number follows from the events or it doesn't, and anyone who re-runs it gets the same answer.

What it catches

  • One-time charges booked as recurring. A setup or implementation fee is not ARR — the most common inflation.
  • Churned and cancelled subscriptions still in the total. Revenue that already left.
  • Trials and $0 plans counted as revenue. Unconverted pipeline dressed up as ARR.
  • Annual plans not normalized. A $12,000/year plan is $1,000 of MRR, not $12,000.
  • Duplicate subscription IDs — the same customer counted twice.

It reads your columns however they're named

Stripe, Chargebee, a spreadsheet — plan_amount, billing_period, subscription_status all get mapped, and it prints every mapping and default it assumed under the result, so a wrong guess is visible, not buried.

What it honestly does not do

It shows whether your reported ARR is derivable from the events you supplied — nothing more. It cannot tell you the events are real, that customers won't churn next month, or that ARR is the right metric. If the billing export itself is wrong, a faithful ARR computed off it reconciles perfectly. The question is always do the reported results follow from the source data?

Built by Aidan Shaughnessy. Runs the same checks as our tested Python engine — the badge above re-runs a reference case against this in-browser code on every load and confirms they match.
You run it on your own numbers · it forms no opinion about your company