Buyouts.ai
All posts
Buyers

How to Verify MRR Before Buying a SaaS Business

Reported MRR is the easiest SaaS metric to inflate. How to reconcile recurring revenue against the processor, bank and billing records before you buy.

By the Buyouts team

July 2026 · 9 min read

Short answer: to verify MRR before buying a SaaS business, reconcile the seller's reported recurring revenue against three independent sources: the payment processor (Stripe, Paddle, Chargebee), the business bank account, and the billing system's active-subscription export. Real MRR is the number that agrees across all three after you strip out one-time charges, refunds, failed payments and annual prepayments spread to a monthly figure. If the reported number is higher than what survives that reconciliation, the gap is your negotiating leverage. Last updated July 2026. Educational only, not financial or investment advice.

Why MRR is the number that gets inflated

Monthly recurring revenue is the headline metric in almost every SaaS listing, and it is also the easiest one to present generously. A seller can count trials as active, include one-time setup fees, book annual contracts at their full value in the month they were signed, or quietly leave churned accounts in the total. None of this has to be fraud. Different tools calculate MRR differently, and an optimistic default in a dashboard becomes a listing number nobody double-checked. When you buy an MRR business, that recurring revenue is the entire asset, so a ten percent overstatement is a ten percent overpayment compounded by whatever multiple you pay.

The job of diligence is not to catch a liar. It is to confirm that the revenue you are paying a multiple on is real, recurring, and likely to still be there next year. That means going past the dashboard screenshot to the systems that actually move money.

How do you verify MRR when buying a SaaS?

Verify MRR by reconciling three sources that are hard to fake together: the payment processor's own reporting, the deposits in the business bank account, and the billing system's list of currently active subscriptions. Pull twelve to twenty-four months from each, line them up month by month, and confirm they agree once you normalize for one-time revenue and annual prepayments. Numbers that reconcile across all three are trustworthy. A number that only appears in a spreadsheet the seller made is not evidence yet.

SourceWhat it provesWhat to watch for
Payment processor (Stripe, Paddle, etc.)Gross and net recurring charges, refunds, failed paymentsOne-time charges counted as recurring; high involuntary churn hidden in failed payments
Business bank statementsCash actually landed, net of processor fees and chargebacksDeposits that do not trace back to subscriptions; personal funds propping up the total
Billing system active subscriptionsCurrent live subscriber count and plan mixTrials or paused accounts counted as active; discounts not reflected in the MRR figure
Tax return / P&LRevenue the seller was willing to report to the IRSReported revenue well below the listing, which is a red flag either way

The reconciliation, step by step

Start with the processor export because it is the closest thing to ground truth for a subscription business. Filter it to recurring charges only, subtract refunds and chargebacks, and you have net recurring revenue per month. Then pull the bank statements for the same period and confirm the processor payouts actually landed. Any month where the processor says one thing and the bank says another needs an explanation before you go further. Getting bank data into a form you can actually work with used to be the slow part; today you can convert a stack of PDF bank statements into a clean spreadsheet in minutes and match deposits against processor payouts side by side.

With processor and bank agreeing, cross-check the billing system's active-subscription list. Count live, paying subscriptions and multiply by their real, discounted prices. This should land close to your processor-derived MRR. If the billing system shows more revenue than the processor collected, you are looking at accounts that are billed but not paying, which is churn waiting to happen.

Finally, handle annual plans deliberately. A customer who paid for a year up front is real revenue, but booking that whole payment as one month of MRR inflates the figure twelve-fold for that account. Spread annual contracts to their monthly equivalent so your MRR reflects the recurring rate, not the billing calendar. Then note separately how much of the base is annual, because a business that is 80 percent annually prepaid has very different cash timing than one billed monthly.

What is a good churn rate for a SaaS acquisition?

For small and mid-market SaaS, monthly logo churn under roughly three to five percent is generally healthy, and net revenue retention above 100 percent is excellent because expansion revenue outpaces cancellations. Churn is not a side metric here. It is the decay rate on the exact asset you are buying, so verify it the same way you verify MRR: from the processor and billing records, not the seller's summary. A business with strong MRR and quietly rising churn is worth less than its headline suggests.

Red flags that should change your offer

  • Revenue that will not reconcile. If processor, bank and billing cannot be made to agree, and the seller cannot explain the gap, treat the lower number as real and price accordingly.
  • Customer concentration. One account that is 20 percent or more of MRR is a single point of failure. Verify that customer's contract, tenure and renewal terms specifically.
  • A recent revenue spike. A jump in the last two or three months before listing can be real growth or it can be timing dressed up for sale. Look at the trailing twelve months, not the last quarter.
  • High involuntary churn. Failed-payment cancellations buried in the processor data mean the true retained base is smaller than the active count suggests.
  • Discounts not in the MRR math. If half the base is on a legacy 50 percent coupon, list-price MRR overstates what actually recurs.

Why verified listings save weeks of this work

Everything above is what a careful buyer does manually on an unverified listing. It is necessary, and on a mass marketplace it is entirely on you. The alternative is buying where the reconciliation is already done. On Buyouts, every listing arrives with verified MRR, ARR, growth and churn confirmed against source systems before it goes live, which turns the reconciliation from a two-week project into a review. That does not remove your own diligence, and it should not, but it means you start from numbers that already survived the test rather than from a screenshot you have to disprove.

If you want to see how this fits the broader process, the buying a SaaS business guide walks through sourcing, offer and close, the SaaS due diligence checklist covers the non-revenue items to verify, and when you are ready to browse deals with the numbers already confirmed, the buy an MRR business page is where the recurring-revenue listings live.

The bottom line

Verify MRR by making three independent systems agree, normalize for one-time and annual revenue, and treat any gap as leverage rather than a rounding error. Recurring revenue is only worth its multiple if it is genuinely recurring, and the fifteen minutes you spend reconciling processor to bank to billing is the highest-return work in the entire acquisition.

The M&A marketplace for AI SaaS

Browse anonymized AI SaaS listings with verified MRR, multiples, growth, and churn. Vetted buyers, escrow-backed closes. See how it works or value your AI SaaS. Educational content only, not financial advice.

Get full access to the AI SaaS deal room

Buyouts is the marketplace built for AI SaaS: verified metrics, vetted buyers, escrow-backed closes. Browse AI SaaS for sale, review pricing, or learn how it works.

Verified metrics · Vetted buyers · Escrow-backed closes

Educational only, not financial, investment, tax, or legal advice · we never guarantee a sale price or return · listings and examples are anonymized and illustrative.