ARR vs GAAP Revenue: Why Your Two Revenue Numbers Disagree

What is the difference between ARR and GAAP revenue?

ARR is a point-in-time snapshot of committed recurring revenue projected forward twelve months. GAAP revenue is the sum of revenue actually earned across a period that has already closed. One looks forward from today. One looks backward across a year.

That is why a company can end December with $30 million of ARR and report $22 million of revenue for the same fiscal year without either number being wrong. The ARR figure describes the business as it exists on December 31. The revenue figure describes what the business earned in January, February, and every month after, when it was smaller.

Teams get into trouble when they present the two side by side as though the difference is an error to reconcile. The difference is growth. Understanding its shape tells you more about the business than either number alone.

What is ARR?

ARR is the annualized value of recurring subscription revenue from active contracts, measured at a single moment. You build it by summing the annualized recurring value of every live subscription and excluding anything non-recurring.

It is a management metric, not an accounting one. No standard defines it, which means construction varies. Some companies include committed contracts that have not started billing. Some annualize month-to-month subscriptions. Some fold in usage minimums, others do not. Two companies reporting $50 million of ARR can be running materially different businesses.

The value of ARR is that it strips out timing. It does not care when a contract started, how it was invoiced, or whether the customer prepaid. It answers one question: what is the annual value of the recurring revenue we hold right now.

What is GAAP revenue?

GAAP revenue is contract value recognized as you satisfy performance obligations, governed by ASC 606, reported on the income statement, and subject to audit. For a standard subscription, it recognizes ratably across the service term.

Recognition follows delivery, not invoicing and not signing. A $600,000 three year contract signed in November and starting January 1 contributes zero revenue in the year it was signed. It contributes $200,000 in each of the next three years. Setup fees, professional services, and usage overages each follow their own recognition treatment.

GAAP revenue is the number that gets audited, taxed, and quoted in public multiples. It is also the slowest possible read on what your sales team accomplished last quarter, which is why nobody runs an operating cadence on it.

Why is ARR usually higher than GAAP revenue?

Because ARR annualizes your ending position while revenue averages your whole year, and a growing company was smaller for most of that year. Work through a simple case. Exit ARR is $21 million. Full year GAAP revenue is $15.5 million. The company grew 75 percent on ARR, and every dollar added in Q4 recognized for at most three months. Nothing is inconsistent. The gap is a direct function of the growth rate and where in the year the ARR landed.

The corollary is useful: the wider the gap between exit ARR and full year revenue, the more of your growth arrived late in the year. That is worth knowing, because late-landing ARR flatters next year's revenue comparison and can make a decelerating business look fine for another two quarters.

When is GAAP revenue higher than ARR?

When a meaningful share of your revenue is not recurring, or when the base is shrinking. Three situations produce it.

A services-heavy company earns implementation, migration, and consulting revenue that GAAP counts in full and ARR excludes entirely. If services are 25 percent of the top line, revenue can exceed ARR even at moderate growth.

A usage-based company bills overages above committed minimums. Only the minimum enters ARR. Everything above it is revenue with no ARR representation.

A shrinking company recognizes revenue across a year from customers who have since churned. Trailing revenue reflects a base that no longer exists. This is the version that matters, because revenue holds up for two or three quarters after ARR turns, and the income statement gives no warning.

Which number should you use for what?

Use ARR to run the business and GAAP revenue to report the business. The table below maps the decision.

How should the forecast handle both?

Model in ARR, convert to revenue with a recognition schedule, and never forecast the two independently. Two separate models will drift, and reconciling them at quarter close is wasted work.

The operating model should start from the committed ARR base, apply churn and expansion to get the retained base, and layer new bookings on top. That structure works because recurring revenue is the most predictable input a forecast has and new business is the volatile part. Our guide on how to forecast revenue walks through how those layers stack, and net revenue retention is the coefficient that decides what the base does before any new deal closes.

The recognition layer is arithmetic once the ARR forecast exists. Apply start dates and term lengths to the forecasted bookings, and GAAP revenue falls out. Doing it in that order keeps the drivers visible. The most common reason a forecast misses is that something in the business or the market changed and the model was still running on old assumptions. A model that surfaces the change in the ARR layer gives you a quarter to respond. A model that only shows recognized revenue gives you the news after the quarter is already over.

The pipeline that feeds new bookings needs the same discipline, which is why pipeline coverage should be read as an input to the ARR forecast rather than as the forecast itself.