Average Deal Size vs ACV: Which Number Belongs in Your Forecast

What is the difference between average deal size and ACV?

Average deal size is the mean value of deals you closed. ACV is the annualized recurring value of a single contract. They only match when every deal is a one year subscription with no services attached, which describes almost nobody.

A $300,000 contract signed for three years has a deal size of $300,000 and an ACV of $100,000. Report the first into an annual recurring revenue plan and you have claimed three years of value in one year. Report the second as your average deal size in a sales review and you have understated what the team sold by two thirds.

Neither number is wrong. Using one where the other belongs is what breaks the model, and it happens most often at exactly the point where a deal value gets multiplied by a win rate to produce a forecast.

What is average deal size?

Average deal size is total booked value divided by the number of deals closed in a period. Most teams calculate it on total contract value including one-time fees, which makes it a measure of commercial size rather than recurring value.

It is a useful operational metric. It tells you whether the team is moving upmarket, whether discounting is drifting, and whether a segment shift is working. Tracked by rep and by segment, it also surfaces coverage problems, because a rep whose average deal size is half the team's needs a different pipeline volume to hit the same number.

The metric has one persistent flaw. It is an average, and deal value distributions in B2B SaaS are skewed. One large contract can pull the average well above what a typical deal looks like. Reporting the median alongside the mean costs nothing and prevents a category of bad conclusions.

What is ACV?

ACV is the recurring value of a contract normalized to a single year. Divide recurring contract value by the term in years, and exclude one-time fees.

ACV exists to make deals comparable. A one year deal at $120,000 and a three year deal at $360,000 represent the same annual commitment, and only ACV shows that. Without normalization, a team that shifts to multi-year contracts appears to double its deal sizes while the annual revenue impact is unchanged.

Ramped contracts are where ACV construction gets sloppy. A deal priced at $100,000, $150,000, and $200,000 across three years can be reported as $150,000 ACV using the term average, or $100,000 using the current step. Both are defensible. What is not defensible is switching between them across quarters, which is common and produces a deal size trend that reflects nothing but reporting choices.

When do the two numbers separate?

Whenever contract terms are longer than a year or non-recurring revenue is attached. The table shows the same four deals measured both ways. Four deals that mostly represent $120,000 of annual recurring commitment produce an average deal size of $293,750. If that figure feeds an ARR plan, the plan is overstated by more than 100 percent. Deal B alone triples its annual contribution, and deal C counts a setup fee as recurring revenue.

The fix is not to abandon average deal size. It is to know which one a given calculation requires and to label the column.

Why is pipeline average deal size higher than closed-won?

Because deals shrink on the way to signature and the CRM amount usually does not follow them down. This is one of the more expensive gaps in a forecast, because it multiplies against every other assumption.

Sellers enter an amount early, when the scope is optimistic and the discount has not been negotiated. Between that entry and the signature, procurement cuts, the scope gets phased, seats get trimmed, and a multi-year term gets shortened. Most deals close for less than the value they carried in the CRM, and nobody goes back to restate the pipeline record.

The magnitude is not small. A pipeline with an average deal size of $80,000 against closed-won deals averaging $40,000 is a pattern worth checking for in your own data. If that gap exists, every expected-value calculation built on pipeline amounts is overstated by half before a win rate is even applied.

Two related distortions compound it. Roughly 10 percent or more of a typical pipeline is stale, meaning untouched for twelve months, and those records carry their original inflated amounts. And of the pipeline carrying close dates inside the quarter on day one of that quarter, about 20 percent actually closes, so 80 percent of the value sitting in the quarter is not realized in it.

Which number should drive a pipeline forecast?

Closed-won ACV, segmented, not blended pipeline deal size. The substitution is simple and it corrects a systematic bias rather than a random error.

Calculate historical ACV on closed-won deals by segment. Apply that to qualified opportunity counts rather than to the amounts sellers entered. The forecast now reflects what your business actually converts at instead of what it hoped for at first contact.

Blending across segments reintroduces the problem. Enterprise and mid-market ACVs can differ by an order of magnitude, and a blended average describes neither. Segment first, then average.

This also changes how pipeline coverage should be read. Coverage ratios in the 3x to 5x range are the standard, and across ORM's customer base most land near 3.5x, with a spread from 1.4x to 5x. But coverage measured in inflated pipeline dollars is coverage of a number that does not exist. Recalculate coverage using closed-won ACV against opportunity counts and the ratio often looks materially worse, which is the more honest starting point. Our piece on why the 3x pipeline coverage rule is wrong covers what coverage can and cannot tell you.

How do you keep deal size honest in the model?

Measure the gap between pipeline and closed-won deal value as a standing metric, and treat movement in it as a market signal. The gap itself is information.

When a new competitor enters and creates pricing pressure, average deal size falls. That is one of the clearest early reads on a change in market conditions, and it moves before win rates do. A forecast built on last year's deal size assumption will miss for a full quarter before anyone traces the cause, because the model was responsive to volume and not to price.

Track closed-won ACV by segment monthly, alongside the pipeline-to-closed ratio and discount depth. When the ratio widens, either sellers are inflating entries more than usual or deals are being cut harder in negotiation, and the two have different fixes. Either way you want to know inside the quarter rather than at the annual planning session. Our guide on how to create a sales forecast walks through where deal value assumptions sit in the model and how to keep them current.