How Buyers Test Your SaaS Pipeline in Diligence

Sales pipeline diligence is the buyer’s audit of your open deals and your sales history, and they run it because forecasts have stopped being believable on their face: the average B2B win rate fell to roughly 19% in 2025, down from about 29% the year before, in the Ebsta and Pavilion benchmark data. Buyers now assume a forecast is optimistic until the raw data proves otherwise.

So somewhere in week two of diligence, an email like this arrives: “Please provide an opportunity-level export of the sales pipeline, all fields, last eight quarters, including closed-lost.” That wording is illustrative, but the request is universal, and the buyer sending it is not checking whether your pipeline is big. They are rebuilding your forecast from raw parts to see how much of it they can underwrite, meaning how much they can safely count on when they wire the money. By the end of this post you will know exactly what they do to your export, column by column, and how to make your forecast survive it.

The Artifact: What Buyers Actually Ask For

Start with the export itself, because every haircut begins in its columns. A serious buyer wants deal-level rows, not summary charts: deal name, dollar amount, stage, created date, expected close date, lead source, last activity, and the buyer-side next step. They want closed-lost history too, because your losses are how they compute your real win rate, which is simply the share of deals you actually close out of the deals you chase.

The eight-quarter window is not accidental. It lets the diligence team watch deals age, move, stall, and die across multiple cycles, and check whether your stage definitions mean anything. A pipeline snapshot can be dressed up. Eight quarters of history cannot.

The Mechanism: Buyers Rebuild Your Forecast With Coverage Math

One division problem does most of the damage.

Sales teams talk about pipeline coverage, which is simply how many dollars of open pipeline you hold for every dollar of your sales target. The folk benchmark says 3x coverage is healthy. But the real requirement is set by your win rate, and the formula is one line: required coverage equals one divided by win rate. Close 25% of qualified deals and you need 4x. Close 20% and you need 5x. Enterprise teams closing 15 to 25% need 4x to 7x to forecast reliably, per Clari’s coverage analysis.

Now put yourself behind the diligence analyst’s screen. Your deck claims next year’s growth on 3x coverage. Your closed-lost history shows a 19% win rate. Their one-line math says you need more than 5x. The gap between what you claim and what your own history supports is not a rounding error to them. It is the exact size of the haircut, and it lands before a single deal is discussed.

Two more history checks follow. Concentration: in the 2025 Ebsta and Pavilion benchmark data, just 14% of sellers drive 80% of revenue, an 11x gap between top and bottom performers. If your pipeline lives in one rep’s column, the buyer prices the risk of that person leaving. And age: deals sitting far past your average sales cycle are not late, they are dead, and the buyer removes them no matter what the stage field says.

The Pipeline Haircut Schedule

Here is the whole process as one table, with illustrative round numbers for a company claiming a $10 million pipeline against its growth story. This is the schedule I walk sellers through before a buyer ever does:

Ten million of pipeline became one million of credited new revenue. That is not buyer cruelty, it is your own history applied without the optimism. The number that survives this schedule is the only version of your forecast that shows up in their model. And the split prices two ways: credited growth supports the multiple you are asking for, while uncredited growth gets restructured into an earn-out, where you get paid only if it comes true.

Red Flags and Green Flags in the Export

What sinks credibility fastest: close dates that all land on the last day of a quarter, stages with no dated activity behind them, a third of the pipeline created in the month before diligence started, win rates that are quietly computed with closed-lost deals deleted, and one heroic deal that is half the total.

What builds it: aged deals already marked dead by your own team before anyone asked, next steps written by the customer’s side, win rates you quote BEFORE the buyer computes them and that match the export, stage definitions that map to observable events, and pipeline spread across reps and sources. The strongest signal in the whole file is a founder whose claimed numbers are slightly tougher on themselves than the buyer’s recount. That founder gets believed on everything else too.

Making Your Forecast Survivable: Three Moves

The mechanism above tells you exactly what to do, and none of it is more pipeline.

Move 1: Run the haircut schedule on yourself, one quarter before market. Purge or mark the aged deals, chase real next steps on the rest, and recompute coverage against your true win rate. Present the post-haircut number as your forecast. It will be smaller and worth more.

Move 2: Reconcile your forecast to your financial model. The growth in the model buyers ask for must be derivable from the pipeline you hand over, at your historical win rate. When the two documents disagree, buyers believe the smaller one.

Move 3: Shift the story toward revenue you already keep. Expansion from existing customers now accounts for over half of new revenue in the same Ebsta and Pavilion benchmarks, and buyers underwrite retention-driven growth far more readily than new-logo forecasts, the same way they credit contracted, recurring ARR over hopeful ARR. The less your price depends on the pipeline, the less the haircut can take.

Frequently Asked Questions

Next Steps