4% of deals close when sellers say they will

Only 4% of deals close on their predicted date.

That's not a data quality problem, or at least not only that. It's a structural one. Sellers have a recency bias that no amount of CRM hygiene fixes. A great call on Tuesday pushes a close date forward. A no-show pushes it back. The date in the field reflects how the seller felt after their last conversation, not when the deal will actually close.

If 96% of close dates are wrong, the coverage ratio built on top of them is also wrong. And the forecast built on top of that is a guess with extra steps.

The volume trap

Pipeline conversations in most companies start and end with one question: do we have enough? The answer usually comes back as a coverage ratio: 3x, 4x, whatever the company's standard is. If the number clears the threshold, the pipeline is declared healthy, and the conversation moves on.

The raw dollar number is the most overrated metric in pipeline management. A 4x coverage ratio filled with stale opportunities, wrong-fit prospects, and deals that haven't had a meaningful touchpoint in 60 days isn't a healthy pipeline. It's a number that passes a threshold while masking a problem.

Will Richings, Senior Director of Customer Insights at Security Scorecard, puts it in terms of quality and volume:

"There's quality, and there's volume. We've taken our finance model, our quota model, and taken it one step further; a pipeline targeting model that looks at historical conversion rates and what we need relative to our growth goals. Is it filled with ICP green accounts? Or is it all fluff?"

The more useful question is what the pipeline is made of, and whether it's actually moving.

Working backwards from go-live

A more reliable way to review a deal: don't ask the rep what the next step is. A rep will always give you the answer that makes the deal sound most likely to close.

Roy Urquez, President at QAD Red Zone, trains his RevOps team to work backwards from go-live instead.

"When someone buys something, they want to go live with it. That's when they start to get value from the product. So ask: what's the go-live date? How long is professional services? How long does legal and procurement take?"

If the timeline doesn't leave room for the close date on record, the close date is wrong.

It's a small shift in how a pipeline review conversation starts, but it changes what information surfaces. Deals that look fine on a dashboard reveal themselves as stuck. Timelines that seemed aggressive turn out to be impossible.

The cost of holding on

Closed-lost deals take 20% to 30% longer to close than closed-won deals. When a deal is going well, it moves. When it's going badly, sellers don't disqualify it; they wait. They send one more email. They hold the close date for another two weeks. The deal stays in the pipeline past the point where it has any realistic chance.

"It's because we're hanging on to hope," as Urquez describes it. "We're saying, no, they'll get back to me."

This is where exit criteria matter. A clear definition of what it takes to move a deal from one stage to the next cuts the number of zombie deals being carried, not because you force reps to clean up the CRM, but because the criteria give them a framework for making the call themselves.

Stage plus age

Stage alone doesn't tell you much. A deal sitting in stage 3 for 45 days is a completely different situation from a deal that reached stage 3 yesterday. Both show up identically in a coverage report.

Urquez runs what he calls a stage-plus-age forecast: how long has this deal been at this stage, relative to how long deals typically take to move through it? Deals that stall within a stage are the leading indicator that a close date is about to slip, often before the rep has acknowledged it. Catching them early means you're having the right conversation at the right time, not two weeks after the quarter has already broken.

Richings builds similar logic into how Security Scorecard monitors rep activity:

"We talk about days in stage. Reps should start to understand: if this deal is in this stage this long, why is that happening to me?"

The goal is to make the data legible enough that reps diagnose their own problems rather than waiting to be told.

You might not need more pipeline

There's a calculation worth running before the next conversation about pipeline coverage. Urquez frames it this way:

"If you increase your average deal size by just 2%, decrease the average days to close by 3%, and lift your win rate by 2%, you're bringing in more revenue and hitting your number quicker, with the same pipeline."

+2%Average deal size

−3%Days to close

+2%Win rate

The point is that velocity metrics tell you things coverage ratios can't. A deal that's been sitting in stage 3 for 45 days is a different problem from a deal that moved from stage 1 to stage 3 in 10 days. Both show up the same in a raw coverage number.

What a quality conversation actually looks like

There's a practical tension in pipeline discipline: push too hard on hygiene and reps flush everything out of fear. You end up with a cleaner CRM, a smaller pipeline, and the same underlying data issues because the reps who were sandbagging are now just entering less.

The answer is specificity. Not "is this deal healthy" but "what happened on the last call, who was in the room, what's the go-live constraint, what does stage exit look like from here." Questions that require a real answer, not a confidence score.

Urquez is direct about what enthusiasm from a rep is worth in that conversation:

"I'm glad you're excited about this deal. That doesn't mean anything to me at the end of the day."

The same problem exists in your CS team

The healthy pipeline lie isn't only a sales problem. It shows up in renewals and expansions too, just with different numbers attached.

A common version: an account looks solid because it renewed last quarter. But licence activation is at 10%. The customer bought the product; they haven't used it. That's a churn risk sitting inside what looks like a healthy book of business.

Mohit, who leads product strategy at Google and previously ran GTM ops there, describes how his team monitors this:

"We track licence usage on a weekly and monthly basis. When usage is up 10 or 15%, we call to find out if they've expanded. When it's down 10 or 15%, we call for a different reason."

The signal is usage movement, not contract status.

Most CS pipeline reviews don't track this. They track renewal date and last contact. Neither tells you whether the customer is actually getting value, which is the only variable that reliably predicts whether they'll renew.

Building a number you can defend

All of this is useful in practice, but it only changes your conversation with leadership if you can put a number behind it. The most durable version of pipeline quality isn't a qualitative argument. It's a stage-by-stage model of how much pipeline you need to enter a quarter with, based on what has actually closed historically.

The formula is straightforward. Take your quarter booking goal, multiply it by the percentage of your closed-won deals that were in a given stage at the start of that quarter, then divide by your historical close accuracy for that stage. The result is your exit rate: the pipeline target for that stage on day one of the quarter.

As an example: a $10M booking goal, where 20% of won deals typically enter the quarter in stage 3, and stage 3 closes at 80% accuracy. $10M × 20% ÷ 80% = $2.5M. That's how much pipeline you need in stage 3 on January 1st to give yourself confidence in the number.

Run it for each relevant stage, and you have a pipeline target that's grounded in your own historical data, not an industry benchmark or a rule of thumb. It also gives you something to track mid-quarter: if you're at 60% of your stage 3 exit rate at the halfway point, you have a specific, defensible reason to raise a flag, not a feeling that things look thin.

Measuring pipeline quality is one problem. Communicating it is another. How do you take a pipeline quality story to a CRO, a CFO, or a board in a way that lands and drives action rather than defensiveness? That's where most teams get stuck, and where the work of building a credible picture gets lost.

That's the question on the table in Boston.