Your pipeline dashboard says 4x coverage. Last quarter's forecast still missed by 30% anyway. If that gap sounds familiar, the problem probably isn't your close rate — it's that nobody was watching the four signals that catch a fake pipeline before the quarter ends, not after.
Why the pipeline number doesn't tell you it's lying
Pipeline value and coverage ratio are self-reported, lagging numbers. A deal that's been silent for six weeks reports the exact same dollar value as one that's closing Friday. Without a RevOps analyst cross-checking activity against stage, the only moment you find out the number was fiction is the moment the quarter closes short, which is also the worst possible moment to find out.
That's not a CRM problem. It's a measurement problem. You don't need a dashboard or a data team to catch pipeline inflation early. You need four proxy signals you can track in a spreadsheet, in about fifteen minutes a week, well before the number shows up wrong in a board deck.
The four signals that catch it early
None of these require new software. They require pulling the same four numbers from your CRM export or your own notes every week, and watching the trend instead of the snapshot.
- Stage-velocity drift. Track the median number of days each open deal has sat in its current stage. If that median creeps upward two weeks in a row without you tightening qualification, deals are stalling quietly while still counting toward your coverage.
- Stage-to-stage conversion compression. Each week, calculate what percentage of deals that entered a stage four weeks ago have advanced to the next one. A falling conversion rate between the same two stages means qualification is loosening upstream, and more of what's entering your pipeline was never going to close.
- Deal concentration, or lumpiness. Calculate what share of total open pipeline value sits in your two largest deals. Above roughly 40%, your forecast is only as real as those two deals, and one loss doesn't just hurt, it collapses your coverage ratio overnight.
- Replacement rate versus burn rate. Compare the dollar value added to pipeline each week against the dollar value removed through closed-won and closed-lost. If additions run behind removals for three straight weeks while total pipeline value holds steady or grows, deals are being kept open instead of closed, which is exactly how inflation builds without anyone deciding to inflate anything.
Running it without a CRM dashboard
Build one spreadsheet with four columns, one per signal, and one row per week. Update it every Friday afternoon, before you close out the week, using whatever export or manual list your CRM or notes give you. This takes about fifteen minutes once the format is set.
The value isn't in any single week's number. It's in the four-week trend. A healthy pipeline shows these four numbers holding roughly flat. A pipeline that's inflating shows at least two of them moving the wrong direction at once, usually two to four weeks before that shows up in a missed forecast.
What to do when a signal flags
If stage-velocity or conversion compression moves, run a deal-by-deal scrub against a checklist like this one, and close what's actually dead instead of waiting for it to age out on its own.
If concentration is high, treat your two biggest deals as the forecast, not as bonus coverage. Everything else in the pipeline is upside, not the base case, until it isn't.
If replacement rate is falling behind burn, you have a top-of-funnel problem masquerading as a forecasting problem. Padding an existing pipeline number won't fix it. Here's more on how coverage math actually works once the number itself is honest.
The one thing to set up this week
Build the four-column spreadsheet today, even with imperfect historical data. Log this week's numbers Friday. In a month you'll have a real baseline, and you'll walk into your next forecast call knowing whether the number on the slide is a fact or a hope, before anyone in the room has to ask.
Frequently asked questions
How often should I check these signals?
Weekly, on the same day each time. Monthly checks miss the early window where you can still fix the problem before it shows up in a closed quarter.
Do I need a CRM to do this?
No. A spreadsheet and your own notes work fine below roughly 20 open deals. A CRM just automates the pull once your pipeline is too large to track by hand.
What if I don't have historical data to set a baseline?
Start logging this week. After four weeks you have a trailing trend, which is what actually matters here, not a perfect historical baseline.
Isn't this the same as a pipeline scrub?
No. A scrub is a one-time cleanup where you close deals that are already dead. This is the ongoing measurement layer that tells you when a scrub is overdue, before your dashboard forces the conversation.
Which signal matters most for an early-stage startup?
Deal concentration, usually. With fewer total deals, one or two large opportunities can single-handedly make an otherwise thin pipeline look healthy, right up until they don't close.
None of these four signals require permission, budget, or a hire. They require fifteen minutes on a Friday and the discipline to look at the trend instead of the number that makes this week's update easier to write.