Money decisions

The pipeline that was seventy percent won at twenty percent probability

Every deal on the board was marked 'commit.' Two-thirds of them slipped. The CRM was a wish list with a forecast column.

The pipeline that was seventy percent won at twenty percent probability
Illustration · Deimar Gutiérrez

A CRO showed me a pipeline review last quarter where forty-three of sixty-one open deals were marked commit. The historical close rate on commit deals was eighty-five percent. The implied forecast was a record quarter. The actual quarter closed at half the implied number.

The pipeline was not lying. The reps were not lying. The stage definitions had drifted until they meant nothing.

The drift starts politely. A rep moves a deal to commit because the buyer said something encouraging on a Tuesday. The CRO sees it and asks no question. The next rep, watching, learns that commit is the stage you use when the deal feels likely. Six months later, every rep uses it for every deal they want their manager to feel good about. The stage has stopped being a forecast and become an emotion.

The fix is not better software. It is exit criteria stated as behavior, not belief. A deal does not move to commit because the rep feels good. It moves because something observable has happened — a mutual action plan signed, procurement contacted by name, security review kicked off. If the thing has not happened, the deal does not advance. The rep loses the right to grade their own confidence.

This is unpopular because it shrinks the pipeline overnight. Forty-three commit deals become twelve. The number is honest. It is also embarrassing. The CRO has to walk into the next board meeting with a coverage ratio that looks worse, against a quota sized to the inflated version. The temptation to revert is enormous.

Companies that hold the line discover, two quarters later, that the close rate on the smaller pile is back to eighty-five percent — because the pile is now made of deals that behave like late-stage deals. Hiring plans built against the forecast stop missing. The board stops second-guessing the numbers because the numbers stop being wrong.

The pipeline you have today reflects how you defined the stages. If the stages are graded by feeling, the quarter will come in wherever feeling lands. Behavior is the only honest definition. Anything else is a wish with a percentage attached.