Your forecast is fiction (until you instrument the funnel)

Every revenue meeting ends the same way: a pipeline number, delivered with a confidence the number hasn’t earned. The forecast is presented as a fact. It is a story, and the story is only as good as the instrumentation behind it.

Here is why the fiction holds together long enough to be believed. A forecast is a projection built on funnel data, and funnel data is a series of stage transitions: lead to MQL, MQL to SQL, SQL to opportunity, opportunity to closed. Those transitions are supposed to be measured. In most companies they are inferred, because the stages were never defined precisely enough to measure. A rep moves a deal to “demo stage” when they feel like it. Another rep’s “demo stage” means something different. The conversion rate between two inconsistently defined stages is not a metric. It’s a rounding error with a decimal point.

The result is a forecast nobody can defend. The VP of Sales brings a pipeline number into a board meeting and gets asked where it came from, and the honest answer is a spreadsheet that aggregates reports pulled from a CRM whose stages don’t mean the same thing across two reps, let alone two quarters. That number isn’t a forecast. It’s a wish with formatting.

This is an instrumentation problem, and it has an engineering answer. You treat the funnel the way an SRE treats a service: define the stages precisely, instrument the transitions, and measure the rates that actually exist. Stage definitions are a schema: what has to be true for a record to be in this stage, written down, agreed, enforced. Conversion rate is the ratio between two adjacent stages, computed from records that actually moved. Velocity is the time between transitions, measured, not guessed. Drop-off is the difference between what entered a stage and what left it, attributed to a reason where possible.

None of this is exotic. It’s the observability discipline every production system already has, applied to the one system that generates revenue. The reason it doesn’t get applied is that the funnel lives in a CRM owned by sales, and sales treats instrumentation as overhead, not as the foundation of every number they quote. So the team runs on lagging reports and intuition, and the forecast stays fiction.

The cost of an undefendable pipeline number is concrete. It’s the board meeting where the number gets cut mid-quarter and nobody can explain why. It’s the hiring plan built on a pipeline that was never going to close. It’s the missed quarter that surprised everyone, because the funnel was telling the truth the whole time and nobody had instrumented it well enough to hear.

A forecast you can defend is a forecast built on a funnel you can see. Make the stages legible, measure the transitions, and the fiction collapses into a number you can actually stand behind.

Your forecast is only as good as the funnel instrumentation behind it. I make the funnel legible. Worth a 30-minute look at your stack? → Book a call.