A pipeline is an opinion, not a measurement.
The numbers came from somebody who had reason to be optimistic.
Two things that get conflated
“Which stage are we at” is a human judgement and should be — the person who spoke to the client knows better than any system. But “won or lost” is an outcome, and an outcome should not be something anyone selects.
When those two collapse into one field, the pipeline stops being defensible. Somebody set it to won, and nothing records when, why, or where it was before.
The two-axis model
A person says the stage; the system derives the outcome
Nine invariant stages whose labels you write yourself — “site survey”, “sample sent”, whatever your work actually contains. The meaning and the order stay fixed; the names are yours.
The outcome is derived and is never written by hand. And every movement writes a history row that is never edited and never deleted — so “three months in negotiation” is a fact rather than somebody's impression.
- Your labels, fixed meaning
- The outcome is derived — there is no field to type it into
- History is append-only: no edit, no delete
Proposal sent
Pars Petrochemical
Financial due diligence
4.8 B T
Alborz Foods
Valuation
1.2 B T
Fieldwork
Kavir Steel
FP&A retainer
320 M T / mo
Reporting
Sepehr Logistics
Statement spreading
180 M T
And why a view
The pipeline falls out of the memory
This is the real difference: the pipeline here is not a separate screen somebody has to maintain. It is a view onto what the team already writes in feeds and does in tasks.
Which is why it stays current — keeping it current is not extra work. If an engagement's last note was forty days ago, the pipeline says so, whatever its stage claims.
- Pulse raises the engagements that have gone quiet
- Indicators can set thresholds and surface a breach through Pulse
- The number you give in the meeting has something behind it
42 days since anyone recorded anything
Kavir Steel
Stage 4 of 9 — Fieldwork
Two fair questions
We have our own stages — do we have to change them?
Not the names: you write the label for every stage. The order and meaning of the nine are fixed, and that is exactly what makes comparison across services and teams possible at all.What if someone sets the wrong stage?
They change it, and the correction is itself a history row. The goal is not to prevent mistakes; it is that no change is ever untraceable.
Build a pipeline you don't have to maintain
Put your real engagements in, leave notes for a week, and see how different the picture is from that forecast meeting.
No card.