Build your first pipeline

A pipeline is a shared claim about how your deals actually progress. If the stages do not mean anything specific, the forecast built on them cannot mean anything either.

This is the short version for a new workspace. The full guide to stages, probabilities and rules is linked at the end.

Who this is for

Whoever is setting up sales in a new workspace — usually a founder, sales lead or ops person.

Before you start

• Permission to create pipelines.
• A rough answer to one question: what has to be true for a deal to move from each stage to the next?

Build one, not five
You can run several pipelines — new business, renewals, partners — each with its own stages and probabilities, and move a deal between them when the motion changes. That is worth doing when you genuinely have different motions. It is not worth doing on day one, when you are still finding out whether your stages describe reality.

Design and build it

1. Write your stages down before you build them
   Name the steps a deal genuinely goes through. If two stages cannot be told apart by a specific thing that happened, they are one stage.
2. Keep it to five or six
   More stages feel more precise and are usually less accurate, because nobody agrees on where a deal sits. Fewer stages that everyone reads the same way beat a granular board people guess at.
3. Give each stage a win probability
   The probability is what turns a board into a weighted forecast. Be honest rather than optimistic — this number is doing arithmetic later.
4. Set your won and lost stages
   Mark which stages mean closed-won and closed-lost, so the system can tell the difference between a deal in progress and a deal that is finished.
5. Add rules only where you need them
   You can restrict skipping or backtracking, lock stages to admins, and require particular fields before a deal advances. Add these once you know where people cut corners, not pre-emptively.
6. Test it with one real deal
   Walk an actual opportunity through from first stage to close. Problems with your stage design show up immediately and are cheap to fix before the team is depending on it.

What you get

• A board your team can drag deals across and actually agree about.
• A weighted forecast built from probabilities you chose deliberately.
• Stage rules that keep the funnel honest where it matters.
• A tested design, because you ran a real deal through it before rolling it out.

Frequently asked questions

How many stages should a pipeline have?

Five or six is a good target. The test is whether two people would independently put the same deal in the same stage — if not, you have too many or they are not distinct enough.

Can I add or change stages later?

Yes, and you probably will. There are guardrails around removing a stage that still holds deals so you do not strand them.

Do I need more than one pipeline?

Only when you genuinely run different motions, such as new business versus renewals. Each pipeline gets its own stages and probabilities, and a deal can move between them.

What is a win probability for?

It weights the forecast. An open deal contributes its value multiplied by the probability of its stage, which is what makes the forecast a number rather than a wish.