Stages are the spine of the whole thing

Before you can track a single deal, you have to answer one question: what are the stages? It sounds trivial, and most teams treat it that way — they copy a template, name a few columns, and start dragging cards. Then three months later the forecast is useless, half the deals are stuck in "Negotiation" forever, and nobody can say what "Qualified" actually means. The stages were wrong from the start.

A sales pipeline is only as honest as its stages. Get them right and the pipeline tells you the truth about your business. Get them wrong and you've built an expensive way to fool yourself.

The one rule: stages are buyer actions, not seller feelings

Here's the mistake nearly everyone makes. They name stages after how the rep feels about the deal: "Interested," "Hot," "Looking Good." These are emotions, not facts, and they can't be verified. A deal sits in "Interested" because the rep is optimistic, not because anything happened.

Good stages are defined by an observable buyer action — something that either happened or didn't, with no judgment call involved.

  • Bad: "Qualified" (qualified according to whom?)
  • Good: "Discovery call completed, budget owner identified" (did it happen? yes or no.)

The test for any stage definition: could two different people look at the same deal and disagree about which stage it's in? If yes, the stage is too vague. Tighten it until the answer is mechanical. This is the same discipline that makes discovery calls actually qualify — concrete, verifiable facts instead of warm feelings.

A default set that works for most small teams

You don't need a custom seven-stage masterpiece. Most small businesses are well served by five:

  1. New — a deal exists; you haven't yet confirmed it's real. Entry criterion: a contact and an estimated value.
  2. Qualified — you've confirmed a real problem, urgency, and a budget owner. Entry criterion: discovery completed.
  3. Proposal — you've sent pricing or a formal proposal. Entry criterion: the buyer has your numbers.
  4. Negotiation — they're engaged on terms, not deciding whether but how. Entry criterion: an objection or term is actively being worked.
  5. Won / Lost — closed, with a reason code captured.

Notice each stage has an entry criterion: the specific thing that must be true to move a deal in. Write these down. They are the actual definition of your pipeline; the stage names are just labels.

Fewer stages than you think

The instinct is to add stages for granularity — "Demo Scheduled," "Demo Completed," "Follow-up Sent." Resist it. Every stage you add is another place for deals to get stuck and another judgment call about where something belongs. A stage earns its place only if deals meaningfully pause there and you'd take a different action depending on whether a deal is in it.

If two adjacent stages always get moved through together, they're one stage. Five well-defined stages beat nine fuzzy ones every time.

Make each stage mean a probability

Once your stages are buyer actions, you can attach a realistic close probability to each — New 10%, Qualified 30%, Proposal 60%, Negotiation 80%. Now your pipeline isn't just a to-do board; it's a forecast you can defend, because the weighted value updates itself as deals advance through verifiable milestones. In Hitt CRM the weighted forecast is computed from exactly this — stage probability times deal value — so the number on your dashboard reflects reality instead of optimism.

Guard the stages with hygiene

Defining stages well is half the job; keeping deals honestly assigned is the other half. A deal that's been in "Proposal" for sixty days isn't in Proposal — it's stalling, and the stage is lying. Set an expectation that every deal carries a current next step and a close date, and treat stale deals as a signal to act, not a thing to ignore. The pipeline hygiene playbook covers the routine that keeps stages trustworthy week over week.

The one-sentence version

Define each pipeline stage as a verifiable buyer action with a written entry criterion, keep the list short, and attach a probability to each — do that, and your pipeline stops being a mood board and starts being a forecast.