Deals stuck in the wrong column make forecasts unreliable. Clear exit criteria fix that: define your sales pipeline stages by what the buyer has done, not by what the rep feels.
For the wider picture, see the B2B sales process.
| Item | Value |
|---|---|
| Core stages in one common model | 8 stages |
| Stages in another published framework | 9 stages, Prospecting to Post-Sale |
| Each stage needs | Entry criteria and exit criteria |
| Lead qualification exit criterion | Lead meets criteria the revenue team agreed on |
| Proposal exit criterion | Proposal reviewed by the decision makers |
Figures come from Avoma and CaptivateIQ, both vendor blogs shown as neutral examples. Stage counts and wording change, so check the sources before you rely on them.
What a pipeline stage actually is
A sales pipeline is a visual representation of your sales process. It shows every active deal, which stage it is in, and what needs to happen to move it forward.
Stage names vary by company. CaptivateIQ says that while naming conventions vary, most sales organizations use a version of eight stages. Avoma lays out nine, running from Prospecting to Post-Sale.
The count matters less than the definitions. A stage is only useful if two reps would place the same deal in the same column.
Entry and exit criteria, explained
Avoma describes the pair this way: each stage has entry criteria (what qualifies a deal to enter) and exit criteria (what signals it is ready to move forward).
Exit criteria are the ones that do the work. Without them, stage changes depend on mood, optimism or the end of the quarter.
Avoma also notes that when each stage is clearly defined with exit criteria, ownership and buyer alignment, you get cleaner sales forecasts and faster sales cycles.
Examples of exit criteria by stage
CaptivateIQ gives exit criteria for several stages. Two of them show the pattern well:
- Lead qualification: the lead meets the qualification criteria your revenue team has agreed on.
- Proposal: the proposal is reviewed by the decision makers.
Notice that the second one depends on a buyer action, not on the rep sending a document. Sending is an activity. Review by decision makers is evidence.
The same source names frameworks such as BANT (budget, authority, need, timeline) to make the qualification test concrete. Your discovery call questions are where much of that evidence gets gathered.
How to write exit criteria that hold up
- Name the buyer action. Write “economic buyer confirmed budget in writing”, not “rep felt good about budget”.
- Make it checkable. A manager should be able to verify it in the CRM, a call recording or an email.
- Assign an owner. Someone is responsible for confirming the criterion is met.
- Keep the list short. Two or three criteria per stage are easier to enforce than ten.
- Review quarterly. Compare stage definitions with deals you actually won and lost.
This fits well with consultative selling, where progress is tied to the buyer’s own problem and decisions.
Common mistakes with vague stages
Avoma warns about vague stages and stale deals. It also says too many teams treat the pipeline like a reporting tool: reps toss in opportunities, slap on a stage and hope it adds up to quota.
- Stages named after rep activities, such as “demo sent”.
- Deals that advance without any buyer confirmation.
- Stale deals left sitting in a stage with no next step.
- Different reps reading the same stage name differently.
The fix for each is the same: a written, observable exit rule that the whole team uses.
A worked example: one deal, three stage checks
Picture a mid-sized software deal. A rep has had a friendly first call and wants to log the opportunity as qualified. The agreed test is simple: does the lead meet the qualification criteria the revenue team has set?
If the team uses budget, authority, need and timeline as its test, the rep has to show evidence for each. A note that says “they seem keen” does not count. A note that says the buyer named a budget owner and a target go-live date does.
Later the rep sends a proposal. Under an activity-based rule, the stage would change the moment the email left the outbox. Under a buyer-based rule, the deal stays put until the decision makers have reviewed it.
- Check one: is there written or recorded evidence for each qualification point?
- Check two: has a decision maker, not only a champion, seen the proposal?
- Check three: is there a dated next step agreed with the buyer?
None of these checks takes long. What they do is stop a deal from drifting forward on hope.
Making the rules stick in your team
Written criteria only help if people use them. The most common failure is a beautiful stage document that nobody opens after the kickoff meeting.
Put the criteria where reps work. Many teams add them as required fields or short checklists in the CRM, so a deal cannot change stage until the evidence is logged.
Then use them in pipeline reviews. Ask “which exit criterion did this deal meet last week?” instead of “how does this one feel?” Over a few weeks the question changes how reps prepare and how they talk to buyers.
- Share the definitions with sales, marketing and customer success so handoffs mean the same thing to everyone.
- Let reps challenge a criterion that does not match how buyers really decide.
- Look at deals that skipped a criterion and ask whether the rule or the deal was wrong.
- Flag deals that sit in one stage with no buyer activity, since those are the stale ones.
What good criteria do for forecasting
A forecast is only as honest as the stages beneath it. If a stage can mean three different things, any weighting or probability you attach to it is guesswork.
Exit criteria turn each stage into a claim that can be checked. Managers can see which deals have earned their place and which are carried by optimism. That is the link behind Avoma’s point about cleaner forecasts.
It also helps coaching. When a deal stalls, you can name the exact buyer action that has not happened yet, and that gives the rep something specific to work on.
Matching stages to how your buyers actually buy
Borrowing a template is a fine starting point, but the stages should mirror the decisions your buyers make. A deal with a single approver needs fewer steps than one that passes through procurement, legal and security review.
Start by mapping the last handful of deals you won. Write down the moments when the buyer visibly committed to something: a meeting with a new stakeholder, a shared document, a signed mutual plan. Those moments are natural exit criteria.
Then look at lost deals. Where did they stop moving? If many died at the same point, that stage probably needs a stricter entry rule so weak deals are filtered out earlier.
- List the buyer commitments that preceded each past win.
- Group them into stages in the order they usually happen.
- Turn each commitment into a checkable exit criterion.
- Test the draft on open deals and adjust where reps disagree.
Handoffs and the stages after the close
Some teams stop the pipeline at closed-won. Others, as CaptivateIQ notes, extend it to include customer onboarding and deal expansion.
If you extend it, the same discipline applies. A handoff to customer success should have its own exit rule, such as a signed agreement and a named owner on the other side, so nothing gets lost between teams.
Clear criteria at the handoff also protect the buyer. They do not have to repeat their story to a new person who was never briefed.
Frequently asked questions
How many sales pipeline stages should I have?
There is no required number. Published models use eight or nine stages, but your count should match the real decision points in your buyer’s process.
What is the difference between entry and exit criteria?
Entry criteria qualify a deal to enter a stage. Exit criteria signal that it is ready to move forward.
Should exit criteria be based on rep activity?
Prefer buyer actions. Activities like sending a proposal show effort, while a decision maker reviewing it shows progress.
Can I move a deal backwards?
Yes. If evidence no longer supports the current stage, moving the deal back keeps your forecast honest.
