Most B2B businesses have pipeline stages. Almost none of them have useful pipeline stages.
The stages exist (usually something like Prospect, Qualified, Proposal, Negotiation, Closed), but they’re defined so loosely that different reps (or the same rep on different days) use them differently. One person’s “Qualified” is another person’s “Closed Won” waiting to happen. One person’s “Proposal” is a deal where a document was sent six months ago and nobody’s heard back.
This is the root cause of most forecasting problems. If the stages don’t mean anything consistent, the pipeline data doesn’t mean anything consistent. And if the pipeline data doesn’t mean anything, the forecast is fiction.
The fix is exit criteria. Not stage names; every CRM has those. Exit criteria: the specific, objective conditions that must be true before a deal moves to the next stage. Here’s how to build them properly.
This post sits alongside What RevOps Actually Means: pipeline architecture is one of the core RevOps disciplines, and it underpins everything from forecasting to rep coaching. It also connects to How to Qualify Beyond BANT, which covers the qualification layer in more depth.
Why stage names aren’t enough
A pipeline stage without exit criteria is just a label. It tells you what the deal is called, not what’s actually happened in it. And when stage progression is based on the rep’s optimism rather than objective conditions, your pipeline becomes unreliable in a specific way: it’s always fuller and more advanced than the deals actually are.
This matters most at the forecast end. If your “Proposal” stage includes deals where a proposal has been sent and deals where a proposal has been sent and the prospect has confirmed they’re evaluating you seriously, those two situations have very different close probabilities. They shouldn’t be in the same stage.
Exit criteria fix this by making stage progression a question of fact, not feeling. Has this happened? Yes or no. Has the prospect confirmed budget authority? Yes or no. Have we had a conversation about timeline? Yes or no.
A working template for B2B pipeline stages
The right stage architecture depends on your deal type, deal size, and sales cycle. What follows is a template that works well for B2B businesses with deal cycles of one to six months and deal values of roughly £20k–£250k. Adjust for your context.
Stage 1: Prospecting
What it means: An account has been identified as a potential fit. There may have been initial outreach or a first touchpoint, but no substantive conversation has occurred.
Exit criteria to move to Stage 2:
- A two-way conversation has taken place (not just an email reply)
- The prospect has agreed to a discovery conversation
- Basic qualification has confirmed the company is a potential fit (size, sector, buying stage)
Do not let deals sit in Prospecting indefinitely. If there’s been no two-way conversation after a defined number of attempts, move the deal to a separate “On Hold” or archive it. Dead prospects inflating the top of the funnel make the whole pipeline meaningless.
Stage 2: Discovery
What it means: A discovery conversation has taken place or is scheduled. The goal of this stage is to understand the prospect’s situation, identify whether there’s a genuine problem worth solving, and qualify whether the deal is worth pursuing.
Exit criteria to move to Stage 3:
- Discovery call(s) completed
- Problem or use case confirmed and documented in the CRM
- Buying process understood: who’s involved, how decisions get made
- Budget existence confirmed (not amount, but whether budget exists or can be created)
- A defined next step agreed with the prospect
Deals that don’t have a confirmed next step should not advance. “I’ll think about it and be in touch” is not a next step.
Stage 3: Qualified
What it means: The deal meets your core qualification criteria. You have reason to believe this could close and that the effort of a proposal or presentation is justified.
Exit criteria to move to Stage 4:
This is where frameworks like MEDDIC, BANT, or your own qualification model come in. At minimum, a deal should not move to Qualified without:
- Budget confirmed or a credible path to budget creation
- Decision-maker identified and engaged (not just a sponsor)
- Timeline confirmed, not “sometime this year” but a realistic window
- Your solution mapped to their specific problem
- Competitive situation understood
If you’re not sure which qualification framework suits your deal type, How to Qualify Beyond BANT covers the options in detail.
Stage 4: Proposal / Presentation
What it means: A formal proposal or commercial offer has been presented. The prospect is actively evaluating.
Exit criteria to move to Stage 5:
- Proposal submitted and confirmed received
- Presentation or walkthrough completed with decision-maker(s)
- Next steps agreed, specifically: a date for a response or a decision meeting
- No outstanding questions that would block a decision
A proposal without an agreed next step is a hope, not a pipeline entry. If the prospect won’t commit to a response date, that tells you something worth knowing before the deal enters the forecast.
Stage 5: Negotiation / Closing
What it means: The prospect has indicated they want to proceed. Commercial terms, legal review, or final approvals are the remaining steps.
Exit criteria to move to Closed Won:
- Verbal or written agreement on commercial terms
- Legal/procurement process engaged and timelines confirmed
- Contract or order form in circulation
- A confirmed close date based on the prospect’s timeline, not yours
Note: deals should only enter this stage if the prospect has genuinely indicated intent. “We’re interested but want to do some more evaluation” is Stage 3 or 4, not Stage 5.
Getting the team to use it
Exit criteria only work if the team uses them consistently. A few things that help.
Make them visible. Put the exit criteria in the CRM itself, as a field description, a checklist, or a note on each stage. Don’t rely on everyone remembering a document they read once in onboarding.
Enforce them in pipeline reviews. When a rep wants to advance a deal, ask the exit criteria questions directly. Has budget been confirmed? Has a decision-maker been engaged? What’s the agreed next step? If the answers aren’t there, the deal doesn’t advance.
Make it about accuracy, not criticism. Reps who feel that accurate pipeline is punished will fudge it. The goal is that the pipeline tells you what’s actually happening, not that it looks impressive. If deals are being moved back, that’s the process working.
What this does for forecasting
Once exit criteria are being applied consistently, the pipeline becomes a different kind of instrument.
Deals in Stage 5 (the negotiation/closing stage) are genuinely in negotiation. Not just ones the rep feels good about. Stage 4 deals have a submitted proposal and a confirmed next step. Stage 3 deals are genuinely qualified, not just interesting conversations.
The result is a pipeline where stage position correlates with close probability in a meaningful way. That’s what makes forecasting possible. You can look at the pipeline and say: we have £X in Stage 5, at our historical Stage 5 close rate of Y%, that’s a reasonable expectation of £Z in the next 60 days.
That’s not perfect (close rates vary, timing slips), but it’s a forecast built on something real. That’s a materially different starting point than the alternative.
If you want to understand how pipeline architecture fits into the broader revenue infrastructure, What RevOps Actually Means covers the full picture. And if your process is being built from scratch, How to Build a Sales Process From Scratch walks through the sequence.