Define each stage by something the buyer has done, leaving out what the seller hopes will happen.
Name each open stage in your pipeline after something the buyer has done. Give each stage two exit criteria that a colleague could check without having been at the meeting. A deal then moves only when there is evidence, and the pipeline report shows where each deal actually stands.
CRMs come with default stages, and they are often named after the seller's work. Salesforce's defaults, for example, are Prospecting, Qualification, Needs Analysis, Value Proposition, Id. Decision Makers, Perception Analysis, Proposal/Price Quote and Negotiation/Review, ending at Closed Won or Closed Lost. Most of these names describe what the seller is doing rather than what the buyer has done. A rep can mark a deal "Negotiation/Review" after the buyer has stopped answering emails, and the CRM will still show it as 80 percent of the way there (an illustrative figure).
When stages are named after seller activities, the rep decides when a deal moves, based on how they feel about it. A buyer-action stage is named after something the buyer has done, so the deal moves only when there is evidence that the buyer did it. Your first step is to take the seller-activity names off your list.
"Proposal" is a weak stage name because it can mean almost anything. "Proposal seen" means the buyer has actually looked at the proposal, ideally in a meeting where you walked them through it. That way of presenting is taught in the course "Present, handle objections, and close", lesson 2.3, "Walk the buyer through it instead of emailing it".
Leave out a prospecting stage as well. A contact who has not agreed to anything is not yet a deal. Prospecting activity belongs in separate prospecting work, which the course "Prospecting: build a pipeline that does not run dry" covers.
Wei Ming is a sales manager. In lesson 4.1, "Why a written process makes everything else work", Wei Ming found that "proposal" could mean almost anything on the team. The team settled on four open stages, each named after a buyer action:
Meeting agreed: the buyer has accepted a first meeting. Problem confirmed: the buyer agrees there is a problem worth solving and roughly what it costs them. Proposal seen: the buyer has seen and discussed a written proposal. Terms agreed: the buyer has accepted price and terms, and only paperwork or approval remains.
Won and lost follow these four. They are outcomes, the end states a deal reaches, and a deal does not sit in them the way it sits in an open stage.
A stage name gives a rough position. Exit criteria are the specific things that must be true before a deal moves to the next stage, so a judgement call becomes a checklist. A good criterion can be checked by someone who was not in the room. "Buyer is interested" fails this test, because interest is a feeling. "A decision maker is named in the CRM" passes, and so do "a budget range has been discussed and noted" and "a decision date has been agreed with the buyer".
Wei Ming wrote two criteria for each stage. Two per stage stops most wrong moves and keeps CRM updates from becoming paperwork. To leave meeting agreed, the first meeting has happened and the buyer has described at least one problem in their own words. To leave problem confirmed, the person who will decide is named and a budget range has been discussed. To leave proposal seen, the buyer has given specific feedback on the proposal and a decision date has been agreed. To leave terms agreed, the contract is signed or the buyer has said no.
Raj, a rep on Wei Ming's team, had six deals marked "proposal". Under the new rules, four of the six had no proposal seen by the buyer, so they belonged at problem confirmed at best. Two of those four also had no named decision maker, so they could not move on until Raj found out who would decide. Raj's pipeline held the same number of deals Raj expected, and most of them were at an earlier stage than the CRM had shown.
Teams often want to add stages such as "demo done", "trial running" or "legal review", and each one seems sensible when it is suggested. A team that accepts them all can end up with nine stages. At that size nobody can name them all, and reps move deals in bulk at month end instead of updating them as things happen.
Aim for a number of stages your reps can recite without looking at a chart. For many teams that is four to six open stages. Extra detail within a stage goes in that stage's activities. Lesson 4.3, "Key activities and tools for each stage", shows how to set those up.
To check your model, ask a rep to name the stages from memory, along with the exit criteria for one of them. If the rep cannot, you have too many stages or the names are unclear. Reduce the stages or reword the names, then ask again.
Most of the work that moves a deal from meeting agreed to problem confirmed is discovery: asking the questions that uncover the real problem, what it costs and who decides. This course does not teach discovery skills. The course "Discovery: questions that find the real problem" teaches them, and its module 5 covers qualification frameworks including BANT and MEDDIC.
Your job as the manager is to set exit criteria that reflect what good discovery produces, so a deal cannot reach proposal seen without it. Wei Ming's criteria work this way. A deal stays at meeting agreed until the buyer has described a problem in their own words, and it stays at problem confirmed until a decision maker is named and a budget range has been discussed.
In the activity below you will write your own team's stages with two checkable exit criteria for each. Start from the stories you collected from your reps in lesson 4.1, and name each stage after what the buyer did at that point in their won deals.
Write down your team's stages, with two exit criteria for each that someone could check.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).