Sales pipeline stages are the fixed steps a deal moves through, from the first qualified conversation to signed or lost. The names matter less than the exit condition on each stage, the specific thing that has to be true before a deal moves forward. This guide gives a five-stage set with those conditions written out, and shows how to build it in FlyCRM, the WordPress CRM plugin from weDevs.
The names below are an example, not a standard. Copy them if they fit and rename them if they do not. If a CRM inside your own site is new to you, our guide to a self-hosted CRM for WordPress covers where the records sit.
Five sales pipeline stages, and the exit condition for each
A stage is a claim about where a deal stands, and the claim needs a test. Without one, the stage means whatever the person updating the board felt on Friday afternoon.
The test is the exit condition, one sentence saying what has to be true before a deal leaves. Two people looking at the same deal should reach the same answer. Deals sit on a kanban board in FlyCRM with a total under each column, and the conditions keep those totals honest.
Qualified
A deal reaches the board when you convert a lead into it, and Qualified means there is something real to sell.
It leaves Qualified when you can write one sentence naming what the customer is buying, roughly what it is worth and when they want it, and the customer has agreed that sentence is right. A sentence you wrote on your own is a hope.
Record the budget range and the name of anyone else who approves the spend. Deals that die in Negotiation usually died here, because a second decision maker appeared at the end who was never asked at the start. A FlyCRM deal holds several contacts with a role each, which is where that person goes.
Proposal
Proposal means the quote is with the customer. Not promised, not drafted. Sent.
It leaves Proposal when the customer responds to what is in it, with a question, an objection, a request to change the scope, or an acceptance. Silence is not an exit. Three weeks with no reply means the deal has stalled.
Every FlyCRM quote gets a public link, so the customer reads it without a WordPress login and accepts or declines there. The plugin records when a quote was opened, which separates a customer who has not read it from one who read it twice and went quiet. An acceptance is recorded against the deal.
Negotiation
Negotiation is where price, scope and terms get changed. Most teams treat it as the last stretch before the money arrives, which is why it collects deals that will never close.
It leaves Negotiation when nothing is left to agree and the person who signs has said yes or no. Not “they seem happy with it”, and not “legal is looking at it”. Those are tasks with dates on them, and FlyCRM keeps tasks, calls and notes on the deal’s own activity timeline.
Re-issuing a quote does not send a deal backwards. If the scope changed and you sent a new version, the deal stays here until somebody decides.
Won
Won is not a stage you work, it is the state a deal ends in, and its test should be the strictest one you have. Your win rate, average deal size and sales cycle are all counted from this column.
A deal is Won when you have a signed acceptance or money in the bank. A verbal “we are going ahead in January” is not Won, it is a deal in Negotiation with a January close date. Counting it as revenue is how a good quarter turns bad in the last week.
Lost
Lost needs a condition too, and the condition is that you know why.
A deal becomes Lost when the customer has chosen something else, has decided to do nothing, or has gone unreachable. Write the reason on the deal before you move it. Six months of those reasons is the most useful sales document a small team owns.
Keep the reasons to a short fixed list so they can be counted, something like price, timing, lost to a competitor or no decision. A tag on the deal does that in the free plugin, since tags work on every record type. A dedicated lost-reason field is a custom field, which is FlyCRM Pro.
Most losses in a small pipeline are not to a competitor. They are to nothing happening, and those two problems have different fixes.
How many stages you actually need
The right number is the smallest set where each stage changes what you do next. If two stages call for the same action from you this week, they are one stage wearing two hats.
Three working stages plus Won and Lost covers most teams under about ten people. A larger B2B team with a real handoff, where one person qualifies and another closes, earns a sixth, because a stage marking a handoff is doing actual work.
Eleven stages is a reporting structure built for a manager. Half its columns stay empty and the rest get backfilled before a meeting, so the forecast ends up reading a rep’s bookkeeping rather than the market.
If a step is something your team does rather than something the customer does, it is not a stage. Contract review and chasing a purchase order number are tasks on the deal, and the stage is still Negotiation until the customer decides.
When a deal skips a stage or goes backwards
Real deals do both, constantly. A referral arrives knowing what they want and lands straight in Proposal. A deal in Negotiation loses its champion to a new job and drops back to Qualified, because the conversation starts again with somebody who has never heard of you.
The instinct is to stop it with a rule, or a CRM that enforces the order. Resist it. A rep who cannot move a deal backwards will leave it where it is, and the board will look healthier than the business.
Let deals move both ways and write one line on the deal saying why. FlyCRM records stage changes, so the deal itself shows that it sat five weeks in Proposal, without anybody having to remember.
Then read the movement as information. One deal sliding back is a bad week. Three sliding back out of the same stage in a month is a pattern, usually a price change, a new competitor, or an exit condition loose enough to let deals through early. A stage most deals skip is not a stage, so merge it.
The quieter problem is the deal that does not move at all. Decide how many days of silence means dead in your business, taking the number from your own average sales cycle rather than from an article. FlyCRM shows that average on the dashboard, and anything past it gets a dated task or a move to Lost.
Win probability, and what the number is for
Each stage in FlyCRM carries a win probability, and every pipeline tool has some version of it. It is also the most misread number in a CRM.
It is a forecasting weight for a group of deals, not a prediction about one deal. Proposal at 40 percent does not mean this customer has a 40 percent chance of signing. It means about four in ten of the deals reaching Proposal were eventually won. Multiply each deal’s value by its stage probability, add the results, and you have a weighted pipeline figure.
Set those percentages from your own closed deals rather than a template. Once you have thirty or so finished deals, count how many of the ones that reached each stage ended in Won. Until then use round numbers and say out loud that they are guesses, because a precise-looking number gets trusted long past the point it has earned.
Recheck them twice a year. A stage set at 60 percent that now closes at 25 overstates every forecast you produce. And do not let people adjust the probability deal by deal to signal confidence, because once it is a mood dial it stops adding up to anything.
The free FlyCRM dashboard shows pipeline value by stage, win rate, average deal size and the open, won and lost deal totals, each against the period before. Sales reports covering the pipeline funnel, sales velocity and how long deals sit in each stage are in FlyCRM Pro, priced on the pricing page as of September 2026.
Where deals enter the pipeline
A pipeline is only as honest as the gate in front of it. If every inquiry becomes a deal, the board is a to-do list with money written on it.
FlyCRM keeps leads in their own list, apart from contacts, with statuses running from New through Contacted, Working, Nurture and Qualified to Unqualified or Converted. Work an inquiry there while you find out whether it is real.
Each lead carries a score you set by hand. It is not a rule engine adding points for email opens, which is what nearly every article on this subject describes.
For a small team that is the better arrangement, not a gap to apologize for. A rule engine dresses up an afternoon of guesses as arithmetic, while a score a person sets is a judgment you can explain. It needs an agreed meaning, so settle what the bands stand for before anybody scores a lead.
FlyCRM also records where a lead came from, holding the campaign, medium, source, content and term it arrived on plus the referring page, and each connected form sets its own source. That answers the question that matters, which is not which source sends the most leads but which sends the ones that reach Proposal.
When a lead is ready, one step turns it into a contact, a company and a deal, with its history. That is where the Qualified condition has to be met rather than assumed.
Setting your stages up in WordPress
Ten minutes with a pen, then five in the plugin.
- Write your stage list on paper first, with the exit condition beside each one, in the words your team already uses. Doing it afterwards is how teams inherit somebody else’s process.
- Install FlyCRM from the WordPress plugin directory and activate it. It needs WordPress 6.5 and PHP 7.4 or newer, and sets itself up with a working pipeline, statuses and a starter tag list.
- Open the pipeline settings and edit the stages to match your list. Rename what is close, remove what you do not sell through, and drag the rest into the order deals really move in.
- Set a win probability on each stage. Round numbers are fine while you have no closed deals to count, as long as everyone knows they are placeholders.
- If you sell in two genuinely different ways, build a second pipeline rather than stretching one over both. FlyCRM allows as many as you need, each with its own stages.
- Put five real deals from last month on the board, including one you lost. Any deal you cannot place is telling you an exit condition is wrong.
- Write the exit conditions somewhere the team can read them without asking. A condition that lives in one person’s head is not a condition.
Reviewing the pipeline once a week
Twenty minutes, the same slot every week, in this order.
- What moved. Stage changes are recorded on each deal, so you read what happened rather than asking people to remember.
- What did not move. Any deal sitting in one stage longer than your average sales cycle needs a decision this week.
- Close dates that have passed. Either the date was wrong or the deal is over.
- What came in. New deal count is the least interesting number in the review and the one that decides next quarter.
- Totals last. Pipeline value by stage, and the weighted figure against your target rather than the raw total.
By the time you reach the totals you already know why they look the way they do, and anything that came out of the review should end up as a dated task on the deal it belongs to.
What to do next
Take last month’s deals and lay them out against your stage list. The ones you cannot place are the argument worth having with your team this week.
FlyCRM is free on WordPress.org, with no cap on records, and a working pipeline is on screen the moment you activate it, so the first job is editing stages rather than building them.
Frequently asked questions
How many stages should a sales pipeline have?
Most small teams need three working stages plus Won and Lost. The right number is the smallest set where each stage changes what you do next. If two stages call for the same action, merge them. A stage nobody updates still counts in the forecast.
What is the difference between a pipeline and a funnel?
A funnel describes volume, meaning how many people move from awareness to purchase across your whole market. A pipeline describes named deals you are working right now, each with a value and an owner. A funnel is a shape you measure. A pipeline is a list you act on.
Can I run more than one pipeline?
Yes. FlyCRM lets you build as many sales pipelines as you need, each with its own stages and win probabilities. Run a separate one when the steps genuinely differ, such as new business against renewals, rather than adding stages only half your deals ever use.
What is a good win rate?
There is no universal number, and any article that gives you one is guessing about your market. The useful comparison is your own win rate three months ago, measured the same way. Fix the definition first, counting deals that reached your first working stage, then watch which direction it moves.
Should I delete lost deals?
No. A lost deal records why somebody said no, and it is the only place that information lives. Move it to Lost, write the reason on the deal, and keep it. In FlyCRM a deleted record goes to trash and can be restored, but the better habit is not deleting it.
Write the Exit Condition Before You Name the Stage
A stage is only real if you can say what has to be true for a deal to leave it.
Five stages cover most businesses, and naming them after what the buyer has done rather than what you hope they will do is what keeps the board honest. Deals will skip stages and go backwards, and that is information about your process rather than a fault in it. Win probability is there to forecast the whole board, not to judge a single deal.
Lay last month’s deals against your stage list this week. The ones you cannot place are the conversation worth having with your team. FlyCRM puts a working pipeline on screen the moment you activate it, so the job in front of you is editing stages rather than building them.
