Sales Pipeline

Before anyone drew a pipeline, the deals a company had in flight lived in the head of whoever was working them. Sofia could be talking to twenty companies and be the only person who knew the state of any of them. When Grace asked how much was coming in this month, the answer was always some version of the same sentence: a few of them look promising.

Three things happen at once in that setup. First, nobody sees the whole picture, so nobody can plan against it. Second, deals die quietly. Far more of them end because a reply got missed and three weeks went by than because a customer said no. Third, when a rep leaves, the deals leave with them. A handover document can list twenty company names and still not say which one was a signature away.

Pagedeck went through this. In the last week of one quarter Sofia said three deals would probably land and one did. The next month she carefully said two and four landed. She wasn't lying and she wasn't bad at her job. There was simply no way to lay twenty deals side by side and count them against the same ruler.

A sales pipeline is the device that takes those deals out of someone's head and puts them on a board. Two things get decided before anything goes on it: the list of stages a deal passes through, and what has to be confirmed before a deal moves from one stage to the next. A team might write the move from demo to proposal as "we know who signs on the customer side and the next meeting is on the calendar." Without conditions in writing, the rep's mood sets the stage, and a total assembled from moods predicts nothing.

Inside the sales field, this is the branch that makes the deals you currently hold visible. The work of building a list of companies worth calling and opening the first conversation fills the front of the pipeline, and the step of checking whether the person on the other end can actually buy filters out what shouldn't be there. The pipeline is where whatever survives that filter sits until it reaches a signature.

The seventeen terms below fall into six branches: what counts as one item (units), where that item travels (stages), how it ends (outcomes), what numbers read the pile (metrics), what the team does with those numbers (operations), and what distorts them (risk).

Units decide what a single row on the board is. The three overlap on the surface and point at different things. A deal is the plain word for one piece of business with one customer. An opportunity is a deal that has been confirmed to have budget behind it and someone willing to spend it, which is why a business card collected at a conference is not one yet and a company with money allocated and a decision date is. A deal stage is the value that says how far along a given deal is, and it's the field that creates the columns everyone looks at.

Stages cut the path a deal travels into ordered positions. Names vary by company, but the shape is usually five.

1. Discovery call — the first conversation, spent working out the customer's situation and problem. It's a session for asking rather than presenting, so the time goes into what they do today and why that hurts.
2. Demo — showing the product and walking through it. Pointing at how the specific pain from the discovery call disappears on screen moves deals forward more often than touring every feature front to back.
3. Proposal — the stage where scope, price, and term go into a document and get sent. Four things usually appear in it: what is delivered and how far it goes, what it costs, what dates it covers, and when and how the money changes hands. The document travels to people who never joined a meeting, so it has to hold up without anyone talking over it.
4. Negotiation — working price and terms into something both sides accept. What moves here isn't only the number: payment timing, contract length, termination terms, and security requirements all come up. On Pagedeck's enterprise deals the last two eat more time than the price does, and they're where the customer's legal reviewer and security lead first appear.
5. Closing — the final stretch that ends in a signed contract. Most of it is confirming signature process and approval chains, which is why a deal that is verbally finished can sit here for weeks.

Outcomes are the two directions a deal closes in. Closed won is a signed contract and closed lost is one that fell through. Losses get a name of their own so they can be counted later. Tag the reason as price, timing, a competing product, or a change of contact, and at the end of a quarter it becomes visible that twelve of the thirty deals lost went the same way. Leave them open instead and those twelve stay on the board inflating the total.

Metrics are the four numbers that read the pile. Sales cycle is the time from first contact to signature; Pagedeck's team-plan deals average 34 days while enterprise deals run past 90, and averaging the two together produces a number neither side can plan with. Pipeline coverage is how many times over the target the pipeline is stacked, found by dividing the total value in play by the goal. A quarter with a 200 million won target and 600 million won in open deals is covered three times over. Three is the common benchmark because win rates tend to land near 30 percent. Win rate is the share of worked deals that close — six wins out of twenty worked is 30 percent. Average deal size is the money one deal is worth on average, and it's what converts a revenue target into a count of deals. A 200 million won target at 10 million won per deal means twenty signatures, and at a 30 percent win rate that means more than sixty deals have to be on the board.

Operations turn those numbers into a plan. Sales forecast is the estimate of next period's revenue built from the pipeline, most often by multiplying each stage by its historical close rate and adding the results: ten deals worth 20 million won each sitting in a proposal stage that closes 40 percent of the time come to 80 million won. Quota is the revenue target assigned to one individual. If the forecast is what the company expects, quota is what a person is accountable for, and putting them next to each other shows where the gap sits.

Risk covers behavior that distorts the arithmetic. Sandbagging is setting a target low, or holding a deal that could close now until the next period. A rep who has already cleared quota and pushes a contract that arrived in the final week into the first week of next month comes out ahead personally while the company's forecast misses twice in a row. This branch stands on its own because a pipeline isn't a record that fills itself — people type it in, and whatever those people are paid on shapes what gets typed. The fix is rarely a rule against the behavior; it's usually a change to how quota and commission are structured so that holding a deal back stops paying.

The six branches form one line. Units decide what gets counted, stages move it, outcomes close it, metrics read the closed record, and operations plan the next quarter from those values. Risk sits outside that loop and edits its assumptions. That's why the first job for a team adopting a pipeline isn't picking a tool but writing one sentence per stage describing what has to be true to move into it. Only after those conditions exist do the other sixteen terms mean the same thing to two people.

"Coverage is at 1.8. What's on the board won't get us to the quarterly target."

Open deal value is short against the goal, so either prospecting steps up or the target gets reset. Said in the middle of a quarter there's still time; said in the last month it's a postmortem.

"This one reached proposal and hasn't moved in three weeks. Let's close it lost and reopen it next quarter."

Take a dead deal off the board. Left there, it swells the total and pushes the forecast up with it, and reopening it later costs nothing because the record of what happened stays.

"Two of the five in negotiation won't close this month. Leave them out of the forecast."

Only deals that can actually close belong in the number — a late stage alone doesn't make it this month's revenue, and a rep saying so in advance is worth more than a rosy number that collapses on the last day.

Pagedeck runs a 30-minute pipeline review on Monday mornings. Sofia reads out the count and value by stage, and Grace checks coverage against the quarterly target. The board flags anything that has sat in the same stage for more than two weeks, and that list gets read first. Last week seven deals were flagged and four of them had gone silent after the contact on the customer side changed jobs. Two were closed on the spot and two were assigned a fresh outreach to the replacement, who in one case turned out to have inherited the same problem and asked for a demo the following week.

Ravi comes to the same meeting with the won and lost records, because he wants to know which channel produced the deals that went the distance. Last quarter, deals that started from conference badges closed one time in fourteen, while deals introduced by existing customers closed four out of six. Next quarter's budget was set on those two numbers.

The most common mistake is treating a thick pipeline as a reason to relax. The total towers over the target, so it looks safe. In practice the close rate differs by stage and decays with age: 100 million won sitting in discovery is not the same 100 million won sitting in negotiation, and a deal that hasn't moved in two months is usually already over. Ignore that and a quarter ends at 4x coverage and under target. Pagedeck did exactly that last year, opening a quarter at 4.2x and finishing at 70 percent of goal, because half the total was parked in discovery where roughly one deal in ten survives. Reading stage-level value alongside the last-activity date makes the illusion disappear.

The other one is reading stage as a measure of how much work the team has done — proposal sent, so move it to proposal; demo delivered, so move it to demo. Our own activity is the part we can see, so it becomes the thing we grade. A stage describes how far the customer has moved, not how far we have. Sending a proposal isn't enough; the customer has to confirm they received it and are reviewing. Miss that distinction and ten unread proposals pile up in the proposal stage, and the forecast built by multiplying that stage by its historical 40 percent close rate misses every month.