# Sales cycle length: how to measure it in Pipedrive and what actually shortens it https://salesbond.app/blog/sales-cycle-length-pipedrive/ Published: 2026-10-01 · Updated: 2026-10-26 Sales cycle length is the number of days from the day a deal is created to the day it is marked won. Count it on won deals only, take the median rather than the mean, and split it by deal size before you quote a single figure. Two or three large, slow deals will pull an average a long way from what a typical deal in your pipeline actually does. In Pipedrive the two dates you need already sit on every deal: the date it was created and its Won time. This post is about that one number: how to get it out of Pipedrive without fooling yourself, and what a small team can do about it. If you want to plug it into the full formula with win rate and deal value, that is covered in [sales velocity in Pipedrive](https://salesbond.app/blog/sales-velocity-pipedrive/). Here we stay on the clock. One position up front. Faster is not the goal. Predictable is. A cycle you can state as “small deals close in about four weeks, large ones in about four months” lets you forecast, staff and say no. A cycle you have shaved by a week but still can’t predict does not. On HubSpot the measurement is the same, won deals only, and we show it in [sales cycle length in HubSpot](https://salesbond.app/blog/sales-cycle-length-hubspot/). ## The average is lying to you Take a five-person team that won 20 deals last quarter. All numbers from here on are made up to be consistent, not pulled from any benchmark. Sixteen of those deals were between $2,000 and $6,000 and closed in 18 to 35 days. The other four were between $30,000 and $45,000 and took 90 to 140 days. The small deals average 26 days, the large ones 115. Add them all up and divide by 20 and the “average sales cycle” is 44 days. Not one of the twenty deals took 44 days. The fastest large deal was twice that. The slowest small deal was nine days short of it. The median tells a truer story. Sort the 20 cycles and the middle value lands among the small deals, at about 27 days. That is what a typical deal in this pipeline does. But 27 is also wrong for the four large deals, which is why the next cut is by deal size: small band median 27, large band median 115. Now the number means something. A $4,000 deal at day 40 is late. A $40,000 deal at day 40 is early. The blended 44 would have told you both were about on time. ![Hypothetical example: 20 won deals on a 0–150 day timeline, mean of 44 days versus median of 27 days](https://salesbond.app/wp-content/uploads/2026/10/sales-cycle-length-pipedrive-2.png) This is why one average is a trap even for a team this small. The mix of deal sizes changes from quarter to quarter, and the average moves with the mix not with how the team is selling. One extra enterprise deal and the cycle “got longer”. Two fewer and it “improved”. Nobody did anything differently. So what is a good sales cycle length? Published benchmarks won’t settle it. They disagree widely, and most of the gap comes from deal size and the segment being sold to, so a figure from someone else’s pipeline tells you little about yours. The reference point that holds up is your own: the median of each deal-size band over the last few quarters. Shorter than that is good news. Longer is a question worth asking. ## How to calculate sales cycle length in Pipedrive You don’t need a report builder for this. The Deals list view and a spreadsheet are enough, and the spreadsheet is where you’ll do the cuts anyway. 1. Open the Deals list view and filter for won deals only, since a lost deal runs a different clock: Status is Won, Won time within the last two full quarters. If that leaves you with only a handful of deals, widen it to 12 months. A median of a few deals is an anecdote. 2. Choose the columns the list shows: deal title, Value, Won time, the deal’s created date, Owner, and whatever field you use to record where the deal came from. Most small teams keep source in a custom field; if you don’t record it yet, skip the source cut this time rather than guessing. 3. Export the filtered list. You get a spreadsheet file. 4. In the sheet, add a column: Won time minus created date, in days. That column is your sales cycle, one row per deal. 5. Take the median of that column. Write it down with the definition next to it: “days from deal created to Won time, won deals only, last two quarters, median”. The definition is the part that stops the next argument. 6. Now cut by value. Sort by Value and look for the natural break; in the example above it sits around $10,000. Two bands are usually enough for a five-person team, three at most. Take the median per band. Any band with only a few deals gets a note saying so, because its median will swing wildly next quarter. 7. Cut by source the same way. Inbound and outbound deals of the same size can run different clocks, and that difference is useful for forecasting even when you can’t change it. Two things about the dates themselves. Pipedrive sets Won time when the deal is marked won, and only users with deals admin permissions can edit it afterwards, from the deal’s detail view or the list view ( [Pipedrive Knowledge Base](https://support.pipedrive.com/en/article/how-can-i-edit-the-deal-won-time)). So the won end of the clock is reliable. The created end is whatever your team does. If reps create the deal after the first call, your cycle starts late and reads shorter than the buyer experienced; if they create it the moment a lead looks warm, it reads longer. Neither is wrong. Pick one convention, put it in the definition, and don’t compare this quarter’s number to a quarter where the habit was different. If you’d rather stay inside Pipedrive, Insights has a deal duration report that shows average time in stages for the deals you filter. It gives you a mean, not a median, and it doesn’t count time spent in stages you have since deleted ( [Pipedrive Knowledge Base](https://support.pipedrive.com/en/article/insights-reports-deal-duration)). Fine for a glance. For the bands and the won-versus-lost comparison below, export. Deal duration is one of a handful of Insights reports worth saving; the [five we would build for pipeline health](https://salesbond.app/blog/pipedrive-reports-pipeline-health/) include it next to slipped close dates and coverage. ## Won versus lost: when an old deal is already past the point of no return Lost deals do not belong in your cycle length. A rejection is a different clock, and mixing it in produces a number that describes neither the sales you make nor the ones you don’t. But the lost deals are worth a separate export, because they answer a question the won deals can’t: how long does this team hold onto a deal that was never going to close? Run the same export with the filter set to lost deals, and compute days from created to the date each was marked lost. Then put the two lists side by side, band by band. Back to the hypothetical team. In the small band, the longest won deal took 35 days. The lost deals in the same band were marked lost at a median of 60 days. Read that plainly: by the time a small deal passed day 35, no deal of that size had ever gone on to close, and the reps carried the losers for another month before admitting it. That month is where the forecast was wrong and where follow-up time went to deals that could not pay for it. So the longest won cycle in a band becomes a line. An open deal older than that line is not a slow deal. It is a deal nobody has had the conversation with yet. It can still close; the point is that it needs a different conversation than “let’s follow up next week”. Either the buyer names a date and a reason the deal is still alive, or it gets marked lost with a reason you can learn from. [Stalled deals and dead deals](https://salesbond.app/blog/stalled-deals-vs-dead-deals/) look the same in the pipeline view and need different handling; a [close date that has been pushed more than once](https://salesbond.app/blog/pushed-close-dates-pipedrive/) is usually the first sign you’ve crossed the line. Check the line each quarter. If the team gets better at qualifying, the longest won cycle in a band should come down, and the line moves with it. ## Three ways a small team can shorten the sales cycle Most advice on shortening the cycle comes as a long list. A team of five can hold three things in its head, and these three are the ones we’d pick, because each one is visible in Pipedrive the following Monday. ### Kill the stuck deals earlier The line from the last section is the tool. Save a filter in the Deals list view: open deals, created before a date that is today minus the longest won cycle for the band. Review it weekly, in ten minutes, and each deal on it gets one of two outcomes: a dated reason it is still alive, or Lost with a reason. This does not shorten the clock on the deals you win. It shortens the real cycle anyway, because the hours that that were going into dead follow-ups go into deals that can close this quarter, and the forecast stops carrying ghosts. ### A dated next step on every deal Deals get long in the gaps, not in the meetings. A deal with no scheduled activity is a deal waiting for somebody to remember it, and on a team where everyone sells and somebody also does support, that can be a long wait. The rule is simple to state and hard to keep: no open deal without a next activity that has a date and a verb. “Follow up” is not a verb anyone can act on; “send revised quote by Thursday” is. At the start of a deal, this is the same discipline as [getting a first touch to new leads the same day](https://salesbond.app/blog/speed-to-lead-pipedrive/). In the middle, it is what keeps a four-week deal from becoming a seven-week one for no reason the buyer caused. ### Agree the buyer’s timeline in discovery The third lever is the only one that changes what happens on the buyer’s side. Ask, in the first real conversation, when they need this working and what has to happen between now and then: who signs, who reviews security, whether there’s a budget cycle. Write the answer on the deal. A [one-page mutual action plan](https://salesbond.app/blog/mutual-action-plan-small-teams/) is the formal version, but even two dated lines in the deal notes change the shape of the sale. Deals with an agreed timeline don’t drift, because drift is now visible to both sides. Deals without one drift by default, and the rep finds out about the procurement step at week nine. Pick one lever first. Run it for a quarter, re-export, and compare the band medians and the won-versus-lost line to last time. If nothing moved, the lever was wrong for your pipeline, and you’ve learned that for the cost of one spreadsheet. ## When a longer cycle is fine A long cycle is a problem when it is unexplained. It is not a problem when it is the price of the deal. If your large band takes four months because four people have to agree and one of them is legal, that cycle is a fact about your customer, and cutting it means either selling to smaller companies or skipping steps the buyer won’t skip. Trying to force it down usually shows up somewhere else: a [win rate](https://salesbond.app/blog/sales-win-rate-pipedrive/) that slides because reps pushed for a decision before the buyer was ready, or discounts offered to “close by Friday” on deals that would have closed at full price in March. The same is true of a cycle that lengthens because you moved upmarket on purpose. Larger deals, longer clock. The measure of whether that was a good trade is revenue per month of selling effort, not the cycle on its own, and that is a question for teh velocity formula, not this one. What you want from the cycle is that each band’s median holds steady and the spread around it stays narrow. That is what makes a [weighted forecast](https://salesbond.app/blog/weighted-pipeline-forecast/) believable: a deal at day 20 in the small band has a known chance of closing, because you know what day-20 deals in that band have done before. A wide, unpredictable cycle breaks that even when the median looks good. So the quarterly ritual is short. Export won deals, take the median per band, export lost deals, find the line and look at whether the line moved. Then pick the one lever the numbers point at. The team that does this for four quarters knows how long its sales take. Most teams only know how long they wish they took.