How Long Should a Deal Stay in Each Sales Stage?

How Long Should a Deal Stay in Each Sales Stage?

Build the answer from your own closed-won history, because stage definitions and deal sizes make cross-company comparisons meaningless. A company whose Stage 3 requires a signed mutual action plan will report longer time in Stage 3 than a company whose Stage 3 means a demo happened. Neither is slow.

The range across real pipelines is wider than most teams assume. When ORM groups opportunities and models a close curve for each group, those curves run from 1 week to 80 weeks. Most of the closing expectation lands before week 12, and very few groups carry meaningful expectation past 52 weeks. That spread exists inside single companies, across segments and deal types, which is why one company-wide time-in-stage target produces bad decisions in both directions.

The practical standard is a median plus a tolerance band. Median time in stage from closed-won deals defines normal. Multiples of that median define when to look and when to act.

How Do You Build the Standard From Your Own Data?

Take every deal that closed won in the last four quarters, measure entry-to-exit days per stage, and report the median by segment. Four steps make it defensible.

Start with closed-won only for the baseline. Including losses mixes in deals that sat in a stage until someone gave up, which inflates the standard and makes every live deal look healthy. Keep the loss data separate as a comparison set, since the difference between winner and loser time in stage is the diagnostic.

Cut by segment before you cut by anything else. Enterprise deals move slower through every stage, so a blended median understates enterprise and overstates SMB, and reps in both segments learn to ignore it.

Use the median rather than the mean. A few deals that sat for a year drag the mean upward, which sets escalation triggers so loose that nothing trips them.

Then add the 75th percentile. The median tells you what normal looks like. The 75th percentile tells you the tail you are willing to accept before a manager gets involved.

What Should the Escalation Triggers Be?

Set them as multiples of your own median rather than as fixed day counts, so the rule scales across segments without maintenance. The last row carries most of the value. Time by itself is a weak predictor, since a large deal can legitimately sit in procurement for months. Time combined with silence is a strong one.

When Is a Deal Actually Stalled?

When nothing has changed in stage, close date, or amount. That is the definition of meaningful activity worth enforcing, and it deliberately excludes notes, logged calls, and emails, because those accumulate while a deal goes nowhere.

Applied across ORM customers, at least 10% of open pipeline has gone untouched by that definition for twelve months. It still shows up in coverage reports, still appears in board slides, and still makes the quarter look better funded than it is.

Most customers run a twelve month rule as a result. An opportunity that reaches a year without a stage, date, or amount change gets closed, and if the account is genuinely still in play a new opportunity is created with a real date. The rule is unpopular for a week and then nobody misses the records.

The same silence predicts slippage earlier than any stage report. The strongest explicit signal is a rep changing a close date, and a deal that slips across a quarter boundary is less likely to close even when it sits in commit. The earliest signal comes before that, when a buyer stops returning email and the record stops changing. Treating deal slippage as a data pattern rather than a rep confession is what moves the detection forward by weeks.

Why Are Deals Sitting Longer Than They Used To?

Because conditions changed, and stage duration absorbs the change before win rate does. Four causes account for most of it.

Market uncertainty produces fewer decisions, which stretches the time from qualified to closed without changing the eventual outcome for many deals. A new competitor creates pricing pressure, which adds a negotiation round and pulls average deal size down. Buyers under cost pressure add approval layers, so procurement and legal absorb weeks that used to be days. Territory changes distract reps, and execution slips even while coverage ratios hold.

Seasonality accounts for a predictable share of the rest. Q2 and Q4 usually run stronger than Q1 and Q3, and the third month of a quarter runs stronger than the first two. A deal that entered negotiation in the first week of a quarter is competing for attention against every deal dated for the quarter end. Measuring time in stage without adjusting for where in the quarter the deal entered produces noise that looks like a trend.

How Does Time in Stage Feed the Forecast?

Convert it into a close curve per group instead of a single probability per stage. Stage probability answers whether a deal will close. Time in stage answers when, and a forecast needs both.

At ORM each opportunity is grouped by a machine learning model, and each group gets a predicted curve for how long it takes to close. That is a different object from a stage weight. Two deals sitting in the same stage with the same amount can belong to groups whose curves peak eight weeks apart, and a forecast that treats them identically will place revenue in the wrong quarter even when it calls the outcome correctly.

The practical version for a team without modeling infrastructure is to segment the pipeline by the characteristics that actually drive timing, usually deal size band, segment, and lead source, then apply historical timing distributions per group. It is cruder, and it still beats one blended rule. Pair it with a sales velocity read so you can see whether the cycle is lengthening while you plan, and rebuild the timing inputs every quarter as part of your forecasting process.