Which Sales Metrics to Track When the Sales Cycle Runs 9 to 18 Months

Why do standard sales metrics fail on long cycles?

Because every outcome metric reports on decisions made a year ago. Win rate this quarter is a verdict on the opportunities created four quarters back, on the qualification bar in force then, under the pricing and competitive conditions of that period.

That gap makes outcome metrics useless as steering instruments. By the time win rate drops, the cohort that produced the drop is closed and the cohort behind it is already halfway through its own cycle carrying the same defect. A team on a three-month cycle can correct within a quarter. A team on a fifteen-month cycle cannot, so the metric arrives as history rather than as information.

Long-cycle organizations need a second layer of measurement that updates weekly and describes deals in flight. Outcome metrics still belong on the board deck. They do not belong in the operating cadence.

What should you measure instead?

Movement, aging against expectation, and the presence or absence of activity. Those three update every week regardless of how long a deal takes to close. Cohort conversion is the one most teams skip and the one that pays. Group opportunities by the quarter they were created, then track what share has closed by month 3, month 6, month 9, and month 12. A cohort converting slower than its predecessors at the same age tells you something changed in sourcing or qualification, and it tells you a full year before the win rate does.

How long should a deal sit in one stage?

Compare it to the expected close curve for its own group rather than to a single company-wide number. A $2M platform deal and a $90K departmental purchase do not share a timeline, so one aging threshold flags the wrong deals in both directions.

ORM groups each opportunity with a machine learning model and predicts a close curve per group. Those curves run from 1 to 80 weeks, with most of the expectation landing before week 12 and very few groups extending past 52 weeks. The practical consequence is that "long cycle" describes an average, not a rule. Even in enterprise portfolios, most groups resolve well inside a year, and the deals that genuinely need 52 weeks are a small minority that gets used to excuse everything else.

Grade each deal against its group curve. A deal at week 30 in a group whose curve peaks at week 14 is behind. A deal at week 30 in a group whose curve peaks at week 40 is on schedule. The same absolute age produces opposite verdicts, which is why fixed aging thresholds generate noise.

When is a long-cycle deal actually dead?

When nothing has changed on it for 12 months. ORM applies a 12-month rule for most customers, and the test is specific. Meaningful activity means a change in stage, close date, or amount. Notes, logged calls, and calendar entries do not count, because those are things the seller does rather than things the buyer causes.

In ORM customer data more than 10% of pipeline sits untouched for 12 months. On a long-cycle team that stale share is easy to defend, because every stalled deal has a story about a budget cycle or a reorg. The 12-month rule removes the argument by making the test mechanical.

The earliest version of the same signal runs on a much shorter clock. In ORM customer data the strongest deal slippage signal is a rep changing the close date, and the earliest signal is the lack of any signal at all. On a long cycle, silence is the metric that moves first and the one most reporting ignores.

How do you forecast a quarter when deals take a year?

Decompose the quarter into its real sources of revenue rather than reading total coverage. Three sources fund any quarter. Carry-over deals already in pipeline on day one and expected to close. Deals created and closed inside the quarter. Deals pulled forward from later periods.

Long cycles compress the middle bucket close to zero, which makes carry-over composition the dominant variable and coverage a weak proxy for it. In ORM customer data, roughly 20% of the pipeline carrying in-quarter close dates on the first day of the quarter actually closes in that quarter, meaning 80% of the day-one value does not land. On a long-cycle team, that is the number that determines the quarter, and no coverage ratio expresses it.

Pull-forward deserves its own line. Closing a future deal early to save the current number costs the next quarter a deal and usually costs margin on the way. Tracking pull-forward as a named category keeps that trade visible instead of letting it disappear into a beat.

What does the weekly meeting look like?

Four numbers, each one a count rather than a rate. Stage entries last week. Close-date changes last week. Deals crossing into stale territory. Deals whose age passed their group curve.

Counts work on long cycles because every open deal contributes to them every week. Rates need closes, and closes are exactly what a long cycle refuses to supply on a weekly schedule. Building the operating rhythm around counts gives a fifteen-month motion the same weekly feedback loop a transactional team gets from bookings.

Keep forecast accuracy on the quarterly view where it belongs, and keep the weekly view on movement. The teams that manage long cycles well are not forecasting better in the final week. They are reading the shape of the year earlier, when the deals in question can still be influenced.