Attrition gets handled in capacity models as a percentage subtracted at the end. Take the roster, shave 15 percent, move on. That treatment understates the cost, because a departure removes capacity in four separate ways and the percentage captures one of them.
What does a single departure actually cost?
Open-seat days, backfill ramp, orphaned pipeline decay, and manager time. The components that can be sized run to five or six rep-months of productive capacity in the model below. The seat being empty is the smallest of the four.Break it out for a rep carrying an $800,000 annual quota:
| Cost component | Duration | Capacity lost |
|---|---|---|
| Notice period at reduced productivity | 3 to 4 weeks | ~0.5 rep-months |
| Open seat before backfill starts | 8 to 12 weeks | ~2.5 rep-months |
| Backfill ramp to full productivity | 12 to 24 weeks | ~2 rep-months of partial output |
| Pipeline decay on orphaned opportunities | Ongoing | Varies, often the largest item |
Why does orphaned pipeline decay faster than expected?
Because a deal with no owner stops generating the signals that predict progression, and by the time anyone notices, the buying process has moved on. Reassignment speed decides how much survives.ORM identifies the earliest deal-slippage signal as the absence of a signal: no stage change, no close date change, no amount change, no notes. Orphaned pipeline produces exactly that pattern by construction. Nobody is updating the record because nobody owns it.
The second signal compounds it. ORM identifies a rep changing a close date as the strongest available slippage indicator, and notes that a deal slipping from one quarter to the next is less likely to close even when it sits in commit. A departure produces a cluster of those pushes once the new owner resets expectations with the buyer.
Three rules that limit the decay:
1. Reassign every open opportunity within 72 hours of notice, before the departing rep's last day. 2. Prioritize reassignment by close date proximity, not by deal size. A $50,000 deal closing in three weeks is more perishable than a $300,000 deal closing in two quarters. 3. Have the departing rep run a live handoff call on any deal with a close date inside the current quarter. A written handoff note does not transfer the context that keeps the deal moving.
How should attrition enter the capacity model?
As scheduled backfill hires with their own ramp curves, not as a percentage haircut on the roster. The two methods produce very different hiring plans.The percentage method: 30 reps, 15 percent attrition, plan for 25.5 effective reps. That treats the lost capacity as evenly distributed and instantly replaced, and it is wrong in both respects.
The cohort method: 30 reps, 15 percent attrition, which is 4.5 departures spread across the year. Each departure creates an open seat for a quarter and a backfill that ramps over two more. Model each as a separate hiring event.
| Quarter | Departures | Backfill starts | Ramped backfill capacity (rep-quarters) |
|---|---|---|---|
| Q1 | 1 | 0 | 0 |
| Q2 | 1 | 1 | 0 |
| Q3 | 1 | 1 | 0.25 |
| Q4 | 1.5 | 1 | 0.85 |
How do you forecast attrition rather than react to it?
Use your own trailing rate split by tenure band, and watch the leading indicators inside each band. A single blended number hides which population is leaving.Split attrition three ways. Reps in their first year leave for different reasons than reps in year three, and voluntary and involuntary departures should be modeled separately since only one of them is predictable from performance data.
Leading indicators worth tracking monthly:
- Attainment trajectory. Sustained attainment well below plan precedes both voluntary and involuntary exits. Set the threshold from your own history rather than a generic cutoff. - Pipeline generation rate. A rep who stops self-sourcing has usually disengaged before they resign. - Territory fairness gap. Reps who conclude their book cannot reach the number disengage, and the gap in modeled potential between books is worth checking before a resignation makes the point for you. - Activity decay. Falling meeting volume with a stable pipeline is a disengagement pattern, and it shows up weeks before a resignation.
None of these are proof. They are inputs to a hiring pipeline that should be warm before it is needed, since a search started on the day of resignation adds a full quarter to the recovery.
What should happen to the forecast when a rep leaves?
The forecast should drop that week for the orphaned in-quarter pipeline, rather than waiting for the quarter to close short. Most teams adjust the capacity plan and leave the current forecast untouched.The adjustment should be specific rather than a blanket haircut. Identify every opportunity owned by the departing rep with a close date inside the quarter, then apply a discount based on stage and days to close. Late-stage deals with an engaged champion often survive a handoff. Mid-stage deals where the rep was the primary relationship rarely do.
A forecast that reprices this immediately preserves the thing that makes it useful. ORM's position is that getting the forecast right in the last week of the quarter helps nobody, because by then the quarter has already happened. A departure is exactly the kind of change a static model absorbs silently and a responsive one reflects the same week. That responsiveness is the core argument in sales forecasting best practices, and it is measurable through forecast accuracy tracked by week rather than only at quarter close.
How do you build an attrition buffer into the hiring plan?
Hire the backfill capacity ahead of the departures, sized from your trailing rate and phased across the year. Waiting for a resignation to trigger a search guarantees the gap.Two structures work. The first is a rolling over-hire, where the plan carries one to two heads above the territory count so that a departure has a partially ramped rep available to absorb the book. The second is a bench of ramping reps in a pooled territory who take over named books as seats open, which suits companies with predictable attrition and enough volume to keep a bench busy.
Both cost money against a plan that assumes nobody leaves. That plan has never described any sales organization. Pair the buffer decision with a look at territory-level pipeline coverage, since the territories most exposed to a departure are the ones already running below the ratio and least able to absorb a stall.
Frequently Asked Questions
How much capacity does one departure cost?
Open-seat days plus the full backfill ramp, which in the model on this page runs to five or six rep-months per departure before counting pipeline decay. Modeling a backfill as continuous coverage understates the loss by most of its value.
What attrition rate should a capacity model assume?
Your own trailing rate over at least eight quarters, split by tenure band. New-hire attrition and tenured attrition behave differently, and averaging them hides which one is driving the loss.
What happens to the departing rep's pipeline?
Much of it goes dormant. ORM identifies the absence of a signal, meaning no stage change, no close date change, no amount change, as the earliest warning that a deal is dying. Orphaned pipeline produces exactly that pattern, so it needs an owner within days rather than weeks.
Should the hiring plan include buffer heads for attrition?
Yes. Add planned backfill hires to the schedule as new hires with their own ramp curves rather than treating them as replacements. A capacity plan with no attrition buffer is planning for a roster that has never existed.
Does attrition affect the forecast or only next year's plan?
Both. The current forecast should drop when a departure creates orphaned in-quarter pipeline, and the forward capacity plan should absorb the ramp gap. Teams usually adjust the second and leave the first untouched until the quarter closes short.
See how ORM turns these insights into action
ORM builds custom revenue forecast models for B2B SaaS companies. Not dashboards. Prescriptive analytics that tell you what to do next.
Schedule a Demo