Rep productivity variance is the spread between the highest and lowest producing reps carrying comparable quotas on the same team. It answers whether revenue comes from a motion the company owns or from a handful of individuals the company employs. Those two situations look identical on a bookings report and behave completely differently when someone resigns.
Measuring the Spread
Rank every ramped rep by closed-won revenue in the period, then read three points: top quartile, median, and bottom quartile. Express the result as ratios.
| Team | Top quartile | Median | Bottom quartile | Top-to-median |
|---|---|---|---|---|
| Team A | $1.8M | $600K | $180K | 3.0x |
| Team B | $900K | $700K | $520K | 1.3x |
Exclude ramping reps from the calculation. Including them manufactures variance that is just tenure, and it hides real spread among tenured reps behind a distribution that looks explainable.
What Wide Variance Usually Means
Three causes account for most of it, and they call for different responses.
Territory quality is the first and most common. When account potential is unevenly distributed, output follows the accounts rather than the rep. Test this by comparing addressable revenue per territory before concluding anything about talent.
Segment mix is the second. A rep working two enterprise deals and a rep working thirty midmarket deals will always show different volatility, and comparing them directly produces a variance number that means nothing.
Skill and process adherence is the third, and it is the only one coaching fixes. It is also the one teams assume first and verify last.
The Forecasting Consequence
Wide variance breaks any model that applies one conversion rate across a team. Pooling a 45% closer and a 15% closer into a 30% team rate produces a forecast that is wrong in both directions at once, too low on strong territories and too high on weak ones. The error rarely cancels out, because the strong reps also carry larger deals.
The fix is grouping deals by attributes that actually predict closing behavior rather than by team membership, then predicting each group separately. That is the difference between a forecast that describes the team and one that describes the quarter. It also changes how forecast accuracy should be reviewed, since a team-level accuracy figure can look acceptable while every underlying territory is badly modeled.
Variance also belongs in the coverage conversation. A team that reports healthy pipeline coverage but holds most of that pipeline in bottom-quartile territories is not covered at all, and standard forecasting practice will miss it until the quarter closes.
Frequently Asked Questions
How do you measure rep productivity variance?
Rank ramped reps by closed-won revenue for the period and compare the top quartile to the median, then the median to the bottom quartile. Ratios are easier to read across teams than dollar gaps. A top quartile producing three times the median is a different management problem than one producing 1.3 times.
Why is the team average a poor substitute?
Averages hide the shape of the team. A team of eight where two reps carry 70% of revenue and a team of eight where every rep carries 12.5% can post the same average and the same total. The first is one resignation away from a bad year. The second is not.
Is high variance always bad?
High variance means the outcome depends on who owns the account rather than on the process. That makes territory assignment the highest-leverage decision on the team and makes any forecast built on team averages unreliable, because the average describes no actual rep.
How does variance affect forecasting?
Models that apply one conversion rate across a team will overstate weak reps and understate strong ones. When variance is wide, deals need to be grouped by patterns that actually predict closing behavior rather than pooled into a single team-level rate.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like rep productivity variance into prescriptive action for your team.
Schedule a Demo