You measure the ROI of sales coaching by comparing what coached reps or deals produced with what similar uncoached ones produced over the same period, then subtracting what the coaching cost.
Why most coaching ROI numbers overclaim
The comparison is the part most calculations leave out. One widely shared vendor calculator sets a baseline for top and middle performers, assumes coaching lifts the middle group's quota attainment by 10%, and divides the gain by the spend. It is easy to follow. It also credits coaching with the whole lift, when a new territory, a price change or a good quarter could explain some of it.
- 01Agree one change per sessionFor the next buyer call
- 02Check it happenedLeading stage: did behavior change?
- 03Compare coached and uncoachedLagging stage: same stage, same period
- 04Subtract what coaching cost
The gap is common. In our Game Winning Kickoffs field study, 21 studies or case studies claimed an effect from the event and none compared attendees with non-attendees. The one formal ROI calculation the study found, for a 2005 franchise event, came out at minus 53%, on a small sample.
A fair measure runs in two stages. The leading stage checks whether coaching changed behavior: in each session the manager and rep agree one change for the next buyer call, and someone checks whether it happened. That is the debrief step of a sales coaching framework, counted. If call scores are part of the check, test the scorer first. In our research on AI call scoring, one model changed 15% to 17% of its scores between identical runs, which is enough to hide a real change or invent one.
The lagging stage compares outcomes. Take deals that got a coaching session at a stage and deals at the same stage that did not, or reps who were coached with reps who were not, and compare stage conversion, win rate or quota attainment. The quality of the coach matters here: in a study of 1,246 reps, frequent coaching from less skilled managers was linked to worse results (Dahling et al. 2016, summarized in our kickoff field study). So measure by coach as well as by rep.
Forty deals a side (illustrative)
A team of 20 reps runs coaching on half the team's late-stage deals for one quarter, chosen at random. The numbers below are made up to show the method.
| Coached deals | Uncoached deals | |
|---|---|---|
| Deals at proposal stage | 40 | 40 |
| Won | 14 | 10 |
| Average deal value | $50,000 | $50,000 |
| Revenue | $700,000 | $500,000 |
The difference is $200,000 for the quarter. If coaching cost $30,000 in manager time and tools over the same quarter, the return is ($200,000 minus $30,000) divided by $30,000, about 5.7 times. On the leading measure, the agreed change showed up on the next call in most coached deals, which supports the idea that coaching caused the gap. With 40 deals a side, one quarter is a signal to keep testing, and a second quarter should agree before anyone quotes the number.
Where coaching ROI differs from sales training ROI
Sales training ROI is usually measured on the event: attendance, satisfaction scores and a quiz. Those numbers are easy to collect and say little about what reps do on calls weeks later. Coaching ROI has an advantage here, because coaching happens on live deals, so each session leaves a record of the change agreed and the deal it was agreed on.
Many sales leaders measure training and coaching as one budget line, so neither can be tested. Keep a separate line for coaching time and tools from the start.
Designing coaching so it can show a return
Tie each session to a live deal at a named stage, so the outcome has somewhere to show up. Agree one change, small enough to check on the next buyer call. Keep the coach and the scorecard the same for the whole test. And put the follow-up at the start of the next session, so the rep knows someone will ask.
A program run this way produces its own evidence. A program of general skills sessions, with no deal attached, can be valuable and still leave nothing to measure.
The sales coaching ROI formula, line by line
The formula is short: return equals the gain from coaching minus the cost of coaching, divided by the cost of coaching. The work is in what goes on each line.
The gain. Use the difference between the coached group and the comparison group. In the example above, the gain is the $200,000 gap. The $700,000 the coached deals brought in includes deals they would have won anyway. If you compare reps, use the same measure for both groups over the same period: quota attainment, stage conversion or revenue per rep.
Manager time. This is usually the largest cost and the one most often left out. Count the hours managers spend preparing for, running and following up on sessions, and multiply by a loaded hourly cost that includes salary, benefits and overhead. Every hour a manager spends coaching is an hour of their job that had other uses, such as hiring, forecasting or their own deals.
Rep time. A rep in a coaching session is not on a call with a buyer. For most teams this cost is small per session and real across a quarter, so count it.
Tools and outside help. Licenses for coaching or call recording tools, outside coaches, and any time from enablement spent building the program. If a tool is shared with other work, such as call recording that the team uses anyway, count only the share used for coaching.
Write each line down before the quarter starts, with how you will measure it. Deciding what counts after the result is in is the easiest way to get the answer you were hoping for.
What to count beyond revenue
Revenue is the cleanest outcome to compare, and it takes the longest to show up. Several other outcomes can carry real value, and each needs the same comparison group before it counts.
| Outcome | How to measure it | The comparison that makes it count |
|---|---|---|
| Ramp time for new hires | Days from start to first closed deal, or to a set level of pipeline | New hires who were coached against similar hires who were not, in the same period |
| Rep retention | Share of reps still in seat after 12 months | Coached reps against uncoached reps on similar teams |
| Forecast accuracy | Gap between committed and closed deals | Coached deals against uncoached deals at the same stage |
| Deal slippage | Share of deals whose close date moved out | Same split, same stage, same quarter |
| Deal size | Average contract value at close | Same split, with discounting tracked separately |
Forecast accuracy deserves a closer look. A session that finds a missing approver may move a deal out of commit and lower this quarter's number. That is a good outcome for the business, and a revenue-only measure scores it as a loss. Track slipped and pulled deals separately so the coaching that prevents a surprise gets credit.
Retention and ramp move slowly and depend on many things besides coaching, so treat them as supporting evidence. If coached reps ramp faster and stay longer, and the leading measure shows the agreed changes landing on calls, the case is stronger than any one number.
How much to trust one quarter's result
A small sample can swing a lot. In the illustrative example, the coached group won 14 deals and the comparison group won 10. If the coached group had won 12, the gap would have halved to $100,000, and the return would have dropped to about 2.3 times. Two deals moved the answer by more than half.
That is why the method asks for a second quarter. Swap the groups, so the reps or deals that were coached become the comparison, and run the test again. If both quarters point the same way, the result is worth quoting. If they disagree, the honest answer is that the team does not yet know, and the leading measure becomes the more useful guide: if agreed changes are landing on the next call, the coaching is doing its part, and the revenue effect needs more deals to show.
Bigger teams get there faster. A team with 200 late-stage deals a quarter can read a smaller gap than a team with 40 a side.
Running the comparison for one quarter: a checklist
- Pick the unit. Decide whether you are comparing deals or reps. Deals give larger samples; reps are easier to run when one manager coaches the whole team.
- Pick the stage. Choose one deal stage where coaching should matter most, such as proposal or the first meeting with the economic buyer.
- Assign at random. Flip a coin or use a spreadsheet to decide which deals or reps get coached. Do not let managers choose, because they will pick the deals they think can be saved.
- Write the cost lines. Manager time, rep time, tools, with the hourly rates you will use.
- Agree one change per session. Record it against the deal, in a sentence.
- Check the next call. Did the agreed change happen? Record yes or no, with the line from the call if you have it.
- Close the quarter and compare. Stage conversion, win rate or revenue, coached against comparison.
- Swap and repeat. Run the second quarter with the groups reversed before quoting a return.
Where ROI measurement goes wrong
Most attempts to measure coaching fail in one of five places, and each has a plain fix.
Nobody wrote down what the session agreed. Without a record of the change agreed on each deal, there is nothing to check on the next call and no way to link a session to an outcome. Sessions squeezed between forecast calls also get cancelled first, so the coached group ends up coached on paper only. Fix: one sentence per session, stored against the deal, and compare only the deals whose sessions happened.
Managers choose which deals get coached. Managers pick the deals they think are winnable or the ones in trouble. Either choice biases the comparison before it starts, and skipping the comparison group because it feels unfair to the reps left out is worse. Fix: assign at random, then swap the next quarter.
The result is read too early. Revenue needs at least one full sales cycle, and a quarter-one read on a six-month cycle is mostly noise. When finance asks for revenue in month two, the program looks like a cost with no return. Fix: agree with finance at the start which leading measure counts before revenue can, and when the revenue read is due.
Manager time is left out of the cost. Without it, almost any coaching program shows a high return, and the number will not survive a finance review. Fix: count preparation, session and follow-up hours at a loaded hourly rate, as in the cost lines above.
One team average hides the coaches. A team average can hide one coach whose reps improved and one whose reps did not, and the Dahling study above found that the quality of the coach changes the result. Fix: report the comparison by coach as well as for the team.
Setting up the comparison
How to measure sales coaching effectiveness covers the measures in full, and the sales coaching framework covers the sessions being measured. What is sales coaching has the basics.
Our AI sales experts, such as Caty, remember every deal a rep brings to them, so each session and the change agreed on it stay attached to the deal you will later compare. Managers get a read across the team (for leaders). If you want to run the coached and uncoached split next quarter, talk to us before it starts, while the deals can still be assigned at random.
FAQ
What is a good ROI for sales coaching? There is no reliable benchmark, because most published figures do not compare coached and uncoached groups. Our kickoff field study found 21 claims of an effect and 0 comparisons. Run your own comparison for a quarter and treat any return above the cost as a reason to keep testing.
How long does it take to see ROI from sales coaching? The leading measure shows within weeks: the change agreed in a session either appears on the next buyer call or it does not. Revenue takes at least one sales cycle. For a 90-day cycle, plan on two quarters before quoting a return.
Which metrics prove sales coaching is working? Two metrics, read together, show whether sales coaching is working. The first is the share of sessions where the agreed change appears on the next call; the second is stage conversion or win rate for coached deals against comparable uncoached ones. Activity counts, such as sessions held, show effort and say little about results.
Should ROI include retention and ramp time? ROI can include retention and ramp time if you measure them against a comparison too. Each needs the same test: did coached reps stay or ramp differently from similar reps who were not coached.
How do you calculate the cost of sales coaching? Add manager time (hours spent preparing, running and following up on sessions, at a loaded hourly cost), rep time spent in sessions, coaching and call recording tools, and any outside coaches or enablement time. Manager time is usually the largest line and the one most calculations leave out.
Can you measure AI sales coaching ROI the same way? Yes, AI sales coaching ROI is measured the same way. The method does not change: agree one change per session, check whether it shows on the next call, and compare coached deals with uncoached ones at the same stage. The cost lines change, with less manager time and a tool cost in its place, and the record of each session is easier to keep because the AI keeps it.
What if the team is too small for a comparison group? Compare deals instead of reps, and run the test for longer. A team of five reps cannot split into two fair groups, but its late-stage deals can be assigned at random to coached and uncoached over two or three quarters. Lean harder on the leading measure in the meantime: if the change agreed in each session keeps showing up on the next buyer call, the coaching is working on behavior, and the revenue comparison will firm up as the deals add up.