Skip across the chapter
Quota Playbook

Role-play coaching: 3 numbers to check before 1:1s

Quota Playbook Editorial Team · Published · 10 min read

Role-play is worth your 1:1 time when you can see it change a result you already care about.

Key takeaways

  • Choose an outcome and record its baseline.
  • Separate repeat practice from better live-call performance.
  • Set a practice count and review date before starting.

It feels like practice, it fits in a calendar slot, and it puts a rep in the position where a hard conversation actually happens. That is the appeal. What it does not come with is proof that it is worth the time. Your 1:1 time is limited, and every slot you give to practice is a slot you do not give to a review of a recent call, feedback on a deal, or a conversation about a rep's goals.

So before you give role-play a standing place in your 1:1s, check three numbers from Mindtickle's published material, and check what each one can and cannot support. The goal is not to find a figure that justifies a decision you already made. The goal is to decide in advance what would change your mind.

Both sources are published by Mindtickle itself, so treat them as one company's account. The first is a customer story about Janssen India. The second is a blog post about AI sales role-play coaching in financial services. Vendor material can still be useful, but read it as a starting point for your own check, not as a result you can carry over unchanged.

How to decide whether role-play belongs in your 1:1

The three numbers below are a place to begin that check, not a verdict on whether it works for your team.

Start by naming the outcome. The sources measure different things. One tracks whether learners went back to practice, and the other tracks scores on live calls. Those are two different questions. A rep who repeats a role-play may still struggle on the phone, and a rep with strong call scores may not practice at all. Pick the outcome that matters to your team before you look at any figure, and write down where it stands now. Without a baseline, a good-looking number tells you very little about your own reps.

Then decide what would count as a fair test. Think about how long you can watch the result, how many reps are involved, and what else changes in the same period. A short pilot with a few reps answers a narrower question than a full year across a whole team. You should be clear about which question you are asking before you read any numbers, because the answer you get depends on the question you asked.

Number one: the analysis covered more than 500 sellers over three years

The Janssen India story says: "To do this, it analyzed the behavior of more than 500 of its sellers using Mindtickle over three years."

Two things to notice. First, the group is the company's sellers, not yours. A finding from more than 500 people at one company tells you about that company. Whether your reps look similar is a question you have to answer yourself, by looking at your team's starting point, its experience, the products it sells, and how it is coached today. Check those differences before applying the finding to your team.

Second, three years is a long window. Consider what else could change during that time, such as the product, the territory, or the people, before attributing a result to practice. A long window can smooth out a bad quarter, but it also gives other changes more room to happen alongside the one you are watching.

Before you use this number, ask how many reps you manage and over what period you can measure. A small team measured over a few weeks may not produce the same kind of evidence, and you should not treat its results as if they do. If your team is small, your own before-and-after comparison will be noisy. That is a reason to be patient with the comparison, not a reason to borrow a larger company's result.

Number two: 71% of learners reattempted after video role-play

The Janssen India story puts it this way: "The team discovered that video role-plays were the most effective practice method, with 71% of learners reattempting at least once."

Look closely at the wording. This number measures a behavior, not a skill. It says learners came back to practice. Repeating an exercise is a reasonable sign of engagement; assess call performance separately. Reattempting, passing and improving are different outcomes.

Consider what this would look like on your team. Suppose you run a role-play exercise for a month and count how many reps go back to repeat it. That count tells you about engagement with the exercise. It does not tell you whether the second attempt was better, or whether the rep handled the next live call differently. To answer those questions, you need a second measure, which is why the outcome you chose in the first step matters so much.

The number is also specific to video role-play. If your 1:1 practice is a spoken exercise, a written scenario, or something else, this figure does not cover it. Treat it as evidence about video, not as a general claim about role-play.

Number three: 61% against 57% for reps with 10 or more role-plays

Reps who complete 10 or more role-plays a year scored 61% on live calls, compared with 57% for reps who complete fewer, according to Mindtickle's 2026 State of Agentic Revenue Enablement Report. The figures appear in a Mindtickle blog post on AI sales role-play coaching in financial services, which attributes them to that report.

This is the most usable of the three for a manager, because it is about a count you can track. The 10-a-year line gives you a threshold to compare against. Still, the comparison is between two groups of reps. The figure does not say that the extra role-play caused the higher scores. Reps who practice more may differ from reps who practice less in other ways. The gap between the two groups is the thing to look at, not either number alone.

For your own team, the practical version of this figure is simple. Count the role-plays each rep completes over a set period, split the reps into two groups around a threshold you choose, and compare the outcome you picked for each group. You will not get the same answer the financial services figures give, and you should not expect to. You will get an answer about your reps, and that is the one you need to act on.

What these numbers do not tell you

They do not tell you how much role-play your team needs. Neither source gives a dose you should copy. Keep these limits in mind when you read them:

  • The three-year analysis covered more than 500 sellers; keep that scope separate from the reattempt and passing-score figures.
  • The financial services figures come from one report, as cited in one blog post.
  • Compare your reps, your product, and your calls before choosing a practice plan.

None of this makes the numbers useless. It means they are good questions, and you should answer them with your own data. A number that makes you ask a sharper question about your team has done its job, even if it does not settle the matter.

If you bring these numbers to your leader, bring the question and the plan too

When you take role-play to your leader, bring the question, the baseline, and the review date with the figures, not the figures alone. A leader who sees only "71%" or "61% against 57%" will reasonably ask whether those numbers describe their own team, and the honest answer is that they describe someone else's. A short proposal fixes that problem before it starts.

The proposal can be brief. Name the one outcome you will track. Say how many reps are in the pilot and which role-play format you will use. State the count of practice attempts you expect each rep to complete, and say when you will review the result. Then say what you will do if the result is unclear. That last part is what makes the proposal credible, because it shows you have thought about the case where the test does not show a clear result.

Be careful about how you describe the source. Say that the figures come from Mindtickle's published material, and that they describe other teams. Keeping that difference clear makes a proposal easier to trust, because it shows you can separate evidence from hope.

How to use these numbers in your next 1:1

Start with a small, measurable block of role-play, not a full program. Use the three numbers as questions to answer with your own data, not as targets to hit.

  1. Choose one outcome. Pick live-call scores, reattempts, or time to ramp. Record where each rep stands now, so you have something to compare against later. Choose only one main outcome for the test, so that a change in one measure does not get lost among several others.

  2. Set a dose you can count. The 10-a-year figure gives you a count to compare with. Ask how many role-plays each rep completes in a set period, and whether that count moves with the outcome you chose. Write the count down. If you only remember it, you will remember the reps who stood out, not the ones who did not.

  3. Define what happens when a rep falls short. The Janssen India story says that "if a rep doesn't achieve a passing score of 80%, action is taken to improve knowledge." That threshold belongs to that team. Choose a threshold suited to yours. The idea is still worth copying: set your threshold in advance and decide the action before you need it, so you are not inventing a response in the middle of a hard 1:1.

  4. Keep a short log. After each 1:1, write one line for each rep: the role-play done, the reattempt count if you track it, and one thing you noticed on the next live call. A log like this takes a few minutes a week, and it is far more reliable at review time than memory. Use these sales coaching log checks to review your record.

  5. Fix a review point. Look at the same measure at the end of the period you set, and compare it with the baseline you wrote down. Do not change the measure halfway through, or the comparison stops meaning anything. If the result is unclear, say so, and extend the test rather than declaring a win.

What a role-play block can look like in a 1:1

Keep the block to one scenario, one practice attempt, and one piece of feedback. A single realistic conversation with specific feedback is easier to repeat and easier to measure than a long list of scenarios in one sitting. Fit the block into your sales one-on-one agenda.

Open the 1:1 by stating the scenario and the one behavior you want to see. Run the role-play, then ask the rep what they would change before you give your own view. That order matters, because it makes the rep the first person to evaluate the attempt. Close by agreeing on whether the rep will reattempt the scenario before your next 1:1, and write that down with the rest of your log.

If you record practice, keep the recording short enough that the rep will watch it. A reattempt can only be counted if it happens, and a small, specific next attempt is more likely to happen than a broad one.

Illustrative example: count repeat practice

Suppose 10 reps try a video role-play and 5 repeat it. Record 5 reattempts among 10 reps. That is an engagement count. Review live calls separately before deciding whether the practice changed call performance.

Today, choose a scenario for your next 1:1 and write down the behavior you will observe.

A check before you commit the time

When should you commit coaching time to role-play?

Commit your coaching time to role-play when you can answer three questions: What outcome will I measure? How many role-plays will each rep do? When will I review the result? If you cannot answer all three, the numbers in this post are not enough to justify a standing slot. A short pilot is a better use of your next few 1:1s than a permanent commitment you cannot evaluate.

Does repeating a role-play prove better call performance?

No, a reattempt records repeat practice rather than a change in live-call performance.

Sources

  1. Rep call score averages: 3 values to compare
  2. Buyer AI research: 5 facts for discovery coaching
  3. Accessible sales decks: 6 checks before coaching
  4. Driver's license on job posting
  5. CRM data ownership coaching notes

More plays