Peer Roleplay Practice: Honest Rules

Why does peer roleplay practice need rules at all?

Because unsupervised practice drifts into conversation, and conversation leaves no evidence. Manager-led roleplay is the ceiling on program volume: one manager cannot sit in every rep's rehearsal every week, so teams either practice unsupervised or skip practice and rehearse on live buyers instead. Unstructured peer practice produces a pleasant conversation and no evidence a manager can act on.

The case against practice environments deserves a fair hearing. Reps do learn on live calls. Real buyers push back in ways no partner invents, and the stakes concentrate attention. The problem is price. Burning a live discovery call to teach a first response to we already have a vendor is the most expensive coaching a sales org can buy, and the rep gets one attempt at that buyer.

Pairs solve the volume problem, not the judgment problem. A pair session costs 24 minutes of rep time and produces two recorded runs, and it costs the manager nothing except the weekly audit. Pairs multiply repetitions; they do not multiply judgment. Rules carry the rest: fixed format, fixed rubric rows, fixed sequence in the debrief, and a manager audit on top.

The 12-minute format for rep-to-rep practice pairs

We recommend a 12-minute block per running rep, split 2 / 6 / 4. Both reps run once, so a full pairs session takes about 24 minutes plus a minute of switching. A time-boxed peer session ends on the clock, not when both reps run out of things to say.

The 2-minute setup is not small talk. The running rep states the scenario, the buyer persona, the call stage, the one behavior being drilled, and the two rubric rows out loud before the run starts. Then set a Sandler up-front contract inside the roleplay itself: purpose, time, both agendas, and the acceptable outcomes including no.

The 6-minute run is recorded and uninterrupted. The partner plays the buyer and stays in character. No coaching mid-run, no restarts, no breaking to explain what the buyer meant. A rep who cannot recover inside six minutes has found the drill.

The 4-minute debrief covers one behavior only. Self-diagnosis first, verdict second, re-run date third.

Which two rubric rows is the partner allowed to score?

One row for the behavior being drilled, and one fixed row for the call stage. Nothing else. Two rubric rows is a limit, not a shortcut: a partner can confirm a behavior appeared, but a partner cannot grade whether the call was good.

Write both rows as behaviors a third party could verify from the transcript. Asked at least one implication question before naming any product capability is scoreable - the SPIN implication question either appears in the transcript or it does not. Handled the objection well is not scoreable, and a partner scoring it will simply reward confidence. Score each row as pass or fail, or on a three-point scale where the middle point has a written definition.

Keep diagnostics off the peer scorecard. Talk/listen ratio is worth investigating on a scored session, but it is an input a coach interprets, not a bar a partner passes someone on.

Rotate the drilled row weekly so pairs do not turn into a single-skill habit. The stage row stays fixed for the quarter, which is what lets a manager read a trend per rep per scenario rather than one blended number.

Self-diagnosis before verdict

The running rep speaks first, always. Give the pair two scripted prompts and forbid improvisation on the sequence: Name the one moment you would take back, and What did the buyer do in the ten seconds after it. Only then does the partner render a verdict on the two rows.

The order matters more than the wording. When the partner leads, the rep spends the debrief defending. When the rep leads, the debrief becomes a check on whether the rep can hear the call.

The partner's verdict is three sentences at most: the row, the score, and one line quoted from the run that justifies it. Then set the re-run date before the pair stands up - same scenario, same two rows, inside the week.

The single-behavior rule is the same one that governs manager debriefs; the mechanics are laid out in Sales Roleplay Debrief: One Behavior, Re-Run.

Pairing rotation for a peer coaching sales team

Publish the rotation, do not let reps pick. Self-selected partners drift toward comfort, and comfortable partners grade generously. Write a fixed rotation for the quarter, rotate pairs weekly, and pair across tenure so a second-year rep faces a new hire's questions and a new hire hears a veteran's first responses.

Cadence we recommend: two pairs sessions a week, 24 minutes each, on the calendar as recurring blocks. Keep them separate from pipeline reviews. Deals are urgent and skills are merely important, so a skills block that lives inside a deal review gets eaten by the deal review every time.

Give every pair the same scenario for the week, drawn from a real call the team lost or nearly lost. Same scenario across pairs is what makes the manager's audit meaningful - otherwise you are comparing scores across different difficulty levels and calling it a trend.

Pairs do not replace the manager's own weekly session. Keep the 30-minute manager-run format (5 setup / 15 drill / 10 debrief), plus roughly 10 minutes beforehand to pick the scenario from a live deal and choose the rubric row.

What does a failing peer debrief sound like?

Three notes and no date. The failing version sounds like this: Good energy. Maybe slow down a bit. Also you probably could have asked about budget earlier. Nice job. Four sentences, three issues, no moment, no rubric row, no verdict, and nothing on the calendar.

The passing version sounds narrower and less friendly. The rep goes first: At 3:40 the buyer said the timing is bad and I moved straight to a demo offer. The buyer went quiet for six seconds. Then the partner: Row two, first response to timing objection - fail. Your line was Totally understand, want me to hold a demo slot for January?, and that came before you asked what changes in Q3. Re-run Thursday at nine, same scenario.

Two other failure modes to watch. The partner breaks character mid-run to explain what the buyer meant, which converts the drill into a discussion. And the partner plays an easy buyer - accepting the first answer, volunteering pain, agreeing to the meeting. Write the buyer's resistance into the scenario card so the partner is following instructions rather than choosing how hard to be.

The manager spot-audit that catches grade inflation

Sample one pair recording per week and re-score the same two rows yourself. Compare your score to the partner's, and log the gap by pair and by scenario, never as one blended accuracy number. Inflation shows up as a directional gap: the partner's score sits above yours on the same row, week after week.

When a pair's scores sit above yours twice in a row, stop auditing and calibrate. Put both reps and yourself on one recording, score independently, then read the gaps aloud row by row. Some of the gap is generosity toward a colleague. Some of it is two people reading the same row differently. Both get fixed the same way: rewrite the row into something more concrete and re-score the recording together. The mechanics of that session are in the Sales Rubric Calibration Guide for Managers.

Audit the re-runs too. Ask one question in your 1:1: which failed drill did you re-run last week, and what changed on the second attempt. If the answer is vague, the pair is scoring passes to avoid the work of scheduling a second run.

The honest limit: repetition is not readiness

Peer practice buys volume; a manager or a scored transcript still has to sign the gate. A partner can verify that a rep said the words, ran the sequence, and held the frame for six minutes. A partner cannot certify that the rep is ready to take solo discovery calls, own tier-one leads, or run a demo unsupervised.

Keep the two roles separate on the calendar. Pairs are practice reps. The gating rep is a named scenario scored against your call stages and rubric - by a manager watching, or on a platform that records, times, and transcribes the session and returns a rubric score per skill with flags linking back to the exact moment. XL Roleplay does the second version; the point is the evidence, not the tool. How to structure those decisions is covered in Sales Readiness Scoring for Evidence.

Two cases where pairs are the wrong tool. A brand-new hire with no agreed first response yet will rehearse a guess into a habit - teach the first response before pairing them. And high-consequence scenarios with technical content, like a security review, need someone in the room who knows the correct answers.

Run this drill this week

Pick one objection your team lost a deal to last quarter. Write a one-paragraph scenario card: buyer persona, call stage, the objection verbatim, and the buyer's instruction to resist the first two responses. Write two rubric rows underneath it. Publish the pairing rotation Monday morning.

Run the 12-minute format twice per pair, once in each direction, twice this week. Recordings go in a shared folder. You re-score one of them Friday.

Pass bar, three parts, all verifiable from the recording alone. The running rep names a specific timestamped moment before the partner speaks. The partner's verdict cites one of the two rubric rows and quotes one line from the run. A re-run is on the calendar with a date and time.

A failing attempt sounds like this: I thought that went pretty well overall - yeah, felt good to me, maybe just be more confident next time. No timestamp, no row, no quoted line, no date. When you hear it on the Friday audit, do not send feedback by message. Sit with the pair, replay ninety seconds, and make them run the debrief again in front of you.