A sales interview role-play predicts performance when it copies the real job closely, is scored against a written rubric agreed before the interview, and includes a second attempt after feedback to test coachability. 'Sell me this pen' and improvised scenarios mostly measure confidence under pressure, which is not the same thing.
Almost every sales interview process includes a role-play. Few are designed well. The typical version is a hiring manager playing an unrealistic buyer, scoring on gut feel, with a scenario the candidate saw for the first time ten minutes earlier. It favours polished talkers and penalises careful sellers who ask good questions. This guide sets out how to build role-plays for account executives, SDRs, customer success managers and solutions engineers that tell you something useful, and how candidates should prepare for them.
Why do role-plays work, when they work?
Role-plays are a type of work sample: the candidate does a version of the job rather than describing it. Selection research has long supported work samples and structured methods over unstructured conversation. A widely cited 2021 re-analysis by Paul Sackett and colleagues, published in the Journal of Applied Psychology, revised earlier estimates and found structured interviews had a mean operational validity of 0.42, work sample tests 0.33 and unstructured interviews 0.19 (Sackett et al., 2021, summarised by Master International). Both structured interviews and work samples outperformed unstructured interviews in that analysis.
Two lessons follow. First, structure matters: the same scenario, the same buyer brief and the same scorecard for every candidate. Second, a role-play is not a substitute for the rest of a structured loop. It should test the skills a conversation cannot, alongside interviews that probe track record and motivation.
| Method | Sackett et al. (2021) | Schmidt & Hunter (1998) |
|---|---|---|
| Structured interviews | 0.42 | 0.51 |
| Work sample tests | 0.33 | 0.54 |
| Cognitive ability tests | 0.31 | 0.51 |
| Unstructured interviews | 0.19 | 0.38 |
Source: Master International summary of Sackett, Zhang, Berry and Lievens (2021).
Why does 'sell me this pen' fail?
Classic improvised role-plays fail for predictable reasons. The scenario bears no relation to the product or buyer, so it rewards generic sales patter. There is no preparation time, so it rewards quick talkers over careful planners. The interviewer improvises the buyer, so every candidate faces a different test. And scoring is a single overall impression, which tends to reward likeability and similarity to the interviewer.
Modern B2B selling is mainly about diagnosis, not persuasion. Gartner's research describes buying groups of six to ten decision makers for a complex B2B purchase, with buyers spending only around 17 per cent of their buying time meeting potential suppliers (Gartner). Salesforce's 2024 State of Sales research reported that 59 per cent of B2B buyers say reps do not take the time to understand their business's unique challenges (Salesforce, July 2024). A role-play that rewards a slick pitch tests the opposite of what buyers say they want.
Which role-play should you use for which role?
Match the exercise to the part of the job that most separates strong performers from average ones. For most AEs that is discovery, not the demo.
| Role | Best role-play | Prep time | Length | What it reveals |
|---|---|---|---|---|
| SDR / BDR | Cold call to a realistic persona, plus a written email sequence | 24 hours for the email; 15 minutes for the call | 10–15 min | Opening, handling brush-offs, earning a meeting |
| Commercial / mid-market AE | Discovery call | 48 hours with a brief | 30 min | Questioning, qualifying, quantifying pain |
| Enterprise AE | Discovery plus account plan or mutual action plan | 3–5 days | 45–60 min | Navigating a buying group, deal strategy |
| Customer success manager | Renewal-risk or QBR conversation | 48 hours | 30 min | Diagnosing value gaps, handling an unhappy sponsor |
| Solutions engineer | Technical discovery and tailored demo | 3–5 days with product access | 45 min | Translating features into the buyer's use case |
| Sales manager / leader | Deal review or coaching session with a 'rep' | 48 hours with deal data | 45 min | Coaching, forecast judgement, inspection |
For AE and SDR hiring see /specialisms/account-executives and /specialisms/sdr-bdr; for technical pre-sales see /specialisms/solutions-engineering.
How to design a sales role-play that predicts performance
Seven steps to a role-play worth running:
- Pick one job-critical skill. Choose the moment that most separates your best sellers from the rest. Usually discovery for AEs, the cold opener for SDRs, and the risk conversation for CSMs. Do not try to test the whole sales cycle in 30 minutes.
- Write a realistic buyer brief. Base it on a real (anonymised) customer. Include the company, the persona's role, their real problem, what they know and do not know, their objections and what would make them agree to a next step. The buyer must reveal key information only when asked good questions.
- Write a candidate brief. Give the candidate enough to prepare: your product overview, the prospect's company and role, and the goal of the call. Withhold what a seller would not know before a first meeting.
- Write the scorecard before the interview. Five to seven criteria, each with a description of weak, acceptable and strong behaviour. Agree it with every interviewer.
- Train the buyer. The person playing the buyer should rehearse, stick to the brief and respond consistently. Give them rules: no information unless asked, one hard objection at a set point, a realistic level of scepticism.
- Add a coachability round. After the first run, give one specific piece of feedback and ask the candidate to rerun the key section. Score the change, not the polish.
- Debrief and calibrate. Score independently before discussing. After each hiring round, compare scores to how hires actually perform and adjust the scenario and rubric.
What should a sales role-play scorecard include?
Below is an example scorecard for an AE discovery role-play. Each criterion is scored from 1 to 4 against behaviours, not impressions. Weightings reflect what predicts performance in most B2B motions; change them to fit yours.
| Criterion | Weight | Strong (4) | Weak (1) |
|---|---|---|---|
| Preparation | 10% | Researched the company and persona; opened with a relevant hypothesis | Generic opener; no evidence of research |
| Agenda and control | 10% | Set a clear agenda and agreed time; steered back when the buyer wandered | Let the buyer drive with no structure |
| Discovery depth | 25% | Open questions, followed up to root cause; uncovered the hidden problem | Surface questions; pitched after the first answer |
| Quantifying impact | 15% | Got the buyer to put a number or consequence on the problem | No sense of cost of the problem |
| Buying process | 15% | Identified decision makers, process, timing and competing priorities | Never asked who else is involved |
| Handling the objection | 10% | Explored the concern before answering; checked it was resolved | Argued or discounted immediately |
| Next step | 15% | Secured a specific, dated next step with the right people | Ended with 'I'll send some info over' |
Notice what is missing: confidence, energy and presence. They matter, but they show up in the behaviours above. Scoring them separately mostly measures how much the candidate resembles the interviewer.
How should you adapt the role-play for SDRs and customer success?
The discovery scorecard does not transfer directly to other roles. For SDRs, the job-critical moment is the first 30 seconds of a cold call and the ability to turn a brush-off into a conversation. Give the candidate a target persona and a short company brief the day before, then run two or three short calls with different buyer reactions: busy, sceptical and mildly interested. Score the opener, relevance to the persona, handling of the first objection, the quality of one or two qualifying questions and whether they ask for a meeting clearly. Add a written task, such as a three-step email sequence, because most SDR work is written.
For customer success managers, the hardest conversations are not friendly check-ins. They are renewals at risk, a sponsor who has left, or a customer who feels the product has not delivered. Give the candidate account data (usage trends, support tickets, contract value and renewal date) and have the interviewer play an unhappy stakeholder. Score how well they diagnose the real issue, avoid over-promising, agree a recovery plan with owners and dates, and spot any expansion signal without forcing it. See /specialisms/customer-success for how we assess CSMs.
How do you test coachability in an interview?
Coachability is one of the traits Mark Roberge emphasised when building HubSpot's sales hiring model, alongside curiosity, intelligence, prior success and work ethic (Freshworks summary of The Sales Acceleration Formula). It is also one of the easiest to fake in conversation: almost every candidate says they love feedback.
A coachability round tests it directly. After the role-play, the interviewer gives one specific, actionable note, for example: 'You moved to the demo after the buyer's first pain point. Go back to that moment and find out what it is costing them before you pitch.' Then the candidate reruns that section. Score three things:
- Reception: do they listen, ask a clarifying question and accept the point, or defend the first attempt?
- Application: does the rerun actually change in the way the feedback asked?
- Transfer: do they apply the principle elsewhere in the rerun, not only at the exact moment discussed?
A candidate who scores 2 on the first run and 4 on the rerun is often a better hire than one who scores 3 both times. They will improve faster under your coaching.
How should the 'buyer' behave?
Most role-plays are ruined by the buyer, not the candidate. Interviewers either make the buyer impossibly hostile to 'test resilience' or so helpful that every candidate succeeds. Both make the scores meaningless. Give the buyer a script with rules:
- Answer what is asked, truthfully per the brief, but do not volunteer key information
- Raise one prepared objection at a set point, the same for every candidate
- Show realistic, mild scepticism; real buyers are busy, not rude
- Reveal the 'hidden' problem only if the candidate asks a good follow-up question
- Agree to a next step only if the candidate earns it by the scorecard's standard
- Never coach during the role-play; save it for the coachability round
Where possible, use the same person to play the buyer for every candidate in a hiring round. Having a second interviewer score silently, rather than the buyer scoring while acting, gives a cleaner result.
Is it fair to ask candidates to prepare?
Yes, within limits. Preparation time is what makes the role-play realistic; sellers prepare for real calls. But respect candidates' time, especially those in demanding current roles:
- Keep preparation to what a seller would do before a real first meeting, typically an hour or two
- Send the brief at least 48 hours ahead, longer for enterprise and SE exercises
- Do not ask for unpaid work that could be used commercially, such as a real outbound campaign for your prospects
- Tell candidates exactly what will be assessed and who will be in the room
- Offer reasonable adjustments, such as extra time or written materials in advance, for candidates who need them
Common mistakes in sales role-plays
| Mistake | Why it hurts | Fix |
|---|---|---|
| Scenario unrelated to your product or buyer | Tests generic patter, not the job | Base it on a real anonymised customer |
| No preparation time | Rewards fast talkers over careful sellers | Send a brief 48 hours ahead |
| Improvised buyer | Every candidate faces a different test | Write and rehearse a buyer brief |
| Single overall score | Rewards likeability | Score five to seven criteria against behaviours |
| Hostile buyer to 'test grit' | Measures tolerance for an unrealistic situation | Realistic scepticism with a set objection |
| Testing the whole sales cycle | Shallow view of everything | Pick the one job-critical moment |
| No feedback round | Misses coachability | Add a rerun after one specific note |
| Never reviewing results | Scenario never improves | Compare scores with on-the-job performance each hiring round |
How should candidates prepare for a sales role-play?
If you are the candidate, treat the role-play as a real first meeting with a real prospect. The strongest candidates we put forward tend to do the following:
Preparing as a candidate:
- Ask what is being assessed. It is a reasonable question, and the answer tells you whether to focus on discovery, the demo or the objection.
- Research the 'prospect' properly. If the scenario uses a real type of company, read how such businesses make money and what typically goes wrong.
- Prepare a hypothesis, not a pitch. Go in with two or three likely problems and the questions that would confirm them.
- Set an agenda. Open by agreeing time, purpose and the outcome you both want.
- Ask, then follow up. The second and third question on a topic are where the useful information is.
- Quantify and qualify. Find out what the problem costs and who else decides.
- Close for a specific next step. A dated meeting with named people, not 'I'll follow up'.
- Welcome the feedback round. If they give you a note, apply it visibly. That is the test.
Candidates looking for their next AE, SDR, CSM or SE role can register at /candidates#register, and we will brief you on the interview process for every role we put you forward for.
Putting the role-play into the full interview loop
A role-play is one stage, not the decision. A sound loop for a mid-market AE might run: recruiter screen on track record and deal fit; hiring manager interview with structured questions on pipeline generation and past deals; discovery role-play with a coachability round; panel with a peer and a cross-functional partner; references with a former manager. Each stage should test something the others do not, and every interviewer should know which criteria they own. Pair the role-play with the checks in our guide to verifying quota attainment, and you have a process that tests both what the candidate has done and what they can do.
Sales role-play interviews: common questions
A simulated sales conversation in which the candidate plays the seller and an interviewer plays a buyer. It lets you observe how the candidate handles a real part of the job, such as discovery, objection handling or a cold call.
Well-designed ones help. Selection research rates work sample tests and structured interviews well above unstructured interviews. The key is realism, a written scorecard and consistency across candidates.
For most AEs, a discovery call based on a realistic buyer brief. Discovery separates strong sellers from average ones more reliably than a demo or a pitch.
Around 10 to 15 minutes for an SDR cold call, 30 minutes for a mid-market discovery call, and 45 to 60 minutes for enterprise or solutions engineering exercises, plus time for a coachability rerun.
At least 48 hours for most AE and CSM role-plays and three to five days for enterprise and SE exercises, with a clear brief and an explanation of what will be assessed.
Give one specific piece of feedback after the role-play and ask the candidate to rerun the relevant section. Score how they receive the feedback and how much the rerun improves.
They can, but a trained interviewer following a written buyer brief, with a second person scoring, gives more consistent results.