sales-hiring

Sales Hiring

Build the four artifacts a hiring manager needs to recruit an SDR/BDR or AE with evidence-backed rigor: a role scorecard, a structured interview loop with question bank, a scored mock-call work sample, and a 30-60-90 ramp plan. Structure beats intuition at every stage - the whole skill exists to enforce that.

Label every benchmark output with a short parenthetical naming its source - peer-reviewed, survey, crowdsourced, vendor data, or a named practitioner - and never present a lower tier as a higher one, or a vendor claim as fact.

If the user turns out to be a candidate preparing for a sales interview rather than a hiring manager running one, stop and point them to mbfinotti/sales-skills@sales-career - this skill never coaches the candidate.

Out of scope, do not drift into:

• sourcing candidates
• writing or distributing job ads
• running an applicant tracking system
• designing the commission plan as a strategy exercise
• SDR-to-AE promotion mechanics

Compensation appears here only as the published range and the base/variable split and ramp draw that shape the candidate funnel and the ramp.

Interview

Ask before producing anything. Stop at these ten questions.

• One question per message.
• Offer the multiple-choice options.
• Skip any question the user's message already answers.

Questions 8-10 exist to re-rank the orderings in this skill before the user commits to a path - ask them here, never later.

1. Team stage: (a) first sales hire, founder still selling; (b) hires 2-4, playbook forming; (c) hires 5-30, repeatable playbook exists; (d) 30+, scaling a known model.
2. Role in mind: (a) SDR/BDR; (b) AE; (c) full-cycle rep; (d) not sure - recommend one for me.
3. Pipeline source mix: (a) mostly inbound; (b) mostly outbound; (c) balanced mix.
4. ACV and typical sales-cycle length: (a) under ~$5K, under a month; (b) ~$5-25K, 1-3 months; (c) ~$25-100K, 3-6 months; (d) over ~$100K, 6+ months.
5. Segment: (a) enterprise B2B; (b) high-velocity inside sales (SMB/mid-market B2B); (c) B2C/consumer.
6. Comp philosophy: base-heavy (60:40 or more) or aggressive variable (50:50, uncapped accelerators)? And will you publish the range in the posting?
7. Hiring jurisdiction(s): US (which states, notably NYC/CO/CA/NY/WA/IL), EU (which countries), UK, other - this drives the compliance checks.
8. Deadline: by what date must the offer be signed? (a) this month; (b) this quarter; (c) no hard date.
9. One-off or compounding: a single hire, or the first of several where the scorecard, question bank and rubric get reused? (a) one-off; (b) compounding.
10. Effort ceiling: how many interviewer-hours per candidate can you spend, and do you have a recruiter and an existing trained panel? (a) under ~5 hours, no recruiter, no panel; (b) ~5-15 hours; (c) more, with recruiting support.

Reading the answers

Every answer above drives an output; use them, don't just record them.

• Q3 + Q4 decide the role recommendation (Q2), and they flip the ordering. Default for ACV under ~$25K on cycles under three months with inbound flow (read > as "more of this axis"; the efficiency line is the recommendation):

  • efficiency: full-cycle rep > AE only > SDR + AE split
  • value (pipeline generated and revenue closed per hire): SDR + AE split > AE only > full-cycle rep
  • effort (loops to run, ramps to write, handoffs to police, time-to-fill): SDR + AE split > full-cycle rep > AE only Above ~$25K ACV on 3-6+ month cycles with an outbound motion, the value line wins and the split leads - one rep cannot both prospect a mapped account list and run six-month deals. Go/no-go, stated as a deletion: while inbound volume keeps a closer's calendar full, delete the SDR + AE split from the menu instead of ranking it last, since a ruled-out option parked at the bottom comes back as headcount. Put it back only once AEs show empty-calendar syndrome or the company commits to outbound as a strategy [survey + practitioner]. If the user picked a role in Q2 that contradicts this, challenge the pick before proceeding and say why.

• Q1 decides the candidate profile.

  • First hire: a founder-led-sales handoff problem. Hire a builder who sells without infrastructure, and treat heavy big-company experience as a risk (the "coin-operated" rep who imports a formula instead of building the company's own). Mark Roberge's regression on his own team found aggression and objection-handling nearly uncorrelated with success; coachability, curiosity, intelligence, and work ethic predicted it (his own single-company regression, not a broader study). For a founder or CEO running this interview personally, SaaStr's repeated framing for the first 2-10 reps is a single gut-check test layered on top of the scorecard, not a replacement for it: would you, the founder, actually buy the product from this person (practitioner opinion).
  • Hires 5-30: a repeatability problem. Profile from the org's own top performers and hire the same rep every time.

  Also warn: doubling the team requires doubling each demand-generation source in proportion, or new hires starve.

• Q5 sets loop length, work-sample content, predictor weighting, comp split and sourcing - see B2B and B2C below.

• Q6 shapes the funnel and the ramp.

  • Base-heavy splits attract risk-averse candidates and suit reps who start with zero pipeline.
  • Aggressive 50:50 plans with uncapped accelerators self-select confident closers.
  • An unpublished "competitive OTE" repels strong candidates, who read it as a signal the quota is set to fail (practitioner opinion).

  Publishing a range is also legally required in a growing set of jurisdictions (Q7).

• Q7 triggers the compliance pass in workflow step 4.

• Q8 promotes the fast-acting options. A this-month date:

  • deletes the 5-6 stage enterprise loop
  • promotes the single-loop roles over the SDR + AE split
  • promotes reusing an existing question bank over writing one

  No hard date leaves every default order below in place.

• Q9 decides whether the work sample stays starved.

  • One-off: the build amortizes over a single candidate pool, so run the mock call but build it thin - no quarterly rubric refresh, no scorer-training programme.
  • Compounding: the frozen rubric and question bank serve every future hire, which promotes building both properly and first - this is the condition in Selection evidence that overrides the efficiency order.

• Q10 deletes rungs rather than reordering them.

  • Under ~5 interviewer-hours per candidate: cut to 3-4 stages and keep only the structured interview and the work sample.
  • No recruiter and no trained panel: every hour in the Selection evidence table becomes the hiring manager's own - re-rank against that, not against a staffed team.

Workflow

Produce the four artifacts in order. Deliver each one, validate it with the user section by section, then move to the next - never dump all four unreviewed.

1. Run the Interview. Confirm or challenge the role choice using Reading the answers, and state the reasoning in one short paragraph the user can veto.

2. Artifact 1 - role scorecard. Build from references/scorecard-template.md. Three parts:

   • a one-sentence mission
   • 3-8 measurable outcomes ranked by importance and quantified (e.g. "$1.2M net-new ACV in year one", "40 SQLs/quarter"), never vague responsibilities
   • 5-7 competencies weighted per segment and team stage

   The scorecard doubles as the interview rubric and the ramp milestone map, so quantify outcomes even when the user resists. Avoid the all-around-athlete profile: narrow, deep competence against these outcomes.

3. Artifact 2 - interview loop + question bank.

   • Design the stage sequence for the role and segment: SDR 3-4 stages over days-weeks; AE 4-6 stages; enterprise AE 5-6 stages over weeks, with an inverted funnel (hiring manager in early) when the candidate pool is small and hot.
   • Place the scored work sample at stage 2 or 3, never last. The two placements are identical on value: value: stage 2-3 == last, a genuine tie because it is the same instrument, the same rubric and the same score wherever it sits. They separate only on hours: interviewer hours burned: last > stage 2-3, at 10+ hours per candidate who cannot sell, plus the extra days of time-to-fill that lose candidates to competing offers. Last placement is therefore deleted from the menu, not ranked below stage 2-3; quality gate 5 enforces it.
   • Write 2-3 behavioral plus 1-2 situational questions per scorecard competency with drill-down probes, from references/interview-question-bank.md. Delete every brainteaser, "sell me this pen", "greatest weakness" and "are you coachable" - they have no demonstrated predictive value [peer-reviewed + measured internal analysis at scale].
   • Assign each interviewer a competency, and keep the same interviewers and the same questions across all candidates for the role. Structure, not headcount, is what raises validity; panels add no validity over a single trained interviewer (peer-reviewed) - a panel's value is governance, not measurement.
   • Define the scoring mechanics now: every score needs an evidence quote, scores are submitted independently before any debrief, and the final number is combined mechanically with pre-set weights. The hiring manager decides within pre-committed thresholds.
   • If the user wants a personality assessment in the loop, apply the assessment rule below and load references/assessment-validity-audit.md.
   • Run the compliance pass for the Q7 jurisdictions with references/legal-landscape.md: remove salary-history questions everywhere, publish the range, and flag any AI screening tool for bias-audit and high-risk-AI obligations.

4. Artifact 3 - work-sample design + rubric. Build from references/mock-call-design.md. Design a mock call where the candidate sells the user's product, not their own:

   • a prep pack sent ahead
   • a hard live scenario
   • one unscripted objection injected mid-call
   • one segment re-run after feedback (the coachability test, replacing the useless self-report question)

   Score on a behaviorally-anchored 0-3 rubric with 5-8 categories, frozen for a quarter, with frame-of-reference training for scorers. Score improvisation and adaptability, never polish. Match content to segment: cold call + objection handling for SDR/high-velocity/B2C, discovery depth + multi-threaded deal strategy + territory plan for enterprise.

5. Artifact 4 - 30-60-90 ramp plan. Build from references/ramp-plan-template.md.

   • Day 0-30: product/ICP/tooling training ending in a certification gate (a real exam, pass required); shadowing and sandbox mock calls.
   • Day 31-60: live activity with coaching, every call reviewed, first opportunities.
   • Day 61-90: self-sourced pipeline, first deals, forecast accuracy.

   Add:

   • a stepped quota ramp (25/50/75/100% over successive quarters)
   • a non-recoverable draw for AEs in months 1-3
   • a peer buddy distinct from the manager
   • check-ins in Situation-Behavior-Impact form

   Include the early-warning list and the action rule: a hire missing all three 90-day milestones (certification, coached improvement, self-sourced pipeline) needs a decision, not hope - involuntary exits cluster early and the productive window is short (survey).

6. Run the Quality gate below on all four artifacts. Iterate until it passes.

7. If your harness has persistent memory, store the scorecard, the frozen rubric, the loop design and the ramp milestones - later coaching and review sessions start from them. Without memory, tell the user to keep the artifacts as the canonical pack and re-supply them.

Selection evidence

Two orderings govern the loop and they disagree, so never quote one as if it were the other. Validity says what an instrument buys; efficiency says what to build first with the hours available. Read > as "more of this axis"; the efficiency line is the build order.

Validities are corrected operational values (Sackett et al. 2022 correction, peer-reviewed; earlier figures were overstated).

• efficiency: structured interview > job-knowledge exam > reference check > cognitive test > work sample
• value: structured interview > job-knowledge exam > work sample > cognitive test > reference check
• effort: work sample > structured interview > job-knowledge exam > cognitive test == reference check (tied: nothing to build for either, under an hour to run)
• compliance cost: cognitive test > work sample > reference check > job-knowledge exam == structured interview (tied: content-valid instruments whose only exposure is the questions inside them - neither triggers an audit, a consent duty, or third-party data handling)

Rows sit in efficiency order, not validity order:

Instrument	Validity	Build	Run per candidate	What the row costs
Structured interview	.42	a week - bank, anchors, interviewer training	an hour per interviewer	Highest validity, and the week is reused by every later hire. Validity climbs with structure (.20 unstructured to .57 highly structured), so the build is what buys the coefficient
Job-knowledge exam	.40	a week to write	near-zero, self-marking	Your own product cannot be tested pre-hire, so this lands as the day-30 certification gate (Artifact 4) - it buys a ramp decision, not a hire decision
Reference check	~.26 or lower	near-zero	an hour per finalist

...