How do you build a sales skills matrix to guide coaching?

Why This Happens — Diagnose Before You Coach

Most managers coach the rep who's loudest in their head — the one who missed quota last week, the one who just complained — instead of the rep and the skill where coaching moves the most revenue. A skills matrix fixes that by forcing you to separate four different problems that all look like "underperformance":

• Skill — the rep doesn't know *how* to do it (can't run discovery, can't handle the budget objection). This is what a matrix is built for.
• Will — the rep knows how but isn't doing it (motivation, burnout, comp misalignment). A matrix can't fix this; a different conversation can.
• Knowledge — the rep doesn't know the product, the buyer, or the competitor well enough to apply the skill. Enablement gap, not a coaching gap.
• System/territory — the rep's patch, lead flow, or quota is broken. No amount of role-play fixes a bad territory.

The matrix is precise about skill, so before you score anyone, route the symptom to the right cause. Score a rep low on "closing" when the real issue is a starved territory and you'll coach a problem that doesn't exist.

The Coaching Conversation

The matrix only matters if reps help build it and then act on it. Use the GROW model (Goal, Reality, Options, Will) to turn a matrix score into a coaching commitment. Here is the verbatim 1:1 script.

Frame the matrix without making it feel like a performance review:

> "I built a sales skills matrix for the whole team — same eight competencies, same rubric, everybody scored the same way off call recordings. This isn't a ranking against your peers, it's a map of where I can help you get sharper. Want to walk through yours?"

Goal — set the target on one cell:

> "On the matrix you're a strong 3 on discovery but a 2 on multi-threading — you're working single-threaded deals and they're stalling at legal. If we got you to a 3 on multi-threading this quarter, what would that do to your stuck pipeline?"

Reality — pull the evidence from the rubric, not your opinion:

> "Here's what a 2 looks like on the rubric: you ask for one contact and run the deal through them. A 3 is you've named and reached the economic buyer plus one champion on every deal over $25k. I pulled three of your calls — on all three you had one contact. Does that match how it feels to you?"

Options — let the rep generate the moves:

> "What are two things you could do on your next discovery call to open a second thread? … Okay, you said ask for an intro to the budget owner. What's a way to ask that doesn't feel like going over your contact's head?"

Will — lock a specific, observable commitment:

> "So this week: on every deal over $25k, you ask your champion, *'Who else needs to weigh in before this can move forward?'* and you log the second contact in Salesforce. I'll review two of your Gong calls Friday and we'll re-score the cell. Deal?"

Notice the script names the rubric level, cites call evidence, and ends with a measurable behavior — that's the whole loop. The rep should leave knowing exactly which matrix cell you're moving and how you'll both know it moved.

The Coaching Plan / Cadence

Build the matrix once, then run it on a rhythm. A 30/60/90 rollout keeps it from dying as a spreadsheet.

• Days 1–30 — Build and baseline. Define 6–10 competencies with your top reps. Write the four-level rubric for each. Score every rep off two recorded calls plus one live observation. This is your skills assessment baseline.
• Days 31–60 — Coach the cells. Each rep gets one target cell. Weekly 1:1s run the GROW script against that cell. You review one or two Gong/Chorus calls per rep per week tied to the skill.
• Days 61–90 — Re-score and reset. Re-rate the targeted cells off fresh calls. Movement gets celebrated and a new cell gets targeted; no movement triggers the diagnose tree again (was it really a skill gap?).

The engine underneath is a tight weekly loop:

Coach one cell per rep per cycle. Trying to fix four competencies at once produces zero. The matrix makes triage easy: pick the lowest-scoring cell with the highest revenue leverage for that rep's role.

Drills & Role-Play

Scores don't move from feedback alone — reps need reps. Tie each drill to a matrix cell:

• Call-review scorecard. Take a real Gong recording, hand the rep the rubric for the target competency, and have them score themselves before you reveal your score. The gap between their score and yours is the coachable moment.
• Scenario role-play. For a low "objection handling" cell, run a 10-minute live role-play where you play a skeptical CFO and the rep handles the price objection three times, getting tighter each round.
• Blind rubric calibration. Have two reps score the same call independently against the matrix rubric. When their scores diverge, the team learns what each level actually means — and your matrix gets more reliable.
• Skill-of-the-week stand-up. Pick the most common low cell across the team and run a five-minute group drill on it in your sales meeting.

Role-play is uncomfortable; do it anyway and do it short. Ten focused minutes beats a 60-minute "training."

What to Measure

Quota is a lagging indicator. The matrix is built to surface leading indicators that prove the coaching is working before the number does:

• Cell movement — the count of competency cells that rose a level this quarter (the cleanest proof your coaching is landing).
• Behavior change in CRM — e.g., multi-threaded deals as a share of pipeline in Salesforce/Clari when you're coaching multi-threading.
• Conversion by stage — discovery-to-demo or demo-to-proposal rate for reps you're coaching on that stage.
• Ramp time — days to first deal for new hires whose matrix you're actively coaching vs. the historical baseline.
• Win-rate on coached deals vs. uncoached, where you can isolate it.

If cell scores rise but pipeline behavior doesn't, you're inflating scores. If behavior changes but win-rate doesn't, the skill you picked wasn't the constraint — go back to the diagnose tree.

Common Mistakes Managers Make

• Rating everyone a 3. A matrix where everyone scores the same is useless. Force distribution against the rubric evidence, not your affection for the rep.
• Coaching to the deal, not the skill. Saving this week's deal feels productive but builds nothing repeatable. Use the deal as a *example* of the skill cell you're coaching.
• No follow-through. Scoring once and never re-scoring turns the matrix into a dead spreadsheet. The re-score *is* the accountability.
• Coaching everyone the same. A top rep stuck on one cell needs a different conversation than a struggling new hire. The matrix tells you who needs what.
• Confusing low will for low skill. A motivated rep who can't do it needs coaching; a capable rep who won't needs a will conversation or a PIP — not more role-play.
• Letting reps grade their own homework with no calibration. Self-scores drift. Anchor every score to call evidence.

How to Weight Competencies for Maximum Coaching ROI

Not every skill in your matrix deserves equal attention. Weight each competency by its actual impact on closed revenue — typically 30-40% for discovery and qualification, 20-30% for objection handling, 15-25% for closing, and 10-15% for pipeline management. Use historical win-rate data from your CRM to validate these weights: pull the correlation between rep scores on each skill and their quota attainment over the past two quarters. A rep scoring 2/4 on discovery (weighted 35%) is a higher coaching priority than one scoring 2/4 on pipeline hygiene (weighted 10%), even if the raw scores are identical. Update weights annually as your market and product evolve.

Common Matrix Design Mistakes That Undermine Coaching

Three errors consistently sabotage sales skills matrices. First, using vague labels like "good" or "excellent" instead of behavioral anchors — "asks at least three budget questions per call" is actionable, "strong discovery skills" is not. Second, building a matrix with 15+ competencies, which dilutes focus and overwhelms reps; 6-8 is the effective range for a quarterly coaching cycle. Third, scoring from memory rather than recorded evidence — managers overestimate rep ability by 1-2 points when relying on recall. Audit your matrix quarterly by comparing scores against actual call transcripts; if more than 20% of scores shift after review, your calibration process needs tightening.

FAQ

How many competencies should a sales skills matrix have? Six to ten. Fewer and it's too blunt to coach against; more and reps can't focus and you can't score reliably. Cover the deal lifecycle — prospecting, discovery, demo/value, objection handling, multi-threading, negotiation, forecasting/CRM hygiene — and weight toward the skills that move your specific sales motion.

Who should build the matrix — the manager alone or the team? Draft it with your two or three best reps so the rubric reflects what "great" actually looks like on your team, then own the scoring yourself off call evidence. Reps co-building the rubric increases buy-in; the manager scoring keeps it consistent and defensible.

How is a skills matrix different from a competency matrix or a scorecard? They overlap. A competency matrix and a skills matrix are effectively the same artifact — reps × skills × rubric ratings. A scorecard is usually a single-call evaluation form; the matrix aggregates many scorecards over time into a coaching map. Use scorecards to feed the matrix.

How often should I re-score the matrix? Quarterly for a full re-score, with lightweight cell-level updates after each coaching cycle. Quarterly cadence shows real movement without turning scoring into a full-time job, and it lines up with most pipeline review rhythms.

Can AI tools build or maintain the matrix for me in 2027? AI call-coaching tools like Gong and Chorus can surface objective signals — talk ratio, question rate, competitor mentions, multi-threading — that make scoring faster and less biased. They don't replace your judgment on the rubric, but they cut the time to a defensible skills assessment dramatically and let you score off real calls instead of memory.

What if a rep scores low everywhere? Don't coach ten cells at once. Pick the single highest-leverage skill for their role and stage, fix that, then move to the next. If scores stay flat across cycles despite genuine effort, run the diagnose tree — you may be looking at a will, fit, or territory problem that coaching can't solve.

Bottom Line

A sales skills matrix turns coaching from a feeling into a system: define 6–10 competencies, write a four-level rubric, score every rep off real call evidence, and coach the one cell per rep with the highest revenue leverage. Re-score quarterly so the matrix shows movement. The matrix diagnoses; the GROW 1:1 acts; the re-score holds everyone accountable.

Related on PULSE

• [Top 10 questions to boost a sales rep's negotiation skills](/knowledge/cg0871)
• [Top 10 questions to assess a sales rep's pipeline management skills](/knowledge/cg0857)
• [How do you certify sales skills before reps face real buyers?](/knowledge/cg0179)
• [How do you use role-play to coach sales skills effectively?](/knowledge/cg0171)
• [Top 10 questions to improve a rep's demo storytelling skills](/knowledge/cg0828)
• [Top 10 questions to develop a rep's upselling and cross-selling skills](/knowledge/cg0893)

Sources

• Gong Labs — What separates top sales reps (call analytics research)
• Harvard Business Review — The Right Way to Coach Salespeople
• RAIN Group — Sales Coaching and Skills Development
• Sales Hacker — How to Build a Sales Competency Framework
• Winning by Design — Sales Skills and Frameworks
• Sandler — Sales Coaching Resources
• Salesforce Blog — Sales Coaching Best Practices
• Richardson Sales Performance — Sales Skills Assessment

*Sales coaching for a sales skills matrix — how to build a sales competency matrix to guide coaching, sales manager coaching guide, rep skills assessment rubric framework, and a sales coaching playbook for 2027.*