── SKILL
coaching-analyst✓ APPROVED

Walk into every 1:1 with the gap already found.

Turns rep data into a specific coaching priority per person, plus the team-level pattern underneath them.

Inside: The per-rep gap logic and the team-pattern rollup.

01 / HOW TO USE

Install it in one line, or paste it in.

1
In Claude Code (one command)
Copy the install line, paste it into your terminal, and restart Claude Code. The skill installs itself to ~/.claude/skills/ and runs automatically when it is relevant.
2
In Claude, ChatGPT, or a Project (no terminal)
Open the file, then upload it to your chat or paste its contents in. A skill is just a markdown file of instructions, so any capable AI can follow it.
New to skills? A skill is a plain-text file that teaches your AI a workflow. Point any capable assistant at it and it follows the steps, on your data.
02 / MAKE IT YOURS

Connect your context. Set it to your motion.

CONNECT YOUR CONTEXT · AND WHY IT HELPS
a CRM

opportunities, stages, forecast categories, close dates, and notes, live.

a meeting or activity tool

meeting quality from transcripts and real follow-up cadence, not just counts.

SET IT TO YOUR MOTION

Map your own methodology to the completeness check. If your framework has named steps or required fields, list them as the methodology fields and the skill counts completeness against those. The method does not assume any one framework.

Set thisWhat it isDefault / Example
CRMWhere opportunities, stages, and notes liveAny CRM with opportunity and activity objects
Methodology fieldsThe fields or steps your completeness check countsYour qualification framework's required fields
Activity sourcesWhere meetings, calls, and follow-ups are loggedMeeting tool + CRM tasks/events
Win-rate segmentsHow you split wins to find where a rep wins vs losesSegmentchanneldeal size
Completeness targetThe team bar for methodology completeness80%
Win-rate flag thresholdWin rate below which a segment gets flagged40%
Multi-thread targetContacts per deal you expect3+
Stage-sticker ruleWhen a deal counts as stuckLonger than the rep's own median in that stage
03 / THE FULL SKILL

Everything the skill does, in full.

── WHAT THIS DOES

This skill turns rep data into per-rep coaching priorities. It reads each rep's pipeline, activity, and win history, then surfaces a specific gap for each person plus the team-level patterns underneath them. It answers "who needs help and on what?" with a named topic per rep, not a vague verdict. It watches the people, not the system.

── THE METHOD
  1. 1
    Per rep, the skill computes:
    • Methodology completeness rate. The share of open opps that pass your check, plus the field the rep misses most.
    • Activity quality. Meeting count times meeting quality (from transcripts) times follow-up cadence. High count of low-quality touches does not score well.
    • Deal velocity. Average days in each stage. Surfaces deals sitting longer than the rep's own median.
    • Win rate by segment. Where the rep wins and loses. Small samples flagged as small.
    • Stale-deal concentration. The share of the book going dark or stale.
    • Coaching theme detection. Patterns that hold across the team, not one rep.

    The team view rolls these up against your targets and names the reps outside the band.

── QUALITY GATES
  • Per-rep gap is specific. Not "needs help with methodology," but "completeness 14% (1 of 7 opps), most-missing step is the decision-dynamics field."
  • Coaching themes are team-level and statistically meaningful. "3 of 5 reps below 40% mid-market win rate" beats "everyone needs to do better."
  • Pre-1:1 brief is forecast-aware. It knows the next forecast call and what the rep called last week.
  • Small samples are labeled with their n.
── OUTPUT (EXAMPLE)
EXAMPLE OUTPUT
COACHING PRIORITIES, Week of June 1

TOP PRIORITY: Rep A
  Methodology completeness: 14% (1 of 7 opps), lowest on team
  Most-missing step: decision-dynamics (5 of 7 opps)
  Stage-stickers: 3 opps in Solution Validation past 21 days (median 9)
  1:1 focus: methodology + multi-thread + activity cadence

Rep B, strong but watch:
  Completeness 87% (highest), win rate 71% (small sample n=7)
  Concern: working hard, but pipeline math does not get there
  1:1 focus: pipeline-creation strategy, not execution

TEAM THEMES:
  1. Completeness target 80%, team avg 56%. Group training.
  2. 3 of 5 reps below 40% mid-market win rate. ICP or messaging check.
  3. Multi-thread averaging 1.4 contacts/deal (target 3+).
── WHERE THE NUMBERS COME FROM

Defaults to re-tune: completeness target 80%, win-rate flag 40%, multi-thread target 3+, stage-sticker rule judged per rep against their own median. Any win rate carries its n so a hot streak is not mistaken for a trend. Activity quality depends on the meeting tool; with none connected, the skill falls back to CRM activity counts and says so.

── EXAMPLE CHAT
EXAMPLE CHAT
You
Who on the team needs coaching this week, and on what specifically?
CLAUDE
Read's done. Your top priority isn't your lowest performer, it's your busiest one. Rep A: methodology completeness 14% (1 of 7 opps), lowest on team most-missing step: decision-dynamics field (5 of 7 opps) Rep B: completeness 87% (highest), win rate 71% (n=7, small sample) Rep A's gap is specific and fixable: they're skipping decision-dynamics on almost every deal, not "struggling generally." Rep B looks like your best rep by win rate, but the sample is only 7 deals, so that number could flip next quarter. One sharpener: connect your meeting tool so activity quality comes from real transcripts instead of raw call counts, which currently can't tell a good call from a long one.
You
What's the team-level theme underneath these two?
CLAUDE
One pattern holds across more than one rep, which is what makes it a theme instead of a coincidence. 3 of 5 reps below 40% mid-market win rate Team avg methodology completeness: 56% vs 80% target That's not an individual coaching problem, it's a segment problem. Worth a group training on mid-market messaging, not five separate 1:1 conversations repeating the same fix. One sharpener: add product-usage data so the team theme can separate "messaging is off" from "we're selling to the wrong accounts."
04 / GO FURTHER

Where an operator takes this next.

The read is step one. Here's where an operator takes it once the manual version proves out.

You built the read once; now it runs itself.

01
Refresh the board before every 1:1, automatically

Point a scheduled Claude task at Salesforce and Gong the morning of 1:1 day and post each rep's brief to their manager in Slack.

02
Catch a stalling rep mid-quarter, not at the QBR

Have the task re-run weekly and flag any rep whose completeness or win rate drops two weeks running.

03
Turn team themes into a training calendar

Feed recurring gaps into a content-brief skill to spin up the enablement session the data is actually asking for.

05 / PART OF A BIGGER PLAY

One skill is the on-ramp.

── GET NEW SKILLS IN YOUR INBOX

One operator-built skill a week. Free.