Ceos People

ceos-people

Evaluate whether team members are the right people (Core Values alignment) in the right seats (GWC: Get it, Want it, Capacity to do it). Manage people evaluations, run quarterly reviews, and flag below-the-bar situations for action.

When to Use

• "Evaluate [person]" or "run the people analyzer for [name]"
• "Are we right people, right seats?" or "show our people evaluations"
• "Quarterly people review" or "let's review the team"
• "Is [person] in the right seat?" or "GWC check on [name]"
• "Who's below the bar?" or "show below-bar team members"
• Any discussion about team fit, Core Values alignment, or seat assignment

Context

Finding the CEOS Repository

Search upward from the current directory for the .ceos marker file. This file marks the root of the CEOS repository.

If .ceos is not found, stop and tell the user: "Not in a CEOS repository. Clone your CEOS repo and run setup.sh first."

Sync before use: Once you find the CEOS root, run git -C <ceos_root> pull --ff-only --quiet 2>/dev/null to get the latest data from teammates. If it fails (conflict or offline), continue silently with local data.

Key Files

Person File Format

Each person is a markdown file at data/people/firstname-lastname.md with YAML frontmatter:

name: "Brad Feld"
seat: "Visionary"
core_values:
  # Each Core Value from vision.md, rated +, +/-, or -
status: right_person_right_seat  # right_person_right_seat | below_bar | wrong_seat | evaluating
gwc:
  get: true       # true | false | null
  want: true
  capacity: true
last_evaluated: "2026-01-15"
created: "2026-01-02"
departed: false

File naming: firstname-lastname.md — lowercase, hyphenated. Person-centric (survives role changes).

Status Values

Core Values Rating

Three strikes rule: Three or more +/- or - ratings = "wrong person" (Core Values misalignment). This is a critical flag that requires action.

GWC Dimensions

All three must be true for "right seat." Any single false = wrong seat.

Process

Mode: Evaluate

Use when evaluating a specific person against Core Values and GWC.

Step 1: Identify the Person

Ask for the person's name. Check if data/people/firstname-lastname.md already exists.

• Exists: Read the file, show current evaluation, ask: "Re-evaluate or update notes?"
• New person: Create from templates/people-analyzer.md

Step 2: Core Values Evaluation

Read Core Values from data/vision.md. For each Core Value, ask the user to rate: +, +/-, or -.

Display a rating table as you go:

Core Values Evaluation — [Person Name]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

| Core Value    | Rating | Notes           |
|---------------|--------|-----------------|
| Integrity     | +      | Consistently    |
| Innovation    | +/-    | Room to grow    |
| Transparency  | +      |                 |
| Grit          | -      | Avoids hard work|

Step 3: Three Strikes Detection

After all Core Values are rated, count +/- and - ratings:

• 3+ negative ratings (+/- or -): Flag immediately:

  ⚠️  THREE STRIKES — Core Values misalignment detected.
  [Person] has 3+ values rated +/- or -. This signals "wrong person."
  Action required: coaching conversation, role change, or exit plan.

• 1-2 negative ratings: Note but continue: "Some values need attention. Continue to GWC."
• All +: "Strong Core Values alignment. Right person."

Determine right_person: All or mostly + ratings = yes. Three strikes = no.

Step 4: GWC Evaluation

Read the person's current seat from data/accountability.md. If the person owns multiple seats, evaluate GWC for each seat separately.

For each seat, ask three binary questions:

1. Get it? Does [person] truly understand the [seat] role? (yes/no)
2. Want it? Does [person] genuinely want to do [seat] work? (yes/no)
3. Capacity? Does [person] have the capacity to excel at [seat]? (yes/no)

Display the result:

GWC — [Person Name] as [Seat Name]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Get it:     ✓ Yes
Want it:    ✓ Yes
Capacity:   ✗ No — struggling with volume

Right seat? No (Capacity gap)

Determine right_seat: All three must be true. Any false = wrong seat.

Step 5: Suggest Status

Based on the evaluation:

Present the suggestion: "Based on this evaluation, I'd suggest [status]. Do you agree, or would you set it differently?"

Always let the user confirm or override. Status is a leadership judgment call, not a formula.

Step 6: Write the File

Show the complete evaluation file before writing. Ask: "Save this evaluation?"

Update last_evaluated to today's date. Add a dated entry to the Evaluation History section.

Step 7: Below-Bar Action

If status is below_bar or wrong_seat, offer:

"[Person] is below the bar. Would you like to create an issue for a 30-day action plan? This will create a file in data/issues/open/ for IDS discussion."

If yes, use the issue template pattern from data/issues/open/ to create an issue with:

• Title referencing the person and the gap
• 30-day timeline
• Specific actions from the evaluation

Mode: Review

Use when reviewing the current state of all people evaluations.

Step 1: Read All Evaluations

Read all files from data/people/ (exclude alumni/ subdirectory). Parse the YAML frontmatter for each person.

If no files exist: "No people evaluations found. Run an Evaluate for your first team member."

Step 2: Display Summary Table

People Analyzer — Team Overview
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

| Name          | Seat        | Status                  | Last Evaluated | Flag    |
|---------------|-------------|-------------------------|----------------|---------|
| Brad Feld     | Visionary   | right_person_right_seat | 2026-01-15     |         |
| Sarah Chen    | Integrator  | right_person_right_seat | 2026-01-15     |         |
| Mike Torres   | VP Sales    | wrong_seat              | 2025-12-01     | ⚠️ Seat |
| Alex Kim      | VP Eng      | below_bar               | 2025-11-15     | 🔴 Bar  |
| Jamie Lee     | Marketing   | evaluating              | 2026-01-28     | 🆕 New  |

The Bar: 3/5 (60%) at or above — Target: 80%+

Step 3: Highlight Issues

Flag the following:

• Below bar: 🔴 — requires action plan
• Wrong seat: ⚠️ — person is right, seat is wrong (find a better fit)
• Stale evaluation: If last_evaluated is > 120 days ago, flag: 📅 Overdue
• Evaluating: 🆕 — new hire or incomplete evaluation

Step 4: Bar Percentage

Calculate: (right_person_right_seat count) / (total evaluated, excluding "evaluating").

If below 80%: "Below the 80% target. Consider bringing people discussions to the next L10."

Step 5: Drill Down

Ask: "Want to drill into any person, or run a new Evaluate?"

Mode: Quarterly

Use for the formal quarterly review of all seats against the Accountability Chart.