okr-definition

OKR Definition

You create Objectives and Key Results at company, team, and individual levels. You research industry OKR patterns and best practices yourself -- do not ask the user for data they would need to look up. Only ask the user for decisions and confirmations.

This skill complements vision-crafting (which defines mission, vision, and strategic priorities) by translating strategic intent into measurable objectives and key results. It also feeds into theme-roadmapping (which organizes execution into themes and initiatives).

Metadata

Field	Value
Name	okr-definition
Version	1.0.0
Primary category	planning
Secondary category	assessment
Output mode	human_readable
Mixins	diagram-rendering, autonomous-research

When to use

• Defining OKRs for a company, team, or individual
• Cascading OKRs across organizational levels
• Assessing and improving existing OKRs
• Translating a vision or strategy into measurable goals
• Setting up a CFR (Conversations, Feedback, Recognition) companion framework

When not to use

• Defining project tasks or work breakdown structures (use task-planning skills)
• Creating KPIs without the OKR framework (use metrics/dashboard skills)
• Strategic planning without measurable outcomes (use vision-crafting)
• Performance reviews or individual evaluations

Required input

Field	Description
Organization/team/product context	What entity the OKRs are for

Optional input

Field	Description	Default
Vision/strategy document	Path to vision-crafting output or strategy doc	Will identify strategic context itself
OKR level	Company, team, or individual	Company
Cadence	Annual or quarterly	Quarterly
Existing OKRs	Current OKRs to assess or build upon	None

Input schema

context:
  type: string
  required: true
  description: Organization, team, or product name and context

vision_input:
  type: file_path | string
  required: false
  description: Vision-crafting output or strategy document

okr_level:
  type: string
  enum: [company, team, individual]
  default: company

cadence:
  type: string
  enum: [annual, quarterly]
  default: quarterly

existing_okrs:
  type: file_path | string
  required: false
  description: Current OKRs to assess or extend

render_mode:
  type: string
  enum: [code, image]
  default: code
  dependency_if_image: "@mermaid-js/mermaid-cli (mmdc)"

Processing rules

Phase 1 -- Setup

Input handling

Follow shared foundation SS7 -- interview mode. When input is missing or insufficient, interview to gather at minimum:

Dimension	Required	Default
Organization/team/product context	Yes	--
OKR level	No	Company
Cadence	No	Quarterly
Vision/strategy input	No	Will research context itself

Exit interview when: Context is clear enough to define objectives and key results.

1. Collect input

Accept one of:

• A company, team, or product name with context
• A file path to a vision-crafting report or strategy document
• Pasted strategic content (vision, mission, priorities)
• No input or vague input -> enter interview mode

2. Detect scope

From the input (or interview results), identify:

• Entity: What organization/team/product the OKRs serve
• Level: Company, team, or individual
• Cadence: Annual or quarterly
• Import source: Vision-crafting output, strategy doc, or none

3. Confirm scope

**Entity**: [name]
**OKR level**: [company/team/individual]
**Cadence**: [annual/quarterly]
**Strategic input**: [imported from vision-crafting / will research]

Ask the user to confirm or adjust. Ask diagram render mode and output path per the diagram-rendering and autonomous-research mixins.

Phase 2 -- Research

Use WebSearch and WebFetch per the autonomous-research mixin.

2a. Industry OKR patterns

Research OKR patterns for this domain/industry:

• Common objectives for similar organizations
• Typical key result metrics and benchmarks
• Industry-specific measurement approaches

2b. Best practices

Research OKR best practices relevant to the context:

• Scoring calibration norms (what 0.7 means for committed vs aspirational)
• Common pitfalls in this industry/domain
• Cadence and review rhythm recommendations

Phase 3 -- Strategic Context

If vision/strategy input is provided

Import strategic context:

• Read the vision-crafting output or strategy document
• Extract: mission, vision, strategic priorities, values
• Identify current challenges and opportunities
• Map strategic priorities to potential objective areas

If no strategic input

Identify through research:

• Organization's stated mission and vision (from website, public sources)
• Industry context and competitive landscape
• Likely strategic priorities based on domain
• Current market challenges and opportunities

Present strategic context summary for user confirmation.

Phase 4 -- Objective Definition

Define 3-5 objectives per level.

Objective quality criteria

Each objective must be:

• Qualitative: Describes a desired future state, not a metric
• Inspirational: Motivates action and ambition
• Action-oriented: Starts with a verb or implies clear direction
• Time-bound: Fits the declared cadence (annual/quarterly)

Objective classification

Type	Description	Expected score
Committed	Must achieve -- failure indicates planning or execution problems	1.0 expected
Aspirational	Stretch goals -- reaching 0.7 is strong performance	0.7 expected
Learning	Experimental -- success means validated learning	0.7 expected, pivot acceptable

Objective table

#	Objective	Type	Strategic priority link	Level
O1	[qualitative, inspirational statement]	Committed	[priority]	Company
O2	[qualitative, inspirational statement]	Aspirational	[priority]	Company

Present objectives for user confirmation before proceeding to key results.

Phase 5 -- Key Result Definition

Define 2-5 key results per objective.

Key result quality criteria

Each key result must be:

• Quantitative: Has a number or measurable outcome
• Measurable: Can be objectively verified
• Time-bound: Has a deadline within the cadence
• Has baseline: Current state is documented
• Has target: Desired end state is specific

Scoring criteria

Each key result uses a 0-1.0 scale:

• 0.0: No progress
• 0.3: Minimal progress
• 0.5: Halfway / some meaningful progress
• 0.7: Strong progress (aspirational target)
• 1.0: Full achievement (committed target)

Key result table

Objective	KR #	Key Result	Baseline	Target	Scoring method	0.3	0.5	0.7	1.0
O1	KR1.1	[quantitative statement]	[current]	[target]	[method]	[threshold]	[threshold]	[threshold]	[threshold]

Phase 6 -- Alignment Check

Cascade validation

For multi-level OKRs:

• Every team OKR must trace to at least one company OKR
• Every individual OKR must trace to at least one team OKR
• No orphan OKRs at any level

Conflict detection

• No two objectives at the same level should conflict (e.g., "reduce costs" vs "increase hiring")
• Key results across teams should not compete for the same resource without acknowledgment
• Flag potential tensions and suggest resolution

Alignment map

Team/Individual OKR	Traces to Company OKR	Alignment strength
[team objective]	O1: [company objective]	Strong / Moderate / Weak

Phase 7 -- Quality Validation

Score overall OKR quality (0-100) by checking against 20 common OKR mistakes:

#	Mistake	Detection	Severity	Points deducted
1	Output not outcome	KR describes activity, not result	Critical	-10
2	Too many OKRs	> 5 objectives per level	Warning	-5
3	Too many KRs	> 5 KRs per objective	Warning	-3
4	Sandbagging	All KRs easily achievable, no stretch	Warning	-5
5	No baseline	KR has no current-state reference	Warning	-3
6	Not measurable	KR cannot be objectively verified	Critical	-10
7	Not time-bound	No deadline within cadence	Warning	-3
8	No alignment	Team OKR doesn't trace to company OKR	Critical	-10
9	Conflicting OKRs	Two OKRs work against each other	Critical	-10
10	Binary KR	KR is yes/no with no gradient	Warning	-3
11	Business-as-usual	OKR describes routine work, not improvement	Warning	-5
12	No aspirational OKRs	All OKRs are committed, no stretch	Info	-2
13	All aspirational	No committed OKRs, everything is stretch	Warning	-5
14	Metric gaming risk	KR incentivizes wrong behavior	Warning	-5
15	Missing strategic link	Objective doesn't connect to strategy	Warning	-5
16	Vague objective	Objective is too generic to inspire action	Warning	-5
17	Lagging-only KRs	No leading indicators, only trailing metrics	Info	-2
18	Duplicate measurement	Same metric appears in multiple KRs	Info	-2
19	Uncontrollable KR	KR depends on external factors team cannot influence	Warning	-3
20	No scoring rubric	KR has no defined 0.3/0.5/0.7/1.0 thresholds	Warning	-3

Quality score

Score range	Rating
90-100	Excellent OKR quality
75-89	Good -- minor improvements needed
50-74	Fair -- significant issues to address
< 50	Poor -- major rework recommended

Phase 8 -- CFR Framework

For each objective, define the CFR companion:

Conversations

Objective	Conversation topic	Frequency	Participants	Purpose
O1	[topic]	[weekly/biweekly/monthly]	[roles]	Check-in / problem-solving / course-correction

Feedback

Objective	Feedback mechanism	Trigger	Direction
O1	[mechanism]	[when to give feedback]	Upward / downward / peer

Recognition

Objective	Recognition trigger	Recognition type
O1	[what achievement triggers recognition]	Public / private / formal / informal

Phase 9 -- Diagrams

Diagram 1: OKR Alignment Tree (flowchart)

flowchart TD
    classDef company fill:#1565C0,stroke:#333,color:#fff
    classDef team fill:#2E7D32,stroke:#333,color:#fff
    classDef individual fill:#F57F17,stroke:#333,color:#fff

    CO1["O1: Company Objective"]:::company
    TO1["O1.1: Team Objective"]:::team
    TO2["O1.2: Team Objective"]:::team
    IO1["O1.1.1: Individual Objective"]:::individual

    CO1 --> TO1
    CO1 --> TO2
    TO1 --> IO1

Shows cascade from company to team to individual objectives.

Diagram 2: Key Results Progress Dashboard (xychart-beta)

xychart-beta
    title Key Results Target vs Baseline
    x-axis ["KR1.1", "KR1.2", "KR2.1", "KR2.2", "KR3.1"]
    y-axis "Score" 0 --> 1.0
    bar [0.2, 0.1, 0.3, 0.0, 0.15]
    bar [1.0, 0.7, 1.0, 0.7, 1.0]

Two bars per KR: baseline (current) and target score.

Diagram 3: Quality Scorecard (xychart-beta)

xychart-beta
    title OKR Quality by Dimension
    x-axis ["Measurability", "Alignment", "Ambition", "Clarity", "Completeness"]
    y-axis "Score" 0 --> 100
    bar [85, 90, 70, 80, 75]

Shows quality scores broken down by dimension.

Render diagrams per the diagram-rendering mixin.

File naming:

• okr-alignment-tree.mmd / .png
• key-results-dashboard.mmd / .png
• quality-scorecard.mmd / .png

Phase 10 -- Report Assembly

Assemble the complete report:

# OKR Definition: [Entity Name]

**Date**: [date]
**Entity**: [name]
**Level**: [company/team/individual]
**Cadence**: [annual/quarterly]
**Objectives**: [count]
**Key Results**: [count]
**Quality score**: [0-100] -- [rating]

## Executive Summary
[Key findings: quality score, objective types breakdown, alignment status, top 3 recommendations]

## Strategic Context
[Phase 3 strategic context: mission, vision, priorities, challenges]

## OKR Table
[Phase 4 + 5 combined: objectives with their key results, baselines, targets, scoring rubrics]

## Alignment Map
[Phase 6 cascade validation + alignment tree diagram]

## Quality Validation
[Phase 7 mistake detection + quality scorecard diagram]

## Key Results Dashboard
[Phase 9 progress dashboard diagram]

## CFR Framework
[Phase 8 conversations, feedback, recognition tables]

## Recommendations
[Prioritized actions traced to specific findings]

## Sources
[Numbered list of web sources]

## Assumptions & Limitations
[Explicit list]

Present for user approval. Save only after explicit confirmation.

Generation rules

Per the autonomous-research mixin, plus:

• Objectives: Must be qualitative, inspirational, action-oriented -- never a metric as an objective
• Key Results: Must be quantitative, measurable, with baseline and target -- never vague outcomes
• Scoring: Must define 0.3/0.5/0.7/1.0 thresholds for every KR -- never leave scoring undefined
• Alignment: Every sub-level OKR must trace to a parent -- never create orphan OKRs
• Classification: Every objective must be classified as committed, aspirational, or learning
• Specificity: "Increase NPS from 32 to 50 by Q2" not "improve customer satisfaction"
• Language: Respond and generate in the user's language unless specified otherwise

Output contract

The report must contain all sections listed in Phase 10. Sections may not be empty -- omit the section header if there is nothing to report for that section.

Failure behavior

Situation	Behavior
No context provided	Enter interview mode -- ask what entity to define OKRs for
Context too vague	Enter interview mode -- ask targeted questions about the organization/team
Vision input malformed	Ask user to verify, attempt partial import
Cannot identify strategic priorities	Report limitation, produce OKRs with explicit assumptions labeled
Existing OKRs provided for assessment	Assess against the 20 mistakes, provide improvement recommendations
Single person / solo founder	Adapt to individual OKRs, skip cascade validation
Non-profit / mission-driven	Adapt objectives to impact metrics, not revenue
Cross-functional initiative	Create shared OKRs with clear ownership per KR
mmdc / web search failures	See diagram-rendering and autonomous-research mixins
Out-of-scope request	"This skill defines and validates OKRs. [Request] is outside scope."

Quality checks

Before presenting output, verify:

[] 3-5 objectives defined per level
[] 2-5 key results per objective with baselines and targets
[] Every objective classified (committed/aspirational/learning)
[] Every KR has 0.3/0.5/0.7/1.0 scoring thresholds
[] Alignment validated -- no orphan OKRs
[] No conflicting OKRs at the same level
[] All 20 OKR mistakes checked
[] Quality score calculated (0-100)
[] CFR framework defined for each objective
[] Recommendations traced to specific findings
[] All Mermaid diagrams render valid syntax (per diagram-rendering mixin)
[] Sources listed for claims (per autonomous-research mixin)
[] Assumptions labeled (per autonomous-research mixin)

Examples

Normal cases

Example 1: SaaS company quarterly OKRs

• Input: "Define Q2 OKRs for CloudMetrics, a B2B SaaS analytics platform with 500 customers"
• Expected: Company-level quarterly OKRs with 3-5 objectives covering growth, product, and customer success. Key results with SaaS-specific metrics (MRR, churn, NPS). Committed and aspirational mix.

Example 2: With vision-crafting input

• Input: "/documentation/cloudmetrics/vision-crafting/vision-report.md"
• Expected: OKRs directly linked to the vision's strategic priorities. Alignment map shows clear trace from each objective to a strategic priority. No orphan objectives.

Example 3: Engineering team OKRs

• Input: "Define OKRs for the Platform Engineering team at CloudMetrics. Company OKR: Achieve 99.95% uptime."
• Expected: Team-level OKRs cascading from the company objective. KRs on reliability, incident response, infrastructure metrics. Alignment map traces each team OKR to the company OKR.

Example 4: Startup annual OKRs

• Input: "Annual OKRs for FreshBite, a pre-Series A food delivery startup in Amsterdam, 3 months post-launch"
• Expected: Company-level annual OKRs focused on product-market fit, growth, and fundraising. Heavy on aspirational objectives. Baselines near zero for many KRs.

Example 5: Assess existing OKRs

• Input: Pasted set of current OKRs for review
• Expected: Assessment against 20 mistakes. Quality score with breakdown. Specific improvement recommendations per objective and KR. No new OKRs generated unless requested.

Edge cases

Example 6: Solo founder

• Input: "OKRs for me, solo founder of a side project"
• Expected: Individual-level OKRs only. Skip cascade validation. Adapt CFR to self-accountability mechanisms. Fewer objectives (2-3).

Example 7: Non-profit

• Input: "Quarterly OKRs for WaterFirst, a clean water NGO operating in East Africa"
• Expected: Objectives focused on impact (lives affected, wells built) rather than revenue. KRs include donor and operational metrics alongside impact metrics.

Example 8: Cross-functional initiative

• Input: "OKRs for a GDPR compliance initiative spanning Legal, Engineering, and Product teams"
• Expected: Shared objectives with KRs assigned to specific teams. Alignment shows how each team contributes. Potential conflicts between speed and compliance flagged.

Failure cases

Example 9: No context

• Input: "Define OKRs"
• Expected: Interview mode. Ask what entity to define OKRs for, what level, what cadence. Do not generate OKRs without context.

Example 10: Out of scope

• Input: "Create a project plan with timelines and resource allocation"
• Expected: "This skill defines and validates OKRs. Project planning with timelines and resource allocation is outside scope. Consider using a project planning skill."