The Product Competency Model
A product competency framework covering customer insight, strategy, execution, leadership, commercial judgement, and the AI fluency modern roles require.
On this page
TL;DR
- Five domains: Product Strategy, Discovery & Insight, Execution & Delivery, Influencing People, and AI Fluency. Product judgement depends on all five.
- AI Fluency now covers six sub-competencies: architecture literacy, evaluation and quality, AI UX, economics and pricing, governance and risk, and builder skills.
- Use the proficiency levels (Foundational through Expert) for self-assessment, coaching, and career conversations. Not as a checklist.
This model describes the skills a product manager needs to build excellent products when AI is part of the working environment. It is not a job description or a performance rubric. It is a shared vocabulary for identifying strengths, targeting growth areas, and structuring development conversations.
The first four domains cover well-established PM competencies. They're included for completeness but kept concise. Domain 5, AI Fluency, gets the depth it deserves. The skillset shift that makes AI Fluency non-negotiable is covered in the product builder chapter, which describes the PM-to-builder transition and the Prompt, Build, Eval loop that underpins it.
The five domains
| Domain | Focus |
|---|---|
| Product Strategy | Connecting daily work to long-term business value and commercial outcomes. |
| Discovery & Insight | Uncovering customer problems through qualitative and quantitative evidence. |
| Execution & Delivery | Translating ideas into shipped, high-quality products with speed and discipline. |
| Influencing People | Improving decisions through context, trust, constructive disagreement, and clear narrative. |
| AI Fluency | Building and managing products where AI is the medium, not a feature bolted on. |
1. Product Strategy
The bridge between company vision and the products you build.
Business and commercial acumen
Connecting product decisions to revenue, margin, and growth. Know how the company makes money. Articulate your product's contribution to the bottom line.
| Level | In practice |
|---|---|
| Foundational | Can link a feature to a business metric and explain why it matters. |
| Proficient | Builds business cases with financial projections. Defends trade-offs with data. |
| Advanced | Owns P&L for a product portfolio. Balances growth, innovation, and profitability. |
| Expert | Shapes business strategy for new markets, including pricing and business model design. |
Market and competitive analysis
Identifying market problems, mapping the competitive landscape, articulating a differentiated position.
| Level | In practice |
|---|---|
| Foundational | Can name the top competitors and describe their key differentiators. |
| Proficient | Maintains deep competitive knowledge. Uses it to inform roadmap priorities. |
| Advanced | Anticipates market shifts and adjusts portfolio strategy proactively. |
| Expert | Identifies market gaps and creates new opportunities through novel product strategies. |
Vision and roadmapping
Articulating a compelling long-term vision and translating it into a prioritised, outcome-driven roadmap.
| Level | In practice |
|---|---|
| Foundational | Can articulate the vision for their area and translate goals into a prioritised backlog. |
| Proficient | Defines and owns a multi-quarter roadmap for a product line. |
| Advanced | Develops a unified vision and cohesive roadmap for a portfolio of products. |
| Expert | Creates multi-year strategy guiding investment and execution across the company. |
2. Discovery & Insight
Customer insight
The practice of deeply understanding the people you're building for.
| Level | In practice |
|---|---|
| Foundational | Participates in customer research. Can cite the top problems with evidence. |
| Proficient | Leads end-to-end discovery with evidence-based understanding of needs and motivations. |
| Advanced | Establishes feedback loops with strategic customers. Coaches other PMs on research. |
| Expert | Recognised authority on the customer and market. Evangelises emerging problems. |
Data fluency
Using quantitative and qualitative data to generate actionable insights, not just dashboards.
| Level | In practice |
|---|---|
| Foundational | Defines success metrics for a feature. Uses basic analytics tools competently. |
| Proficient | Defines coherent metrics for a product line. Tells compelling stories with data. |
| Advanced | Establishes portfolio-level KPIs. Drives accountability for outcomes across teams. |
| Expert | Defines new business metrics reflecting long-term value. Coaches the organisation on data literacy. |
Solution shaping
Translating customer problems into intuitive, well-scoped solutions.
| Level | In practice |
|---|---|
| Foundational | Collaborates with design and engineering to define requirements for a user flow. |
| Proficient | Leads solution design for a product line, balancing user needs with technical constraints. |
| Advanced | Guides consistent solution quality across a portfolio. |
| Expert | Defines architectural and design principles that enable the long-term product vision. |
3. Execution & Delivery
Requirements craft
Translating problems into clear, actionable, testable requirements.
| Level | In practice |
|---|---|
| Foundational | Writes clear user stories with well-defined acceptance criteria. |
| Proficient | Scopes complex, cross-functional initiatives into shippable increments. |
| Advanced | Establishes requirements standards across teams. Mentors other PMs. |
| Expert | Defines requirements for multi-year strategic initiatives with ambiguous scope. |
Delivery leadership
Guiding cross-functional teams through the development lifecycle.
| Level | In practice |
|---|---|
| Foundational | Keeps team and manager informed of progress and blockers. |
| Proficient | Proactively manages stakeholder expectations. Drives end-to-end delivery. |
| Advanced | Aligns senior leaders on shared vision for a portfolio of products. |
| Expert | Builds C-suite trust and alignment for complex, multi-year programmes. |
Launch and lifecycle management
Getting products out the door and managing them once they are live.
| Level | In practice |
|---|---|
| Foundational | Collaborates with GTM teams for launch readiness. Monitors key metrics post-launch. |
| Proficient | Leads successful launches. Optimises based on data and customer feedback. |
| Advanced | Oversees launch strategy for a portfolio. Ensures consistent practices. |
| Expert | Defines the organisation's approach to launches. Guides high-risk, high-visibility releases. |
4. Influencing People
Influence is the discipline of improving decisions across people with different goals, context, authority, and risk. Influence Is Internal Product Discovery provides the operating method behind this competency.
Stakeholder management
Understanding how other people make decisions and building trusted relationships across the organisation. Product leaders connect the work to the goals, incentives, constraints, and evidence that matter to each decision-maker.
| Level | In practice |
|---|---|
| Foundational | Keeps the immediate team and manager informed. Asks what context other functions need before a decision. |
| Proficient | Builds strong cross-functional relationships, surfaces disagreement early, and adapts evidence to the stakeholder's decision without distorting it. |
| Advanced | Understands executive and board-level goals, pre-aligns consequential decisions, and influences senior leaders through domain authority and constructive challenge. |
| Expert | Operates as a trusted advisor to the C-suite. Improves the decision system across functions rather than relying on personal persuasion. |
Narrative and communication
Crafting stories that move people to action.
| Level | In practice |
|---|---|
| Foundational | Can articulate product progress clearly and concisely. |
| Proficient | Uses narrative memos to align stakeholders on complex initiatives. |
| Advanced | Synthesises and communicates vision and strategy for a portfolio to senior leadership. |
| Expert | Coaches executives through compelling, data-driven narratives that shape company direction. |
Decision framing and alignment
Helping a group make the right decision rather than seeking approval for a preferred solution. This includes restoring context, exposing assumptions, stating trade-offs, and recommending what should stop.
| Level | In practice |
|---|---|
| Foundational | States the decision required, the customer problem, available evidence, and the main uncertainty. Separates facts from recommendations. |
| Proficient | Frames options and trade-offs, asks what led others to a conflicting view, and updates the recommendation when stronger evidence appears. |
| Advanced | Aligns leaders on portfolio choices, kills weak work, and connects local product decisions to company strategy, risk, and capacity. |
| Expert | Designs decision forums, escalation paths, and strategy communication that let teams move quickly without losing coherence. |
5. AI Fluency
AI fluency is a baseline expectation for a product manager whose product or workflow uses models, agents, or AI-enabled data systems. The required depth depends on the role, but delegating all understanding to engineering is no longer sufficient.
This domain covers six sub-competencies. Each one has its own proficiency table.
5.1 AI architecture literacy
Understanding model harnesses, context, tools, permissions, state, evaluation boundaries and when routing is justified. You don't need to build every component, but you do need enough architectural literacy to make informed product decisions.
| Level | In practice |
|---|---|
| Foundational | Understands the difference between a single model call and a multi-step workflow. Can explain what an API, a prompt, and a context window are. |
| Proficient | Can diagnose a simple AI system and design a bounded multi-step workflow. Understands context, tool and permission boundaries, and can decide when routing adds enough quality or resilience to justify its cost and complexity. |
| Advanced | Guides shared harness and platform patterns across a product portfolio. Makes informed decisions about tools, integrations, model substitutability, state, evaluation boundaries, and managed versus self-hosted infrastructure. |
| Expert | Defines organisational AI architecture strategy. Connects product requirements to platform, security and operating constraints, and guides build, buy and partner decisions as the model market changes. |
5.2 Evaluation and quality
Building and maintaining eval suites, monitoring production quality, detecting drift, and running regression tests. Evals are day-one infrastructure, not a post-launch afterthought.
| Level | In practice |
|---|---|
| Foundational | Understands why AI products need structured evaluation beyond manual testing. Can review eval results and interpret pass/fail rates. |
| Proficient | Designs eval frameworks with seed examples drawn from real production failures. Defines quality rubrics for model outputs. Monitors key quality metrics. |
| Advanced | Establishes eval infrastructure across a product portfolio. Implements drift detection, regression suites, and automated quality gates in CI/CD. Traces agent paths, not just final outputs. |
| Expert | Treats the eval suite as competitive IP. Builds data flywheels where production feedback continuously strengthens evals. Defines quality standards that shape the organisation's AI product culture. |
5.3 AI UX and interaction design
Designing experiences where AI is the medium. Inline assistance vs. destination products. Generative UI. The copilot-to-autopilot spectrum. Handling uncertainty, latency, and errors gracefully.
| Level | In practice |
|---|---|
| Foundational | Understands the difference between deterministic and probabilistic UX. Can articulate why AI products need different interaction patterns than traditional software. |
| Proficient | Designs AI experiences that handle uncertainty well through calibrated risk signals, supporting evidence, graceful fallbacks and human checkpoints. Knows when to use inline AI versus a dedicated AI surface. |
| Advanced | Guides AI UX strategy across a portfolio. Makes informed decisions about the copilot-to-autopilot spectrum, generative UI, and progressive autonomy. Designs for the 90th-percentile failure case, not just the happy path. |
| Expert | Defines the organisation's AI interaction design principles. Shapes industry thinking on trust calibration, transparency patterns, and the boundary between assistance and automation. |
5.4 AI economics and pricing
Modelling inference costs (COGS), understanding margin structures, pricing AI features, and navigating the cannibalisation paradox where AI can simultaneously increase value and reduce billable usage.
| Level | In practice |
|---|---|
| Foundational | Understands that AI features carry variable inference costs. Can read a cost-per-token breakdown and explain why it matters for product decisions. |
| Proficient | Models inference COGS for a product. Understands the margin trap (high gross revenue, negative unit economics at scale). Designs pricing that accounts for variable cost structures: per-seat, usage-based, outcome-based. |
| Advanced | Manages AI economics across a portfolio. Navigates the cannibalisation paradox (AI automates tasks users currently pay for). Models the audit tax (compliance and monitoring overhead that compounds with each AI feature). |
| Expert | Defines the organisation's AI pricing and monetisation strategy. Builds margin advantages through model routing, caching, and cost optimisation. Shapes portfolio investment decisions using total cost of AI ownership, not just inference spend. |
5.5 AI governance and risk
Risk-tiered classification, data governance, security, compliance, and responsible AI practices. Not a legal exercise. A product discipline.
| Level | In practice |
|---|---|
| Foundational | Can articulate the core risks in their product's AI features: data privacy, bias, hallucination, security. Knows that governance is a product responsibility, not just a legal one. |
| Proficient | Applies risk-tiered classification to AI features (low, medium, high, critical). Designs for transparency and fairness. Works with legal, security, and compliance to meet regulatory requirements. Documents model lineage and training data provenance. |
| Advanced | Establishes governance across a product portfolio. Defines risk appetite by use case and ensures sensitive data, agent permissions, approvals, monitoring, and recovery match the consequence. Articulates regulatory and security trade-offs to leadership. |
| Expert | Defines the organisation's responsible AI and assurance strategy. Connects product, security, legal, risk, and operational controls into a system teams can use. Engages with industry bodies and regulators on emerging standards. |
5.6 Builder skills
Prototyping with AI coding tools, writing effective instructions, debugging AI systems, and judging whether an AI-generated output is fit for purpose. Building a working prototype shortens the distance from uncertainty to evidence.
| Level | In practice |
|---|---|
| Foundational | Can prototype a simple AI workflow using current coding or workflow agents. Writes effective instructions. Can distinguish a good model output from a mediocre one. |
| Proficient | Builds functional prototypes that demonstrate product concepts to stakeholders and users. Debugs prompt failures systematically. Crafts prompts for complex, multi-turn interactions. Uses evals to validate prototype quality. |
| Advanced | Builds prototypes sophisticated enough to run as internal tools or early-access products. Debugs agentic workflows, including tool failures, context limits and routing errors. Establishes shared context, instruction and evaluation practices for the team. |
| Expert | Operates as a technical co-founder for AI product initiatives. Builds proof-of-concept systems that de-risk major investments. Coaches product teams on builder skills. Has refined taste for AI output quality that sets the bar for the organisation. |
These competencies translate into concrete career expectations; see the product builder chapter for how they map to role definitions and hiring criteria at each level. The agentic security playbook provides the production depth behind AI security decisions.
How to use this model
Self-assessment. Rate yourself honestly across all five domains. Identify two or three sub-competencies where moving up one level would have the most impact on your current role.
Growth planning. Pick one sub-competency per quarter. Set a specific goal ("design an eval framework for my product's core AI workflow") rather than a vague aspiration ("get better at AI").
Coaching conversations. Use the proficiency levels as a shared language between manager and report. They make feedback concrete. "You're operating at Proficient in AI economics; this is the evidence required at Advanced" is more useful than "work on your commercial skills."
Hiring and team development. Map your team's current coverage. Identify gaps. Prioritise hiring or development based on which missing competencies create the most risk. Use the hiring product builders playbook to turn these capabilities into comparable evidence, then use AI-native team design to decide whether the gap needs development, a specialist, or another team shape.
Develop these competencies through deliberate practice on real work, using the frameworks in this handbook as scaffolding. The AI fluency spectrum describes the scope dimension that proficiency levels alone don't capture. AI adoption explains how individual capability becomes an organisational workflow. Career scenario planning helps choose which capability to deepen without betting on one forecast. The AI accountability principle is what makes the practice professional rather than merely productive.
v3.1 · Updated July 2026