What Is a Product Manager? Role, Types, and Key Skills

A product manager is responsible for defining what a product should do, who it serves, and why it matters. As companies compete more fiercely on product quality, the product manager role has become one of the most strategically influential positions in any organisation.

Product managers sit at the intersection of business, technology, and user experience. They do not write code or design interfaces, but they make the decisions that shape what gets built. By translating user needs into prioritised requirements and aligning cross-functional teams around a shared direction, the product manager is ultimately accountable for whether a product delivers real value or misses the mark entirely.

This article explains what product managers do day-to-day, how specialisations like the technical product manager and AI product manager differ from the general role, and what core skills separate good PMs from great ones.

‍

What Do Product Managers Do?

The core job of a product manager is to identify which problems are worth solving and then lead a team to solve them well. In practice, that involves a broad range of responsibilities spanning strategy, execution, and communication.

On any given day, a product manager might be conducting user interviews, writing and prioritising backlog items, reviewing analytics to understand how customers are actually using the product, or presenting a roadmap update to senior stakeholders. The role does not sit neatly in any one function — it connects all of them.

McKinsey research on product managers in the digital world found that almost 80% of PMs are involved in both design activities and go-to-market decisions. That breadth reflects how central the role has become across the entire product lifecycle, well beyond the development phase alone.

There is also a significant discovery gap across the profession. According to Product Led Alliance's product management statistics, only 34% of product teams regularly collect customer insights and use them to guide prioritisation — despite this being the benchmark standard. That gap represents both one of the most common failure points in product management and one of the clearest indicators of what separates high-performing PMs from average ones.

At its core, product management is about accountability for outcomes, not output. Success is not measured by how many features are shipped in a quarter but by whether those features moved the metrics that matter: retention, conversion, activation, or revenue growth. That accountability to outcomes rather than activity is what makes the role fundamentally strategic.

Alongside strategy and discovery, product managers spend considerable time managing stakeholder expectations, resolving prioritisation conflicts, and ensuring engineering and design teams have the clarity they need to move fast. The best PMs create the conditions for great work rather than just directing it.

If your organisation is building out its product capability, the Product People hiring guide for product managers walks through how to evaluate candidates across full-time, interim, and contract models, including what to look for at each seniority level.

‍

Technical Product Manager vs Standard Product Manager

Not all product managers work the same way. The technical product manager is a distinct specialisation that has grown significantly alongside the rise of complex software infrastructure, developer platforms, and API-first products.

A standard product manager focuses on customer-facing outcomes. They work closely with UX designers, marketing, and commercial stakeholders, and their decisions are grounded in user research and business goals. A technical product manager, by contrast, is embedded more deeply in the engineering layer of the product.

The key distinctions break down across several dimensions:

Focus: Standard PMs prioritise user-facing features and go-to-market strategy. Technical PMs concentrate on the underlying systems, APIs, developer tools, and platform architecture that make the product function.

Background: Technical PMs typically hold degrees in engineering or computer science, enabling them to engage credibly in architecture discussions and evaluate technical trade-offs. Standard PMs often come from business, design, or marketing backgrounds.

Stakeholder mix: Technical PMs spend more time working alongside engineering teams, data infrastructure specialists, and platform architects. Standard PMs are more frequently engaged with sales, marketing, and executive leadership.

Compensation: The specialisation commands a meaningful premium — technical product managers earn roughly $135,000 per year on average in the US, compared to around $91,000 for standard PMs.

The distinction is not about seniority. It is about the nature of the product being managed. A company building a payments API or a developer platform needs a technical PM. A company building a consumer-facing app may need a more customer-centric profile. Many scaling organisations eventually need both, working in tandem across platform and product layers.

As the Atlassian guide to agile product management captures well, the PM role is defined less by a fixed job description and more by the specific problems a given product needs to solve. The shape of the role follows the shape of the product.

‍

The Rise of the AI Product Manager

The AI product manager is the fastest-growing specialisation in the field. As organisations embed machine learning, generative AI, and intelligent automation into their products, they need PMs who can navigate a fundamentally different kind of product development.

Traditional product management deals with deterministic systems. Build feature X, and users get behaviour Y consistently and predictably. AI product managers work with probabilistic systems: the product's behaviour evolves as models are trained, updated, and fine-tuned on new data. That changes nearly everything about how you define requirements, test functionality, and measure success.

Core responsibilities of an AI product manager include:

• Identifying use cases where AI creates a meaningful advantage over rule-based approaches
• Translating business problems into machine learning problem framings that data scientists can act on
• Defining metrics that capture both model performance and business outcomes
• Monitoring for model drift — the gradual degradation in accuracy as real-world data shifts away from training distributions

The skills required also shift significantly. AI PMs do not need to write models, but they need statistical literacy: an understanding of distributions, confidence intervals, evaluation metrics, and why a high accuracy score can be deeply misleading when datasets are imbalanced. They must understand when a training set is insufficient, how to design valid A/B tests for probabilistic outputs, and how to communicate uncertainty clearly to non-technical stakeholders.

According to Product Led Alliance's research, 73.4% of product professionals expect PM roles to become more hybrid across product, design, and engineering — a trend driven largely by AI's increasing role in product development. Only 6.1% of organisations have embedded AI as a core product capability so far, which means the AI PM skillset is likely to be a significant differentiator in hiring over the next few years.

For companies that need this expertise but do not yet have it in-house, bringing in an interim product manager with AI and platform experience is a practical way to close that gap quickly while building toward a permanent appointment.

‍

FAQs

‍

‍Conclusion

The product manager role is one of the broadest and most strategically important positions in modern product-led organisations. Whether the context is a consumer app, a developer platform, or an AI-powered service, the PM's ability to connect user insights to business goals and translate both into clear priorities for engineering, design, and go-to-market teams is what separates companies that build what the market wants from those that build what seems internally obvious.

For organisations hiring, building, or scaling product capability, defining the right PM profile for your product context is the first decision. Getting it right early saves significant time, cost, and momentum down the line.

‍