product vision & strategy

Your CEO walks into the product review. Halfway through your feature update, she interrupts: “Forget the backlog. What is our product vision?”

You freeze. You have a roadmap. You have sprint goals. You have a Jira board with 200 tickets. But you cannot articulate — in two sentences — why your product exists and where it is going.

This is the most common failure mode in product management. Not bad execution. Not poor prioritization. A missing strategic spine. Without it, your roadmap is a shopping list. Your team ships features that do not compound. Your product drifts.

This page gives you the structure to fix that.

The strategy stack

Most PMs confuse vision, strategy, and roadmap. They use the terms interchangeably, which creates chaos when the leadership team expects different things from each.

Here is how they relate — think of them as a stack where each layer constrains the one below it:

Vision — the why. Where you want to be in 5-10 years. Technology-free, jargon-free. Your vision should survive a platform shift. If your vision mentions “AI” or “cloud” or “mobile,” it is too specific. Those are implementation choices, not destinations.

Strategy — the how. The set of deliberate choices about where to play and how to win. Strategy is what you say no to. If your strategy does not explicitly exclude markets, segments, or approaches, it is not a strategy — it is a wish list.

Roadmap — the what, when. The sequenced plan for executing the strategy over the next 6-18 months. A roadmap without a strategy above it is just a list of features sorted by who shouted loudest.

Goals/OKRs — the how much. Measurable outcomes tied to roadmap initiatives.

Roman Pichler captures this well: vision is the why, strategy is the path, roadmap is the steps, goals are the milestones. Each layer must trace back to the one above it. If a roadmap item cannot be linked to a strategic choice, it does not belong on the roadmap.

Writing a product vision that is not garbage

Most product vision statements are useless. They sound like this: “To be the leading platform for empowering businesses to unlock their potential through innovative solutions.”

That could describe any product in any industry at any time. It tells your team nothing. A good vision statement has three properties:

1. It describes a future state for the user, not the company.

Bad: “To be the #1 invoicing platform in India.” Good: “Every small business owner in India gets paid on time, without chasing.”

The first is a company ambition. The second is a user outcome. Your team can rally around a user outcome. They cannot rally around a market share number.

2. It is durable — free from technology and implementation.

Bad: “An AI-powered platform for automated invoice management.” Good: “Small businesses spend zero time on payment collection.”

AI is a tool. It may or may not be relevant in five years. The problem of getting paid on time will exist as long as commerce exists. Your vision should outlast your current tech stack.

3. It is specific enough to exclude.

Bad: “Making business easier for everyone.” Good: “Every small business owner in India gets paid on time, without chasing.”

The second excludes large enterprises. It excludes businesses outside India (for now). It excludes problems that are not about payment collection. Good. That specificity is what makes it useful for decision-making.

Strategy is the set of choices you make

Vision without strategy is a poster on the wall. Strategy is where the real work happens.

A product strategy answers four questions:

Who are we serving? Not “users” — a specific segment with specific needs. Zerodha chose retail traders who wanted low-cost, no-nonsense stock trading. They explicitly did not chase HNIs wanting wealth management or institutions wanting algorithmic trading. That choice shaped everything — their UI, their pricing, their support model.

What is our value proposition? What do we offer that alternatives do not? This is not a feature list. It is the reason a user picks you over the next best option. PhonePe’s value proposition was not “another UPI app” — it was UPI payments that work reliably at a merchant’s shop, even in tier-2 cities with patchy connectivity. That specificity drove their offline merchant strategy.

Where do we not play? The hardest part of strategy. Every feature you build for a segment you are not targeting is a resource you took from the segment you are. When Freshworks started, they chose small businesses explicitly. They did not try to compete with Salesforce on enterprise features. That constraint let them build fast, ship cheap, and grow from the bottom.

What is our unfair advantage? Why can we win where others cannot? This could be technology, distribution, data, network effects, regulatory positioning, or cost structure. More on this below.

Competitive moats: why they matter more than features

Features get copied. Swiggy launches a feature on Monday and Zomato has it by Friday. If your strategy depends on having better features, you do not have a strategy — you have a temporary lead.

Moats are structural advantages that get stronger over time. There are a handful that actually work in practice:

Network effects. Each new user makes the product more valuable for existing users. WhatsApp is useless if your contacts are not on it. Naukri becomes more valuable to recruiters as more job seekers join, and vice versa. Network effects are the strongest moat because they create switching costs for the entire network, not just one user.

Data compounding. Your product gets better as it collects more data — and that data is hard for competitors to replicate. Google Maps is better than any competitor because billions of users have contributed driving data, reviews, and location corrections over 15 years. A new mapping product starts at zero.

Switching costs. The cost (in time, money, or effort) of moving to a competitor. Tally has survived decades in India not because it is the best accounting software, but because millions of accountants have their entire financial history in Tally. Migrating that data — and retraining the accountant — costs more than the software itself.

Cost advantages. You can deliver the same value at a structurally lower cost. Jio did not just offer cheaper plans. Their cost structure was fundamentally different — a greenfield 4G-only network with no legacy infrastructure to maintain. That is a structural cost advantage, not a pricing decision.

Brand and trust. In regulated or high-stakes domains, trust is a moat. CRED built a moat around aspirational identity — it is not just a credit card bill payment app, it is a status signal. That brand positioning means they can launch new products (CRED Store, CRED Cash) with built-in distribution that a new fintech cannot replicate.

The question for your product is not “which moat do we have?” It is “which moat can we build?” Most products start with no moat at all. The strategy is to build one before competitors catch up.

The vision-strategy-roadmap trace

Here is the test most product teams fail: can you trace any single roadmap item all the way back to the vision?

Take a concrete example. Say you are building an ed-tech platform for professional upskilling in India.

The roadmap item (peer-matching) traces to the strategy (cohort-based, not self-paced). The strategy traces to the vision (affordable skill development for Indian professionals). The goal measures whether the roadmap item actually advanced the strategy.

Now imagine someone proposes adding a certificate marketplace to the roadmap. Trace it back. The strategy explicitly says “we do not compete on certifications.” The marketplace contradicts the strategy. It does not belong on the roadmap — no matter how many users request it.

This is the power of a strategy stack. It gives you a reason to say no that is not “I don’t feel like it” but “it contradicts our strategic choices.”

When the CEO says “what’s our product vision?”

This happens to every PM at some point. The right response is not to panic and produce a vision statement overnight. It is to run a structured process:

Step 1: Audit what exists. Your company already has implicit strategic choices — they are just not written down. Look at the last 12 months of product decisions. What segments did you invest in? What did you say no to? What do your best customers have in common? The strategy is already there, buried in decisions.

Step 2: Draft the stack. Write a one-sentence vision, a one-paragraph strategy (who, what, where we do not play, unfair advantage), and map your current roadmap against it. You will immediately see roadmap items that do not fit. That is the point.

Step 3: Get alignment, not approval. Do not present the vision as a finished document for the CEO to approve or reject. Present it as a draft for the leadership team to debate. The debate is the point. If your CEO, CTO, and head of sales have different mental models of who the customer is, discovering that now saves you a year of misaligned execution.

Step 4: Use it to decide. A vision document that lives in Confluence and never gets referenced is worthless. The strategy stack earns its keep when you use it to resolve disputes. “Should we build X?” becomes “does X trace back to our strategic choices?” If yes, prioritize it. If no, kill it or park it.

Common mistakes PMs make with strategy

Confusing strategy with goals. “Grow revenue 40% this year” is a goal, not a strategy. Strategy is the set of choices you make to achieve the goal. Two companies can have the same revenue goal and completely different strategies — one grows by going upmarket, the other by geographic expansion.

Copying someone else’s strategy. “Slack did it, so we should too” is not a strategy. It is cargo-culting. Slack’s strategy was built on their specific context — their user base, their market position, their competitive dynamics. Your context is different. Borrow mental models, not playbooks.

Strategy without tradeoffs. If your strategy document does not explicitly say “we will NOT do X,” it is not a strategy. A strategy that tries to serve every segment, enter every market, and build every feature is indistinguishable from having no strategy at all.

Updating the roadmap but not the strategy. Strategies should be reviewed quarterly and updated annually. If your strategy is two years old but your roadmap changes every sprint, there is a disconnect. Either the strategy is still valid (in which case your roadmap changes should trace to it) or it is stale (in which case update it first).

Making strategy a solo exercise. The PM does not own the strategy alone. Strategy is a leadership-level decision that involves the CEO, CTO, head of sales, and head of design. The PM’s job is to facilitate the process, draft the document, and ensure alignment — not to decide in isolation.

Test yourself

Where to go next

• Translate strategy into execution plans: Roadmap Planning
• Understand the competitive landscape: Market Research & Competitive Analysis
• Connect strategy to measurable outcomes: Metrics & KPIs
• Go back to foundations: Product Thinking