A PM spends two weeks building a forty-slide competitor deck. The review goes well, people nod, someone says “great work.” Six months later, nothing on the roadmap has changed, and the same competitor quietly takes the segment anyway. This happens more often than anyone likes to admit.
Most competitive analysis produces information. Strategy needs decisions. Those are not the same thing, and mixing them up is how a genuinely useful exercise turns into a slide deck nobody opens again.
This guide covers how to tier competitors properly, which framework actually answers your question (not all of them, and not all at once), where evidence legitimately comes from, and how to turn a competitor’s move into a decision, including the decision to do nothing. There’s also an India worked example, because most competitive analysis content online is written for a US SaaS company that doesn’t exist.
What Competitive Analysis Actually Is (and What It Isn’t)
Competitive analysis is the structured practice of identifying who else solves your customer’s problem, evaluating how they solve it, and using that evidence to make specific decisions about your positioning, roadmap, pricing and go-to-market. Not simply to describe the market. Describing the market is what a Wikipedia page does.
It isn’t a feature checklist. It isn’t a one-off deck you build once a year before the offsite. It isn’t a SWOT square drawn on a whiteboard by four people with opinions and no evidence. And, worth saying once because the phrase gets overloaded, it isn’t SEO or ads research either, that’s a separate discipline with its own tools (Semrush, Ahrefs and friends already own that corner of the internet). This piece is about product and business strategy.
Competitor Analysis vs Competitive Analysis vs Competitive Landscape
People use these interchangeably in practice, and honestly that’s fine. But it helps to know the shades:
● Competitor analysis, depth on specific named rivals: their product, pricing, positioning, motion.
● Competitive analysis, the wider practice, including substitutes, new entrants and how the industry is structured.
● Competitive landscape, the map itself: who exists, in which tier, holding how much of the problem.
● Market research, the parent activity. Competitive analysis is one input, alongside customer research and market sizing, neither of which this article covers in depth.
The Output Should Be a Decision, Not a Deck
Every competitive analysis should end with a short, written list: decisions it changed, decisions it confirmed, and decisions it deliberately did not touch. If all three lists come up empty, the analysis failed, regardless of how polished the slides looked. This is arguably the single most useful habit in this entire article, and it costs nothing to adopt. Competitive judgement, by the way, is exactly the kind of thing a product manager is actually expected to own, not something you outsource to a research team and forget about.
Practical habit: open the document with the decisions, not the competitor profiles. The profiles belong in the appendix. Nobody in a strategy review wants to read twelve competitor bios before finding out what you’re proposing.
Signal to Strategy: The Chain Most Analyses Break
This is the part almost every article on this topic skips entirely, and it’s the whole point. A signal on its own is trivia. It only becomes useful once it survives four questions: what happened, what does it mean, what’s at risk, and what do we do about it.
Four Kinds of Competitor Signal
● Product signals: changelogs, release notes, app-store update text, new SKUs, deprecated features, API docs.
● Pricing and packaging signals: pricing-page changes, new tiers, seat-vs-usage shifts, discounting patterns, free-tier limits.
● Go-to-market signals: new channels, partnerships, homepage messaging changes, ad libraries, channel hiring, geographic expansion.
● Capital and talent signals: funding rounds, job postings (an underrated leading indicator, roles get hired two to three quarters before a product ships), leadership changes, filings.
Rank signals by lead time. Job postings lead the market by a couple of quarters. A changelog is roughly coincident with what’s actually shipping. A press release is lagging, it’s the last thing to know, not the first. Weight your attention accordingly.
The “So What?” Test
Signal: a competitor adds a free tier capped at three users. Insight: they’re chasing bottom-up adoption in small teams, not enterprise procurement. Implication: your sub-ten-seat segment faces a zero-price anchor within a couple of quarters, and your SMB win rate is the metric actually at risk. Decision: don’t match the free tier, reprice the entry plan as a flat per-team fee instead, and extend the trial. Owner named, review date set.
An observation with no measurable metric at risk isn’t an insight. An insight with no owner and no date isn’t a decision, it’s a nice sentence in a Google Doc. This discipline of moving from raw finding to something actionable isn’t unique to competitors either, it’s basically how a structured business analytics process is supposed to work anywhere in a company.
Competitive analysis should legitimately inform positioning, roadmap sequencing, pricing and packaging, segment focus, and partnership choices. It should not be the primary input for what customers actually need (that’s customer research) or how big the opportunity is (that’s market sizing). Different jobs, different tools.
Map the Landscape Before You Analyse It
The most common beginner mistake isn’t a bad framework, it’s analysing the wrong three companies deeply because those are the only ones the CEO ever mentions in a meeting. Tiering forces you to include the substitutes that actually take your revenue, which are usually less glamorous than a funded rival.
| Tier | How to spot them | Tracking depth |
| Direct | Same job, same segment, similar solution shape, sales calls and win/loss records will surface them | Deep, full profile, monthly |
| Indirect | Same job, different shape or segment | Medium, quarterly |
| Substitute / DIY | Spreadsheets, WhatsApp groups, an intern, an agency, or plain doing nothing | Medium, but highest strategic value |
| Aspirational | Sets category expectations even without competing for your deals | Light, watch for norms, don’t copy their roadmap |
| Potential entrant | Has the distribution or data to enter tomorrow | Light watchlist, named triggers only |
The “do nothing” competitor deserves its own paragraph. In a lot of Indian B2B categories, the real incumbent isn’t a funded rival, it’s Excel plus a WhatsApp group. You don’t win that deal with a feature comparison, you win it by making the status quo more painful than switching.
Worth borrowing from Clayton Christensen here: define competition by the job the customer is hiring a product to do, not by industry category. A small retailer choosing between a billing app, a UPI app that already shows transaction history, and a paper ledger is really comparing three answers to the same job, and the jobs-to-be-done framing explains why that comparison matters more than the category label.
A practical rule that no template online seems to mention: profile three to five competitors deeply, keep a shallow watchlist of around ten, and delete anyone you haven’t looked at in two cycles. An untended tracker is worse than no tracker, it produces confidently stale beliefs, which is a special kind of dangerous.
Pick the Framework That Answers Your Actual Question
Frameworks aren’t a checklist you march through in order. Each one answers exactly one question. Running all of them on every competitor is precisely how analysis paralysis begins, more on that failure mode later.
| Framework | Question it answers | Where it fails |
| Porter’s Five Forces | Is this industry structurally attractive? | Says nothing about individual rivals or what to build |
| Comparative SWOT | Where do we stand versus one named competitor? | Useless as a solo brainstorm with no evidence |
| Perceptual / positioning map | How do customers see us on the two things they actually trade off? | Falls apart if the axes are chosen for convenience, not customer relevance |
| Feature comparison matrix | Where are the real capability gaps, and which matter? | Invites feature-parity chasing without a weighting column |
| Win-loss analysis | Why do we actually win and lose deals, in the buyer’s words? | Needs disciplined interviews, CRM dropdowns lie |
Quick answer to a question that comes up a lot: is a SWOT analysis a competitive analysis? No. SWOT is one tool that can live inside a competitive analysis, and it’s only useful when it’s comparative, you against a named rival, with every cell backed by a source and a date. A solo SWOT drawn on a whiteboard is a brainstorm wearing a business-casual outfit.
And the “4 P’s” people ask about is usually the marketing mix, product, price, place, promotion, applied to a competitor. It’s a fine quick lens for go-to-market comparisons. It’s a marketing framework though, not a strategy one, and it won’t tell you what to build next.
Five Forces Looks at the Industry, Not Your Rivals
Bargaining power of buyers, suppliers, threat of new entrants, threat of substitutes, and rivalry among existing players, that’s the full set. In Indian B2B SaaS, buyer power is high because procurement cycles are price-sensitive and long. In D2C, entry barriers are often low because manufacturing and fulfilment get outsourced. But here’s the correction most articles skip: Porter’s framework tells you whether an industry is worth being in. It cannot tell you whether to build feature X. Different question entirely, and the framework is nearly five decades old at this point, so it also under-weights network effects and platform dynamics, worth keeping in mind if you’re analysing a marketplace or an AI product.
Positioning Maps Live or Die on Axis Choice
The whole game is picking axes customers actually trade off against each other, not axes that look tidy on a slide. “Features vs price” is a lazy default. “Setup effort vs depth of control” or “assortment breadth vs delivery speed” are the kind of axes that come from win-loss interviews, not a whiteboard session on a Friday afternoon.
The Comparison Matrix Template
| Dimension | Weight (1-5) | Us | Competitor A | DIY / substitute |
| Core job performance | 5 | |||
| Time to first value | 4 | |||
| Pricing model | 5 | |||
| Integrations | 3 | |||
| Support & SLA | 2 |
A few rules worth following exactly: weights come from customers, not your own team’s opinions. Every filled cell needs a source and a date, or it gets struck through, no exceptions. A gap on a weight-2 dimension is not a roadmap item, however loudly sales asks for it. And keep the DIY column, most online templates skip it entirely, and it’s often the column that quietly explains most of your losses.
Where the Evidence Comes From, and Where the Line Is
In rough order of reliability: public disclosures (quarterly results, shareholder letters, exchange filings, DRHPs for listed or IPO-bound Indian companies), the competitor’s own surfaces (pricing pages, changelogs, careers pages, take dated screenshots since these change quietly), buyer-side evidence (G2 reviews, app-store reviews filtered to the 2 and 3-star range, where the honest trade-offs actually live), hiring and capital signals, your own win-loss and churn data (already yours, chronically under-used), and paid tooling for traffic or download estimates, treat those as directional, not factual.
The Ethical and Legal Boundary
No misrepresenting who you are to extract information. No inducing anyone to breach an NDA. No accessing systems you have no right to. The SCIP Code of Ethics is the professional standard here, and it explicitly requires disclosing your identity before interviews. Worth knowing for India specifically: under the Competition Act, 2002, exchanging commercially sensitive information like future pricing directly with a competitor can raise cartel-conduct concerns, gathering public information is fine, chatting with a rival about prices is a different story entirely. The Competition Commission of India publishes compliance guidance on exactly this line.
A simple gut check: if you wouldn’t be comfortable with the competitor reading a transcript of how you got the information, don’t get it that way.
A Worked Example: Reading India’s Quick Commerce Battle
Quick commerce works well as an example because the players are familiar to almost every Indian reader, and because two of the three major names sit inside listed companies that publish investor disclosures, which means this isn’t a hypothetical US SaaS company invented for a blog post. This walkthrough is illustrative of method, not a scorecard on any company.
Tier the landscape first: direct competitors are the well-known quick commerce apps, indirect includes grocery on the larger e-commerce platforms, substitute is the neighbourhood kirana store plus a WhatsApp order, and any player with dense last-mile delivery infrastructure is a potential entrant worth watching.
Choose axes for a positioning map: assortment breadth against delivery promise, with basket value as bubble size, not “price vs quality,” which tells you almost nothing useful here. Then collect a handful of dated signals, dark-store footprint changes, category expansion, shifts in platform or delivery fees, and run the so-what test on each. That’s the whole method, applied to a category everyone already has an opinion about.
For anyone building this out for real, pull current figures directly from Eternal’s investor relations disclosures and Swiggy’s investor relations page, cite the exact quarter, and note your retrieval date. Numbers in this space move fast enough that anything undated is close to useless within a quarter.
The decision a smaller player in this space should land on usually isn’t matching dark-store count, that’s an expensive game only the best-capitalised player wins. It’s picking a specific geography or category where delivery speed matters less than assortment depth, and defending that ground properly instead.
What to Do When You’re the Smaller Player
Most readers of an article like this work at, or want to build, the smaller company. So here’s the part nobody on the SERP bothers to address.
● Be different, not better. Matching a larger rival on their strongest dimension is the most expensive way to lose slowly.
● Win one segment completely before chasing a whole market. Depth beats shallow presence across five categories.
● Find the distribution asymmetry, a channel, a community, a language, a price point the incumbent structurally can’t reach.
● Use packaging as a weapon. You can restructure pricing in a week. A large incumbent usually can’t, without disrupting existing revenue.
● Serve deliberate non-consumption, customers the incumbent has decided aren’t profitable enough to bother with. Classic entry wedge.
One habit to avoid entirely: using a competitor’s roadmap as your own backlog. Do that and you guarantee arriving second to every feature and first to nothing, which is a fairly bleak place to build a company from.
The Two Ways Competitive Analysis Destroys Value
Every top-ranking page on this topic treats competitive analysis as unambiguously good. It isn’t, and naming the failure modes is more useful than another list of benefits.
Feature-Parity Chasing
Competitor ships X, sales asks for X, X enters the roadmap, six months of parity work follows, and no real differentiation gets built in the meantime. It happens because features are easy to compare and positioning isn’t. The fix: every competitor-triggered roadmap item has to name the segment it protects and the metric it moves. Can’t name one? It’s parity work, and it goes below the line. Keep an explicit not-building list too, with reasons, it’s the cheapest defence available in a roadmap review.
Analysis Paralysis
The analysis expands to fill whatever time is available, the deck grows, the decision date slips, and the market moves on without you. Timebox it instead, a first-pass competitive read should take days, not weeks. Write your decisions as hypotheses first, then gather only the evidence that could actually change them. Research that can’t move a decision is entertainment, expensively formatted entertainment.
The Response Ladder
When a competitor makes a move, run it through five filters, relevance, evidence quality, metric at risk, positioning fit, and opportunity cost, then land on one of these:
1. Ignore, different segment or nothing at risk. Log the reason and don’t re-litigate it next month.
2. Monitor, set a named trigger and a review date.
3. Neutralise, remove the objection with documentation or a comparison page, without building the feature.
4. Match, only when it protects a chosen segment and moves a named metric.
5. Reposition or leapfrog, change the axis of competition entirely. Highest return, least used.
If more than a third of your responses land on Match, you’re quietly running someone else’s roadmap. Most competitor moves should end at Ignore or Monitor, that’s not laziness, that’s judgement.
Turn It Into a Cadence, Not a One-Off Deck
| Rhythm | What happens | Time cost |
| Continuous | Automated alerts on pricing pages, changelogs, job boards | Near zero after setup |
| Weekly | Triage the alert queue against the relevance filter | 15 minutes |
| Monthly | Refresh the comparison matrix, log any decisions changed | 1-2 hours |
| Quarterly | Refresh positioning map, win-loss coding, sales battlecards | Half a day |
| Annually | Re-tier the whole landscape from scratch | 1 day |
Every tracked competitor needs a kill criterion, the condition under which you stop tracking them. Trackers that only grow, and never shrink, become noise. And noise is exactly what produces analysis paralysis in the first place, which brings the whole thing full circle.
AI-Era Competitive Intelligence: What to Automate, What to Verify
AI-assisted research has become standard practice fast in Indian product and business teams, Scaler’s India AI Workforce Report tracks this shift in more detail if you want the wider picture, and it’s changed how quickly a first-pass competitor read gets built.
What genuinely automates well: change detection on public pages, alerting, transcribing win-loss interviews, summarising long filings, clustering review-site complaints into themes. Tools built for this, alongside general-purpose AI automation for business workflows, can save a genuine day of grunt work. What doesn’t automate: choosing the axes of a positioning map, weighting the comparison matrix, deciding which segment is worth defending. Judgement is the job. Collection is the chore.
The honest problem nobody selling these tools wants to mention: LLM-generated competitor research routinely produces stale pricing, invented features, and confident numbers with no real source behind them. Retrieval helps but doesn’t fix this, a tool can retrieve the right page and still summarise the pricing tier wrong.
A short verification protocol worth actually following:
1. Every factual claim traces to a primary source URL.
2. Every claim carries a retrieval date.
3. Pricing and availability get verified on the competitor’s own page, never from a summary.
4. Any number that would change a decision gets a second, independent source.
5. AI output is treated as a first draft of the questions to verify, not the answer itself.
Where This Shows Up in Interviews and on the Job
Competitive judgement is what separates a PM who reports on the market from one who actually shapes a roadmap. It surfaces in quarterly planning, pricing reviews, and the awkward sales escalation where a rep swears the deal was lost purely on price.
“How would you compete with X” and “a competitor just launched Y, what do you do” are standard product-sense interview questions. What’s actually being scored: do you clarify the segment before answering, do you tier competitors instead of just listing brand names, do you name the metric at risk, and can you land on a decision with the trade-off said out loud, including the option of doing nothing.
A genuinely useful portfolio exercise for anyone switching into product: build one public competitive teardown of a category you actually use, tiering, a positioning map, a weighted matrix, three decisions and one explicit thing you’re refusing to build. Two pages, no more. It demonstrates judgement far better than a certificate ever will, and it doubles as an interview artefact. If you’re mapping out what comes after that kind of upskilling, career paths after a PGP in Business and AI is worth a look.
Frequently Asked Questions
What do you mean by competitive analysis?
The structured practice of identifying who else solves your customer’s problem, evaluating how they solve it, and using that evidence to make specific decisions about positioning, roadmap, pricing and go-to-market, rather than just describing the market.
What is the difference between competitive analysis and competitor analysis?
Used interchangeably in practice. Where a line gets drawn, competitor analysis goes deep on named rivals, competitive analysis covers the wider landscape, substitutes, new entrants, industry structure. Both should end in a decision.
What are the 5 steps of a competitive analysis?
Tier the landscape, pick the framework that matches your question, gather dated evidence from primary sources, build a weighted comparison matrix and positioning map, and write a decision log covering what changed, what got confirmed, and what you deliberately won’t do.
Is a SWOT analysis a competitive analysis?
No. SWOT is one tool that can sit inside a competitive analysis. It’s only useful when comparative, versus a named rival, with dated evidence in every cell. Solo, it’s a brainstorm.
How often should you do a competitive analysis?
Continuous alerts, weekly triage, a monthly matrix refresh, quarterly positioning-map and battlecard updates, and a full annual re-tier of the landscape. A cadence, not a once-a-year project.
Is competitive intelligence legal in India?
Gathering public information is standard, legitimate practice. Misrepresenting your identity, inducing an NDA breach, or exchanging sensitive information like future pricing directly with a competitor is not, the last can raise cartel-conduct concerns under the Competition Act, 2002.
The competitors were never the analysis. The chain from signal to insight to implication to decision is. An exercise that changed nothing was research. One that changed three things, and deliberately refused a fourth, was strategy.
Tier your landscape this week. Fill the matrix for three competitors. And write down the one decision you’re refusing to make, that last part is harder than it sounds, and more useful than almost anything else in this article.
Turning competitor signals into positioning, pricing and roadmap decisions is exactly the judgement structured product and business programmes are built to develop. Scaler’s online PGP in Business & AI covers strategy frameworks, product decision-making and AI-era analysis with mentorship from working practitioners.
