Most competitive analysis is a spreadsheet: your features down the left, competitors across the top, checkmarks in the grid. It takes days to build, it's obsolete in a month, and it answers exactly one question badly: "who has more stuff?"
Customers don't buy checkmarks. They buy solutions to problems, chosen under constraints, influenced by price, trust, switching costs, and what their peers use. Good competitive analysis studies that. Here's how to do it in a way that actually changes your decisions.
What Competitive Analysis Is For
Before doing any of it, know which decision you're informing. The useful ones:
- •Positioning: What can we credibly claim that competitors can't?
- •Roadmap: Where are we losing deals or users, and does that change our priorities?
- •Pricing and packaging: What does the market consider table stakes vs. premium?
- •Strategy: Which segments are overserved and underserved? Where should we not fight?
Analysis without a decision attached is a hobby. The feature matrix fails mostly because it's built for no decision in particular.
Start With Losses, Not Websites
The highest-signal competitive intelligence isn't on competitors' websites. It's in your own lost deals and churned customers.
- •Win/loss interviews. Talk to prospects who chose a competitor. Ask what the decision came down to, who else was in the running, and what almost changed their mind. Losses are uncomfortable and pure gold. Sales can get you these conversations; it's one of the best reasons to invest in that relationship.
- •Churn interviews. Customers who left for a competitor will tell you exactly which promise pulled them away, and whether it was kept.
- •Sales call recordings. Search for competitor names. You'll hear objections and comparisons in customers' own words.
A dozen of these conversations beats a hundred hours of website archaeology, because you learn how the choice actually gets made in your market. Often the finding is humbling: deals hinge on onboarding speed, security review friction, or one integration, not the flagship features anyone compares.
Then Study the Competitors Themselves
With the choice criteria understood, direct competitor study becomes targeted instead of encyclopedic. Sources, in rough order of value:
Use the product
Sign up for trials. Go through onboarding. Try the core workflows. An hour inside a competitor's product teaches you more than any analyst report, and almost no PM actually does it. Note what they've clearly invested in and what's neglected. Neglect is strategy too.
Read their public reviews
G2, Capterra, App Store reviews, Reddit threads. Sort by negative. Competitors' unhappy customers are describing the gaps you could win on, for free, in detail. Do the same for your own reviews to see what they'd say about you.
Watch their pricing and packaging
Pricing pages encode strategy: who they're targeting, what they consider premium, where they're moving upmarket or down. Screenshot them quarterly; the diffs tell stories. (Useful input for your own pricing work.)
Track their hiring and announcements
Job postings reveal roadmaps months early. A competitor hiring five ML engineers and a healthcare compliance lead is telling you their next two bets. Changelogs, launch posts, and conference talks fill in the rest.
The Analysis: Three Questions That Matter
Once you've gathered, resist the urge to make the big matrix. Answer three questions instead:
1. Where do we win, and why?
Not "what features do we have," but what deals, segments, and use cases consistently break our way, and what drives it. This is the raw material for positioning and for doubling down. Strength you don't understand is strength you can't protect.
2. Where do we lose, and does it matter?
Every loss reason goes in one of three buckets:
- •Fix: losses in segments we care about, for addressable reasons
- •Accept: losses in segments outside our strategy (losing enterprise deals is fine if you're deliberately SMB-focused)
- •Reframe: losses driven by perception rather than reality, which are marketing problems, not roadmap problems
The "accept" bucket is the one weak competitive analysis never uses. If you treat every loss as a gap to close, competitors set your roadmap, and you'll be a mediocre copy of three different companies simultaneously.
3. Where is the market going that nobody owns yet?
The most valuable competitive insight is the space between competitors: the underserved segment, the workflow everyone handles badly, the emerging need nobody's positioned for. This comes from combining competitor study with your own customer interviews. Competitors tell you where the fight is today. Customers tell you where it'll be next.
Keep It Alive Without Making It a Job
Competitive analysis rots fast. A quarterly mega-report is stale before it's read. Better: a lightweight, continuous system.
- •One living doc per major competitor. Positioning, pricing snapshot, strengths and weaknesses in your deals, recent moves. One page each, updated when something changes.
- •A monthly 30-minute sweep. Changelogs, pricing pages, review sites, job postings. Calendar it.
- •A channel for field intelligence. A Slack channel where sales and CS drop competitor mentions. You'll get a running feed of ground truth, and posting a monthly summary back keeps them contributing.
- •Win/loss conversations on a drumbeat. Even one per month compounds.
Total cost: a few hours a month. That's sustainable, which matters more than thorough.
One distribution tip: share what you learn in small, regular doses rather than big reports. A short monthly note ("what competitors did, what it means for us, what we're doing about it") gets read and builds your reputation as the person who understands the market. A 40-slide quarterly deck gets skimmed once and forgotten. Same research, radically different impact, and the note takes a tenth of the effort.
The Traps
- •Competitor obsession. If your roadmap is mostly "they shipped it, so we must," you've outsourced strategy. Fast followers can win, but only when following is a choice, made per-item, against your own thesis.
- •Feature parity as a goal. Customers don't want parity. They want their problem solved better. Parity is the most expensive way to be unremarkable.
- •Dismissing upstarts. The scrappy tool that's "not a real competitor" because it lacks your enterprise features is often serving a segment better than you, and segments grow. Disruption looks like a toy right up until it doesn't.
- •Analysis as anxiety management. Some competitive research exists to make leadership feel informed rather than to inform decisions. If nobody can name the decision the analysis serves, decline politely and do discovery instead.
- •Believing the demo. Competitor marketing shows the product at its best, configured by experts, with perfect data. Your customers will compare it to your product at its worst, on their messy real-world setup. Weight actual user reviews and win/loss evidence over anything a competitor says about themselves.
The Bottom Line
Study losses before websites. Use competitors' products. Sort their bad reviews. Answer three questions: where we win, where we lose and whether it matters, and what nobody owns yet. Keep it alive in small weekly doses and let it inform real decisions, not a slide that gets presented once and archived.
Competitive questions come up constantly in PM interviews ("how would you compete with X?"), and this structure, especially the fix/accept/reframe framing, will serve you well there too.
When you're ready to size up a new market of your own, browse open PM roles at productmanagerjobboard.com.