The No BS Guide to Startup Competitor Research

Josh, a founder I coached, burned $100K building a deep analytics dashboard for CFOs before he read a single competitor review.

The signal was public the whole time. Competitor forums were full of users complaining that the existing tools were too complex. Josh had bet on more depth when the market wanted less. Once he ran the competitor research, he pivoted to a simpler interface, won back the losses, and landed paying customers within months.

An afternoon of reading complaints would have saved him six figures. This guide is that afternoon, structured.

What is competitor research for?

Pattern-matching. You read where the incumbents fail, find the slice they can’t fix without breaking their own business model, and ship the 10x version of that slice for one buyer type.

The reframe most founders miss: competitors are evidence the market is real. If nobody competes for your customer, you’re either too early or the customer doesn’t exist. CB Insights pins poor product-market fit at 43% of startup failures, ahead of bad timing at 29% and weak unit economics at 19%. Competitor research is how you stay out of the 43%.

Who are your real competitors?

Most founder competitor lists read “tools that look like ours.” That list is wrong. Map five tiers:

• Tier 0: the status quo. The customer does nothing, runs a spreadsheet, or pays an intern. Usually your biggest competitor. “We just use Excel” ends more deals than any rival product.
• Tier 1: direct. Same buyer, same job, same budget. HubSpot against Pipedrive.
• Tier 2: indirect. Same job, different buyer or delivery model. Notion against Asana.
• Tier 3: adjacent. Same buyer, different job today, with a credible path into your lane.
• Tier 4: substitute. A tool bent into “good enough.” Airtable stands in for purpose-built CRMs below 50 seats, right up until it breaks.

Write three names per tier before you build a homepage. If your Tier 0 row is blank, you haven’t talked to enough customers yet.

Where do you find what competitors get wrong?

Skip their feature pages. Marketing copy tells you what a competitor wants to be. Complaints tell you what it is.

Start with the 1-star and 2-star reviews on G2 and Capterra, and read the text instead of the stars. G2 lets vendors run gift-card campaigns for positive reviews, so ratings are marketing. Nobody pays for a 1-star review. The angry text is the most honest data on the page.

Then widen the net:

• Reddit. Search site:reddit.com "[tool]" sucks and "switched away from [tool]".
• Hacker News. Search the tool name on Algolia, sorted by date. Comments cost reputation there, so the signal runs higher than review sites.
• Google operators. "why I left [tool]" and "switched from [tool] to" surface migration stories the review sites never catch.

The bar for signal: 30 or more unique complaints, across three or more sources, about the same feature, inside 90 days. Below that you’re pattern-matching noise.

This complaint mining is Step 2 of the 4-step startup idea validation framework. Quote three customer phrases verbatim before you write a single feature spec. Their words become your homepage copy.

How do you know a gap is worth building on?

Score each complaint pattern on five dimensions, 1 to 5 each:

1. Frequency. How many unique users complained in the past 90 days?
2. Intensity. Are they grumbling, hacking together workarounds, or already paying for a partial fix?
3. Specificity. Does the complaint cluster by role, vertical, or company size?
4. Budget. Are the complainers already paying real money for the parent tool?
5. Incumbent-can’t-fix. Would closing the gap force the incumbent to hurt its own revenue?

18 or higher out of 25: pursue. Below 15: kill it and keep mining.

The fifth dimension is the whole game. The strongest wedge is one the incumbent can’t close, because closing it means firing their own customers. Linear beat Jira exactly there. Creating an issue took 15 clicks in Jira and two keyboard shortcuts in Linear. Jira couldn’t copy the speed. Its enterprise revenue depends on infinite configurability, and opinionated speed breaks that promise. Pylon found the same shape of gap: Zendesk is a consumer ticketing model retrofitted onto B2B, so Pylon rebuilt support for B2B software teams whose customers live in Slack Connect. It raised a $32M Series B from a16z in 2024.

The test that ends the debate: finish the sentence “We’re [tool] for [segment].” Linear is Jira for fast software teams. Pylon is Zendesk for B2B SaaS. If you can’t fill the blanks in one clean sentence, you don’t have a wedge yet.

Should you price against your competitors?

No. Price against your positioning.

ClickFunnels charges 10x what other funnel builders charge and still holds a major share of the market. Rivals with more features at lower prices don’t dent it, because ClickFunnels frames itself as the obvious choice for one specific buyer.

Positioning runs in a fixed order, alternatives first and category last. Name what the customer does if you don’t exist (your tier map from above). Name what you have that those alternatives lack. Translate each attribute into an outcome the customer can see. Name the slice of buyers who care most. Then pick the category frame.

Narrow an existing category instead of inventing a new one. “CRM for investment banks” explains itself in four words. A new category needs 18 to 24 months of expensive education, and Quibi burned $1.75B in six months trying to shortcut that. Once the position is set, the startup pricing strategy guide covers how to set the number.

What if you have no competitors?

Two possibilities. Both should worry you.

Either they exist and you haven’t found them, or nobody spends money on this problem. “We have no competition” in a pitch deck reads as “we haven’t done the research.”

Every real problem already has a solution taped together somewhere. James, a client of mine, built his B2B SaaS product for agencies, polished it, and launched to silence. When he finally called the people he’d built it for, he learned agencies were already paying VAs $5 an hour to do the same job, and his product wasn’t 10x better than that duct tape. He didn’t have a marketing problem. He had a “nobody gives a shit” problem. He rewrote the value prop around a pain the VAs couldn’t touch and landed his first three paid clients in three weeks.

The duct tape is your Tier 0. It’s cheap, it’s already trusted, and it’s what you have to beat.

Where competitor research breaks

• You’ve talked to zero customers. Run 5 to 10 customer discovery interviews first. Competitors tell you what’s already in the market. Only customers tell you what’s missing.
• You copy features 1:1. You’ll ship a slower clone of the incumbent and lose on distribution. The research finds the gap. It doesn’t write your spec.
• You run it once and stop. Markets move quarterly. Three free trackers keep it continuous: F5Bot emails you when your competitors get named on Reddit or Hacker News. The Wayback Machine shows every pricing change they’ve made. Their LinkedIn job posts show where the money goes next: a sales hiring spike means new territories, an engineering spike means platform work, both 6 to 12 months before it ships.
• You’re selling enterprise software with a competitive-intelligence budget. Klue and Crayon run $15K to $40K a year and exist for that job. This workflow is for founders spending $0 to $50 a month.

The full workflow, with the tier worksheet, all 13 complaint surfaces, and the scoring rubric, is in the No BS Startup Guide.

Book a free strategy call if you want a second set of eyes on your competitor map or the wedge you think you’ve found.

Founders who fear competition haven’t done the reading. Competitors are proof the money is real. The research tells you which slice of it is yours.

Related startup guides

Use the product-market fit validation guide to test your wedge with wallet votes instead of compliments. Once the gap is confirmed, the startup GTM playbook shows how to pick the channel and positioning that exploit it.