Gap analysis for developer-tool companies
Updated September 2026
Gap analysis is the part of market intelligence most developer-tool teams skip, because it needs the boring half first: a named list of competitors and a record of where each of you has shown up. This page is how we run it and what to do with the output.
Quick answer
MarketHQ runs gap analysis continuously and turns each absence into a prioritised action plan; the three kinds of gap that matter are conversations you missed, venues where a rival keeps appearing and you never have, and AI answers that recommend someone else. Each gap becomes one action, and the point is closing them, not counting them.
Key points
- A gap is competitor-present and you-absent, in the same place, at the same time.
- Three kinds: missed conversations, venues you have never appeared in, and buying questions an assistant answers without you.
- Every gap ships with one move, so the output is a plan rather than a report.
What sources does MarketHQ cover?
Free: Hacker News, GitHub, RSS, Stack Overflow, Dev.to, Lobsters. Paid plans add Reddit, LinkedIn, YouTube, X, Quora, Web, Perplexity, Product Hunt, Broader web crawl, Bluesky.
- Grouped Slack daily digest
- Drafted replies from your own strengths
- Gap analysis and action plans
- Weekly AI-answer visibility (paid plans)
- Chat over your own evidence
- API and MCP on every paid plan
- Flat capacity pricing, tiers differ by tracking volume only
- 7-day trial for $0.99 on paid plans
- Free tier with no card and no end date
What is gap analysis for a developer-tool company?
It is the comparison between where your category gets discussed and where you actually appear, run against a named list of competitors rather than against the whole internet.
A market report tells you what is happening. A gap analysis tells you what is happening without you. The unit is not a mention but an absence: a Stack Overflow answer recommending a rival SDK, a subreddit where two competitors get named monthly and you never have, a comparison thread that closed before anyone read it.
That framing suits developer tools, because the buying conversation is public and technical. It happens in issues, threads and answers rather than analyst calls, and it leaves a record you can check. That record is the evidence behind every gap we show, so a claim about where you are missing can be audited.
Which gaps actually matter for a developer tool?
Three: conversations you are missing, places a competitor is seen and you are not, and buying questions where an AI assistant names someone else.
Conversations you are missing are live, high-intent threads where a tracked competitor is named and you are not: someone comparing options, asking which tool solves a problem, or complaining about a rival. These have a deadline. A comparison thread is worth answering this week and close to worthless next month.
Places they are seen and you are not are cumulative rather than urgent. Every mention resolves to a venue: a subreddit, a GitHub repository, Hacker News, the publication behind an RSS item. The gap list is the venues where a competitor has appeared and your brand never has. That is a presence problem, not a reply you forgot.
The third kind sits inside AI answers. Each week we ask the buying questions in your category and record which brands come back. A question where a competitor is named and you are not is the same gap, in the place a growing share of shortlists gets built.
How do you find gaps without reading every thread?
Track a short list of competitors and problem phrases alongside your own name, then look only at the matches where they appear and you do not.
Most monitoring optimises for recall: catch every mention, forward it, let a human sort it out. That works until the volume outgrows the person. Gap analysis inverts it. Your own mentions are the least interesting rows in the set. The signal is the complement.
So the setup is deliberate rather than exhaustive. Name the two or three competitors you actually lose to, not the twenty in the category. Add the problem phrases a buyer uses before they know any brand name: connection pooling, rate-limit errors, migrating off something. The overlap between those lists is where gaps live, and it stays small enough to read.
- Two or three real competitors, not the whole category.
- Your own brand, which mostly functions as the absence test.
What turns a gap into an action plan?
One move per gap, scored against the dimension it belongs to, so the list arrives ordered by what would change if you did it.
A gap on its own is a complaint. We score your position across presence dimensions such as AI visibility, search presence and community share of voice; perception dimensions such as sentiment balance and authority footprint; and product dimensions such as problem presence and switching momentum. Each dimension carries the move that would raise it.
In practice the moves are small and specific. Answer this thread. Publish the comparison page this question keeps asking for. Fix the docs page a competitor's answer is beating. The plan is useful because it is short: a list of forty gaps is a structured way of doing nothing.
How often should a team run gap analysis?
Weekly for the conversation gaps, quarterly for anything whose fix is a page someone has to write.
The two halves move at different speeds. Missed conversations decay. A thread answered four days late reads as marketing, and one answered four weeks late does not get read. Those belong in a daily digest and a weekly review.
Venue and AI-answer gaps move slowly and cost more to close, because the fix is content or documentation rather than a reply. Treat those as a quarterly planning input. Running the whole thing weekly and acting on none of it is the common failure, and it is worse than not running it.
Limits: where does gap analysis not help?
When the gap is the product itself, when your category is argued out in private, or when nobody has time to act on the output.
If buyers keep choosing a competitor because it ships a feature you do not, gap analysis will say so clearly and repeatedly, and it will not build the feature. At that point the output is a roadmap input, not a marketing task.
Coverage has real edges too. We read public developer communities and the public web. A category argued out in private Slack groups, customer calls or closed Discords leaves no public record, and any tool claiming to see it is guessing.
What does a week of gap analysis look like for an API company?
Illustrative example, not measured: a small team selling a managed queueing API, tracking two competitors and four problem phrases.
Say a week turns up 90 matches. Sixty-two name a competitor, eleven name you, and the rest are noise from a similarly named company. The gap view marks six of the sixty-two as high-intent: three Reddit comparison threads, two Stack Overflow questions where a competitor's SDK is the accepted answer, one GitHub issue asking which service handles retries better.
It also shows four venues where both competitors have appeared and you never have, and two buying questions where assistants named a rival and not you. That is twelve items rather than ninety: six replies someone can write this week, four a content decision, two a documentation gap. Every figure here is invented to show the shape of the output.
Illustrative example
“Illustrative example, not measured: a small team selling a managed queueing API, tracking two competitors and four problem phrases.”
Say a week turns up 90 matches. Sixty-two name a competitor, eleven name you, and the rest are noise from a similarly named company. The gap view marks six of the sixty-two as high-intent: three Reddit comparison threads, two Stack Overflow questions where a competitor's SDK is the accepted answer, one GitHub issue asking which service handles retries better. It also shows four venues where both competitors have appeared and you never have, and two buying questions where assistants named a rival and not you. That is twelve items rather than ninety: six replies someone can write this week, four a content decision, two a documentation gap. Every figure here is invented to show the shape of the output.
Frequently asked questions
What counts as a gap in gap analysis?
A place where your category is being evaluated, a tracked competitor is named, and your brand is not. It can be a live thread, a venue you have never appeared in, or a buying question an AI assistant answers without you. Absence is the unit, not volume.
How is gap analysis different from mention monitoring?
Mention monitoring shows you the rows that contain your name. Gap analysis shows the rows that do not. Both need the same collection underneath, but the useful output is the complement: where competitors appear and you are missing, ordered by whether acting on it would change anything.
Which sources does MarketHQ read for gap analysis?
Hacker News, GitHub, Reddit, Stack Overflow, Dev.to, Lobsters and RSS, plus paid social and web lanes. Every gap keeps the evidence it came from, so you can open the original thread instead of taking the conclusion on trust.
Do I need a paid plan for gap analysis?
Plans are Free, Lite $29, Startup $99 and Growth $199 a month, with a 7-day trial for $0.99. The paid tiers differ by tracking capacity rather than by feature, so the analysis does not sit behind an upgrade. Reading it from the API or MCP needs a paid plan, Lite and up.
Can an agent read the gap analysis?
Yes, on paid plans. Gaps and action plans are available through the API and the MCP endpoint on every paid tier from Lite up. Free workspaces get the chat interface and the Slack digest, but not programmatic access.
Related
- MarketHQ vs Syften — the head-to-head against the closest feed-shaped competitor
- MarketHQ vs Brand24 — the same decision against a broad media-monitoring suite
- Track mentions on Hacker News, GitHub and Reddit — the collection half, which has to work first
- Best developer community monitoring tools — the category judged on coverage and triage
- All use cases — the other jobs this product is bought for
- Pricing — what the four tiers cost
Sources
- MarketHQ pricing: https://markethq.ai/pricing (verified 2026-09-13)