Competitor Analysis Example: 142 Keyword Gaps Across 3 Competitors
This example shows the aiwebpageseo Competitor Analysis for a SaaS marketing site against 3 competitors. You at DR 340, top competitor at DR 620. 142 keyword gaps (they rank, you don’t), 38 content gaps, 21 link gaps and 8 schema gaps. Strongest opportunity: build vs-comparison pages and reclaim missing trade-publication links.
| Metric | You | competitor-a.com | competitor-b.com | competitor-c.com |
|---|---|---|---|---|
| Domain Rank | 340 | 620 | 410 | 280 |
| Referring Domains | 847 | 2,140 | 1,210 | 620 |
| Total Backlinks | 12,480 | 42,500 | 18,700 | 9,200 |
| Indexed Pages | 1,820 | 5,400 | 2,100 | 1,200 |
| Avg CWV pass rate | 86% | 72% | 91% | 68% |
| Schema coverage | 92% | 74% | 81% | 61% |
| AI citation rate | 41% | 62% | 38% | 29% |
| Keyword | Volume | Best competitor pos | Difficulty | Action |
|---|---|---|---|---|
| async standup tool | 3,200 | competitor-a #2 | 42 | ⚠ Build comparison page |
| remote team retrospective software | 1,800 | competitor-a #3 | 38 | ⚠ Create content cluster |
| distributed team OKR tracking | 1,200 | competitor-b #4 | 32 | ✅ Low effort win |
| cross-timezone meeting scheduler | 980 | competitor-c #2 | 28 | ✅ Low effort win |
| project management for engineering teams | 4,800 | competitor-a #1 | 62 | 🔴 Hard but high value |
"vs Asana" / "vs Linear" comparison pages
Competitor-a ranks top-3 for 14 "vs [Major Tool]" queries. You have 0 comparison pages. Build 5-8 high-quality comparison pages with feature tables, pricing comparison and use-case differentiation. Expected new organic clicks: 2,400-4,800/month within 6 months.
21 high-authority links your competitors have, you don’t
TechCrunch, Forbes, ProductHunt, IndieHackers, plus 17 trade publications link to competitor-a but not you. Outreach + thought-leadership content (case studies, original research) is the path. Reclaim 5-10 within 6 months.
SoftwareApplication schema gap
Competitor-a has SoftwareApplication + Offer schema on every product page. You have it on 22% of yours. Do not expect a rich result from this — there is no SaaS enhancement to earn. What accurate SoftwareApplication and Offer markup does is disambiguate: it states unambiguously that the page describes a piece of software and that a given figure is its price. Add it where it truthfully describes the page, and nowhere else.
Read the table row by row and a pattern emerges that the composite framing tends to hide. You lead on Core Web Vitals. You lead on schema coverage, by a wide margin. You trail on referring domains, on indexed pages, on keyword coverage, and on AI citation rate. In other words: everything a crawler can fix, you have already fixed. Everything that requires somebody to write something, you have not.
That is the most common shape in competitor analysis and it is worth naming plainly, because the instinct on receiving this report is to do more of what you are already good at. Pushing schema coverage from 92% to 98% is measurable, satisfying, and worth almost nothing. The competitor at 74% coverage has a higher citation rate than you do, which should end the theory that markup volume drives citations on its own.
The gap is 142 keywords and 38 topics where they have published something and you have not. That is a content deficit, and no technical work closes it.
340 against 620 looks like a chasm. Three corrections before anyone treats it as a target.
Domain Rank is DataForSEO's model, not a search engine's
Every third-party authority score is one vendor's reconstruction of a link graph. Google uses none of them and has said consistently that it has no single site-wide authority number of this kind. The score is useful for putting two sites on a common scale. It is not a dial, and optimising it means optimising something no search engine can read.
They did not acquire 2,140 referring domains — they earned them
Somebody published things people wanted to cite, over years. The link count is the residue of that. Chasing the count directly leads to buying links, which is a policy violation and, more to the point, futile: purchased links are the ones Google is best at recognising and discounting.
Which of their pages hold the links?
It is almost never the product pages. Find the handful of URLs that attracted the references — original research, a free tool, a definitive guide — and you have found what the 21-link gap is actually made of. Then ask what you could publish that would deserve the same treatment.
One reassurance while we are on links: do not go hunting for toxic backlinks in response to this. Google's guidance is that most sites should never use the disavow tool, and since Penguin 4.0 it devalues links it does not trust rather than penalising the site they point at. Disavow only for a manual action, or for links you built yourself.
The table sorts by volume and the eye goes straight to the 4,800-a-month head term where competitor-a sits at position one. That is the worst possible first target: they hold it with years of accumulated references, the difficulty is 62, and you would be attacking their strongest position from your weakest.
Sort instead on two things the report cannot compute for you.
Could you plausibly write the best answer that exists for this query?
Look at what currently ranks. If the top results are thin, generic, or written by people who have never used the thing they are describing, the query is winnable regardless of its difficulty score. If the top result is a genuinely excellent piece by somebody who knows more than you do, it is not — and no amount of technical work changes that.
Would the person searching this ever buy from you?
A 1,200-volume term describing exactly what your product does is worth more than a 4,800-volume term describing the category loosely. Traffic that cannot convert is a cost: it consumes writing time, it dilutes your topical focus, and it flatters the dashboard.
Ranked that way, the two "low effort win" rows rise to the top and the head term drops to a twelve-month ambition — which is what it actually is. Volume is the most visible column and the least useful one for sequencing work.
Competitor-a holds top-three positions on fourteen "vs" queries and you have none. The opportunity is real: comparison searches come from people close to a decision, which makes them among the most commercially valuable queries you can hold.
The way these pages fail is entirely predictable. A vendor writes a comparison in which the vendor wins every row, the competitor's product is described in terms its own users would not recognise, and the "when to choose them" section is either absent or transparently insincere. Readers have seen a hundred of these and they discount them on sight. So do the people deciding what to link to.
The version that works states, in plain terms, where the other product is genuinely better and who should buy it instead. That sounds like giving ground and it is the single most persuasive thing on the page, because it is the only sentence a sceptical reader believes — and having believed it, they believe the rest. It is also the sentence an assistant can lift and quote, because it is specific and self-contained, which is exactly what retrieval systems surface.
Keep the feature table honest and current, and mark the page up with what is actually on it. There is no ComparisonTable type in schema.org — do not invent one — and no rich result awaits a SaaS comparison page. The markup describes; it does not promote.
You have 92% schema coverage. Competitor-a has 74%. Their AI citation rate is 62% and yours is 41%. If denser structured data drove citations, that result is impossible.
It is worth stating the defensible position clearly, because the overclaim is everywhere. There is no verified evidence that AI engines preferentially cite pages carrying more structured data. What is true is that retrieval systems generally parse the HTML they are served rather than rendering it in a browser, so JSON-LD present in the initial response reaches them, and it disambiguates — it states that a number is a price, that a name is an author, that a page describes a piece of software. That is a clarity benefit and it is real. It is not a competitive weapon, and you cannot out-schema somebody who has written better answers.
Which points at the actual explanation for the citation gap: competitor-a has published pages that answer the questions being asked, in prose that survives being lifted out of context. You have 38 topics where you have published nothing. The markup was never the difference.
Is a Domain Rank gap of 340 to 620 the reason they outrank me?
It is unlikely to be the direct reason, because no search engine uses Domain Rank. It is DataForSEO's model of a link graph, useful for putting two sites on a comparable scale and used by nobody at Google, who have said repeatedly that they have no single site-wide authority score of this kind. What the gap does indicate, indirectly, is that more people have chosen to reference competitor-a than have chosen to reference you. That is a real difference and it is worth understanding — but the thing to investigate is which of their pages earned those references and why, not the number itself, which cannot be optimised because nothing consumes it.
How should I pick which of the 142 keyword gaps to work on?
Not by search volume, which is what the table sorts on and what everyone reaches for first. Pick by whether you can plausibly produce the best available answer to that query, and by whether the people searching it would ever buy from you. A 4,800-volume head term where competitor-a sits at position one with years of accumulated links is a bad first target, however tempting the number. A 980-volume term where the existing results are thin and nobody has written a serious answer is worth more, because you can actually win it and the traffic converts. Rank the list by winnability times commercial relevance, and the order changes completely.
Will building comparison pages actually deliver the projected traffic?
The projection in the report is a scenario, not a forecast, and it should be presented internally as one. It assumes the pages get written well, get indexed, and reach the positions competitor-a currently holds — none of which is guaranteed, and the last of which depends on competitors who are not sitting still. What is defensible is the direction: comparison queries have high commercial intent, competitor-a is capturing them, and you have no pages targeting them at all. That is a genuine gap. Write the pages because the gap is real, and treat any number attached to the outcome as an illustration.
Should I copy the competitor's content cluster structure?
Copy the observation, not the artefact. The useful finding is that competitor-a has decided a set of topics matters and has published on them systematically. What you should not do is reproduce their page inventory as a checklist, because a cluster of mediocre pages built to match somebody else's sitemap is exactly the thin, derivative content that a core update reassesses downwards. The pages that win are the ones that answer the question better than the existing results — which requires knowing something the existing results do not say. If you have nothing to add on a topic, publishing on it anyway is a cost, not a gain.
What is the fastest way to close a 21-link gap?
There is no fast way, and the ways that look fast are the ones that get sites into trouble. Paid links and link exchanges are policy violations, and the links you buy are the ones Google is best at identifying and ignoring. The slow route is the only one: publish something worth citing — original data from your own product, a genuinely thorough guide, a free tool — and then tell the people who write about this subject that it exists. Note also that the same publications linking to competitor-a linked to something specific, so the first useful step is to find out what that something was.
Does higher schema coverage explain their AI citation rate?
There is no verified evidence that it does. Your schema coverage is 92% against their 74%, and their AI citation rate is higher than yours — which is the opposite of what the schema-density theory would predict, and a useful illustration of why that theory should be treated with suspicion. The defensible claim about structured data is narrower: retrieval systems generally parse served HTML rather than rendering it, so JSON-LD in the initial response reaches them and disambiguates what your text means. It clarifies; it does not persuade. What is likely driving their citation rate is that they have written more pages that directly answer the questions being asked.
If I am ahead on Core Web Vitals and schema, why am I behind overall?
Because those are hygiene measures and hygiene is a floor, not an advantage. Once a page is fast enough and correctly marked up, further improvement in either has sharply diminishing returns — Google has been consistent that Core Web Vitals act as a tie-breaker between comparable pages rather than lifting a weaker answer above a stronger one, and no schema property confers authority. Your 86% CWV pass rate and 92% schema coverage mean nothing technical is holding you back. The gap is in content and in the references that good content attracts, which is the uncomfortable finding in almost every competitor analysis and the only one that matters.
Related Demo Reports
Run Competitor Analysis on Your Own Site
Get your real competitor analysis with side-by-side authority metrics, keyword gaps, content gaps, link gaps and schema gaps.