AI Competitor Research
Identifies the real organic rivals competing for your audience and scores the pressure they put on you — traffic, share of voice, top-3 rankings and keyword overlap.
Last updated 2026-08-06
Summary#
AI Competitor Research finds the domains genuinely competing with you for the same organic audience and measures each one: organic keywords, monthly traffic, share of voice, top-3 rankings and how many keywords you share. It rolls that into a single competitive pressure score. Its internal tool id is ai_competitor_research.
Purpose#
Most companies keep a competitor list built from sales calls and conference programmes. The set of domains actually taking your search traffic is usually different, and often smaller than you think.
This tool exists to produce the measured version of that list. The decision it supports is where to point your effort: which single rival to attack first, and whether the category is dominated by one player or spread across many.
Overview#
You give it either a domain or a category keyword, and the two paths differ.
Given a domain, Metric Vault reads that domain's own organic performance, then pulls the competitor set search data associates with it — up to 50 candidates — excluding the giant general-purpose sites that appear for everything. Each candidate carries a shared-keyword count with you.
Given a keyword, it reads the live top 20 results for that query and treats the distinct domains there as the competitive set.
Either way, each competitor is then enriched with its own organic metrics, sorted by monthly traffic, and given a share of voice — its slice of the combined traffic of the measured set.
The pressure score combines three real measurements: how much of that traffic the top rival holds, how deep your keyword overlap with the set runs, and how large your own share is. High pressure means a small group is taking most of the traffic in the keyword universe you share.
Despite the name, this tool measures organic search competition, not per-engine AI visibility. The report says so, and points at Visibility Overview for the AI side.
Benefits#
- A competitor list built from measurement rather than memory.
- Share of voice puts each rival's size in proportion.
- Shared-keyword counts identify who competes with you directly versus who is merely large.
- Every rival card carries a written attack plan and a one-click handoff into Gap Finder.
- Works from a keyword when you do not have a domain, so you can size a market you are not in yet.
- Costs 1 credit, which keeps it under the monthly premium threshold.
Use Cases#
Finding out who you are actually competing with. The measured set often contains names your sales team has never heard.
Choosing an attack target. The rival with the widest keyword overlap is the fastest place to win, not the largest rival.
Sizing a market before entering it. Run it on a category keyword to see how concentrated the traffic is.
Reporting competitive position. Share of voice and pressure band answer "how are we doing against them" in one number each.
Feeding the growth queue. This is one of the source tools Growth Actions reads.
Requirements#
- A signed-in account.
- A paid plan — 1 credit per run, and any cost above zero is blocked on Free.
- A domain or keyword with measurable organic presence. A brand-new site or an obscure phrase returns an honest empty result rather than an invented one.
- Available credits and hourly headroom.
Permissions#
Any workspace member on a paid plan can run it. Free accounts are refused with This tool needs a paid plan. Free includes the 10 technical SEO tools; upgrade to Pro to unlock the rest. The recommendations panel under the result requires Pro or above — see Get Recommendations.
Cost#
1 credit per run. The run button reads Analyze AI Competitors · 1 report.
Below the premium threshold of 3, so the monthly quota never blocks it — only the hourly fair-use cap of 100 light-tool calls per hour. A shared-cache hit still charges the credit; the cached window for this tool is two days. Some of the underlying data, notably the competitor set itself, is cached for up to a week. See How credits work and Result caching and freshness.
Navigation Path#
Dashboard → AI Visibility → Competitor Research
The in-view title is AI Competitor Research.
Inputs#
| Field | Accepts | Required | Default | Validation | Notes |
|---|---|---|---|---|---|
e.g. crm software or salesforce.com | A domain or a category keyword | Yes | Empty | Empty input shows Enter a value first. | The two are handled differently. A domain gives you shared-keyword counts and your own metrics; a keyword gives you the live top-20 domains and no personal comparison |
TRY: chips (crm software, salesforce.com, project management, asana.com) | — | — | — | — | Clicking one fills the field and runs immediately, charging a credit |
| Country picker (topbar) | Country list | No | United States | — | Sent with the run |
Pressing Enter in the field runs the tool.
Step-by-Step Guide#
- Open
AI Visibilityand choose Competitor Research. - Enter your own domain in the field marked
e.g. crm software or salesforce.com, or enter a category keyword. - Click Analyze AI Competitors · 1 report, or press
Enter. - Wait for
Analyzing… this can take a few seconds for live data.to clear. - Read the pressure gauge and the plain-English note beneath it.
- Work through
Top Attack Targetsfrom the top card down. - Use Run Gap Finder on the rival you have chosen to get the exact keyword list.
- Export or share the report from the buttons at the top of the view.
Reading the Results#
The header. Your domain, a Competitor Intelligence badge and a DataForSEO Labs badge, then a summary line: how many rivals were measured, which one is the top threat, and what share of voice it holds. It closes by pointing you at Visibility Overview for the per-engine AI breakdown — because that is not what this report measures.
The five KPI cards.
| Card | Reads | Good | Bad | Action |
|---|---|---|---|---|
Pressure | The score out of 100 with its band | Low | High | See the band table below |
Your Share | Your slice of your traffic plus the competitor set's | 25%+ | Under 5% | This is the honest size of your position in the measured universe |
Your Traffic | Your estimated monthly organic visits | — | 0 | A zero here means the target was a keyword, not your domain |
Your Keywords | Your indexed organic keyword count | — | — | Compare against the rivals' counts in the table below |
Competitors | How many rivals were measured | 8 or more | 1 or 2 | A thin set usually means a very narrow niche or a weak domain signal |
Your Share, Your Traffic and Your Keywords read zero when you entered a keyword rather than a domain. That is expected — there is no "you" to measure.
Competitive Pressure Score. A 200-pixel gauge with the score, the band, and a sentence of plain English.
| Score | Band | What it means | What to do |
|---|---|---|---|
| 70-100 | High | A small set of rivals is capturing most of the traffic in your shared keyword universe | Concentrate. Pick one rival and take keywords from them rather than spreading effort |
| 40-69 | Moderate | Competition is real but distributed, with no single dominant rival | Pick the rival with the most shared keywords and focus there |
| 0-39 | Low | You either dominate the niche or it is highly fragmented | Scale content. There is no concentrated threat to defend against |
The note under the gauge says the same thing in the product's own words. Quote it in your reporting — it is the interpretation, not a metric you have to explain.
Share of Voice — Top Rivals. A bar per rival showing its share of the measured set's combined monthly traffic, badged REAL TRAFFIC. Read the shape: one bar far ahead of the rest means a category owner and a clear target; a flat set of similar bars means a fragmented market where content volume wins.
Top Attack Targets — ranked by real overlap. The most actionable block, showing up to five rivals ranked by how many keywords you share. Each card carries an overlap label:
| Label | Shared keywords | What it means |
|---|---|---|
Direct rival | 5,000 or more | You are fighting over the same audience on most queries |
Adjacent rival | 1,500-4,999 | Substantial overlap; a real competitor in part of the market |
Niche overlap | 300-1,499 | You meet on a specific segment |
Loose overlap | Under 300 | Large or visible, but not competing with you directly |
Each card shows four figures — Shared kw, Traffic, SOV, Top-3 — a bar showing its overlap relative to the other targets, and a written attack plan. For a rival above 1,500 shared keywords the plan is to run Gap Finder and clone their top-3 page format on the highest-volume terms you do not rank for; below that it is to pull their top pages and replicate the format where it fits your audience.
Two buttons finish each card: Run Gap Finder, which carries both domains into Gap Finder, and Visit site.
The right target is rarely the largest rival. A Direct rival with moderate traffic is a better first move than a Loose overlap giant, because you already rank on the same terms and only need to move positions.
All Competitors. Every measured rival in a table:
| Column | What it means | How to read it |
|---|---|---|
Domain | The rival, linked | Click through to see what they actually publish |
Keywords | Their indexed organic keyword count | Compare against your own KPI card |
Traffic | Their estimated monthly organic visits | Estimates — use them to rank rivals, not to quote absolutes |
SOV | Share of voice, with a bar | Their slice of the measured set's traffic |
Top-3 | Positions 1 to 3 they hold | This is the number that hurts. Top-3 rankings are where the clicks are |
Shared | Keywords you both rank for | The directness of the competition |
A rival with high traffic and low shared keywords is big but not in your way. A rival with moderate traffic and high shared keywords is directly in your way. Sort your thinking by the Shared column.
What This Tells Us. Written findings synthesised from the measured numbers, badged REAL DATA. One names the leading rival and its share; one compares your own traffic and keyword count against the set when you entered a domain; one flags the top rival's top-3 count as the rankings to target.
Recommended Action Plan. Four ordered steps, badged PRIORITIZED: run Gap Finder against the leading rival, audit the content behind their top-3 rankings, target the rivals with the most shared keywords first, and track these domains in Competitor Manager with a monthly re-run.
What the report deliberately does not include. Per-AI-engine visibility has no measured source in this data, so it is omitted rather than estimated. Run Visibility Overview for that.
What to do with it. Pick the top attack target. Run Gap Finder against it. Take the twenty-five highest-volume terms where they rank and you do not, and build those pages. Re-run monthly and watch your share of voice.
Examples#
Example: You run ourcrm.com. The header reports 11 rivals measured, top threat rival-a.com at 34% share of voice. Pressure reads 72/100 — High, with the note that a small set of competitors is capturing most of the traffic in your shared keyword universe. Your Share is 6%, Your Traffic 18,400, Your Keywords 4,900. The share-of-voice bars show one dominant rival and a long tail. Top Attack Targets ranks rival-b.com first — 6,200 shared keywords, Direct rival, 41,000 monthly traffic, 18% SOV, 820 top-3 positions — with the plan to run Gap Finder and clone their top-3 page format. You click Run Gap Finder, get the term list, and build the twenty-five highest-volume pages you are missing.
Screenshots#
Tips#
- Enter your own domain, not a competitor's, if you want the shared-keyword counts and the
Your Sharefigures. - Sort your attention by the
Sharedcolumn, not byTraffic. Overlap is what makes a rival beatable. - Use a category keyword when you are evaluating a market you have not entered — the live top 20 is a fair proxy for who owns it.
- Pair this with Visibility Overview on the same brand. Organic pressure and AI visibility often tell opposite stories, and the gap between them is the strategy.
- Re-run monthly. Competitor sets move slowly, and the data behind this tool is cached for up to a week.
Best Practices#
- Attack one rival at a time. The report ranks them so you do not have to spread effort.
- Take the attack plan literally: pull their top-3 pages, match the format and depth, then beat it.
- Add the measured rivals to Competitor Manager so subsequent tools reuse the same list.
- Treat traffic figures as relative. They are estimates that rank rivals correctly, not analytics you should quote to two significant figures.
Common Mistakes#
- Expecting per-engine AI visibility here. This tool measures organic search competition; the report states plainly that per-engine AI data is omitted rather than fabricated.
- Chasing the largest rival. A
Loose overlapgiant shares almost none of your keywords. - Entering a keyword and then wondering why
Your Shareis 0. With no domain there is nothing of yours to measure. - Reading
Competitors 3as a good sign. A thin set usually means a weak domain signal or a very narrow niche, not an empty market. - Ignoring the
Top-3column. Positions 4 to 10 earn a fraction of the clicks of positions 1 to 3.
Limitations#
- Competitor and traffic data come from Google's United States, English-language index. The country picker does not change the underlying market.
- Very large general-purpose domains are excluded from the competitor set on purpose; they appear for everything and tell you nothing.
- From a domain, up to 50 candidates are considered and the top 12 by traffic are measured. From a keyword, the live top 20 results yield up to 8 domains.
- Traffic, keyword counts and top-3 counts are provider estimates.
- Shared-keyword counts are only available on the domain path. Running from a keyword produces no overlap figures.
- The pressure score is a formula over those real measurements, not a market study.
- No per-AI-engine visibility is included.
- Saved results are kept for 90 days.
Troubleshooting#
| Symptom | Likely cause | Fix |
|---|---|---|
Enter a value first. | Empty field | Type a domain or a keyword |
Please sign in to run this tool. | Signed out | Sign in and retry |
This tool needs a paid plan… | Free plan | Upgrade — see Plans and pricing |
No measured data was found for this query. | No competitor with measurable organic metrics was found | Check the spelling, or try a more established domain or a broader keyword. See A tool returned no data |
Your Share, Your Traffic and Your Keywords all read 0 | You entered a keyword, not a domain | Enter your domain to get your own metrics |
Shared is 0 for every rival | You ran from a keyword, so no overlap was computed | Re-run from your domain |
| Only two or three competitors | A narrow niche, or a domain with a thin ranking footprint | Try a category keyword to see the wider market |
| A rival you expect is missing | It was filtered as a mega-domain, or it has no measurable organic overlap | Track it manually in Competitor Manager |
| Same numbers as yesterday | Shared cache hit — two days for this tool, and up to a week for the competitor set | Expected. See Result caching and freshness |
Hourly fair-use limit reached… | Over 100 light-tool calls this hour | Wait for the hourly reset |
FAQs#
Should I enter my domain or a keyword? Your domain, in almost every case. It gives you shared-keyword counts, your own traffic and keyword totals, and a real Your Share figure. Use a keyword only when you are sizing a market you are not yet in.
How is the pressure score calculated? From three real measurements: the top rival's share of voice, the depth of your total keyword overlap with the set, and your own share. It rises when rivals concentrate the traffic and falls as your own share grows.
What exactly is share of voice here? Each rival's estimated monthly organic traffic as a percentage of the combined traffic of the measured competitor set. It is a relative measure within that set, not a measure of the entire category.
Why does this sit under AI Visibility if it measures organic search? It is the competitive half of the AI-visibility workflow: you use it to identify who to compare yourself against, then use Visibility Overview to see how those same rivals fare inside AI answers. The report's own note is explicit that per-engine AI data is not included.
Why are big sites like marketplaces missing? They are filtered out. They rank for everything, so including them would swamp the share of voice and hide the rivals you can actually act against.
How does this differ from Competitor Battle? Competitor Battle compares two named domains head to head in depth. This tool discovers who your competitors are and ranks them.
Can I export the competitor table? Yes. Export PDF, Export Excel, Export CSV, Export JSON and Share public link are at the top of the view.
See also
Was this article helpful?
Thanks — feedback noted for the docs team.