Competitor Keyword Analysis: How I Actually Do It (After 7+ Years in SEO)

I have run competitor keyword analysis for DTC supplement brands, pet food companies, fintech startups, podcast advertising agencies, and my own consultancy. Same core process every time. But I need to be honest with you up front: the game this analysis feeds into has fundamentally changed, and most of the articles ranking for this topic are still describing a version of SEO that no longer exists.

Here’s what changed. AI Overviews now sit on top of the search results and answer informational questions before anyone clicks. Tools like ChatGPT and Claude have become the starting point for millions of searches that used to happen on Google. Top of funnel traffic, the lifeblood of the old bulk-blogging playbook, is getting absorbed before it ever reaches a website.

I’ve watched it happen in Search Console across the sites I manage. Informational queries that reliably drove clicks for years now drive impressions and nothing else. The click just doesn’t come. The AI answered it.

So why write a guide on competitor keyword analysis at all? Because the analysis matters more now, not less. When the total pool of clicks shrinks, you cannot afford to spend a single article on the wrong keyword. Precision is the whole game. And the fastest path to precision is knowing exactly which high intent search queries your competitors are converting on, and which ones they’ve left exposed.

This guide covers my actual workflow, my honest opinions on the tools, the KPI hierarchy I use to measure what matters, and how to run this analysis for both traditional search engine optimization and the new layer people are calling AEO and GEO. (and these are the differences)

What Is Competitor Keyword Analysis?

Competitor keyword analysis is the process of identifying the keywords your competitors rank for in organic search, then using that data to find gaps and opportunities in your own SEO strategy.

You’re reverse engineering the search engine visibility of the sites that are beating you, so you can figure out where they’re strong, where they’re weak, and where they’ve left the door open.

In practice, it involves three phases: collecting competitor keywords, cleaning the data, and assessing what’s worth targeting. Most people do the first phase and skip the other two. That’s why most competitor keyword research ends up as a spreadsheet nobody opens again.

In the AI era, I’d add a fourth phase: filtering everything through intent and business value, ruthlessly. A keyword gap analysis used to produce a content calendar. Now it produces a shortlist, because half the keywords in any gap report are informational queries that AI Overviews will answer without sending you a click. The analysis got harder to act on and more valuable at the same time.

The KPI Hierarchy: Measuring What Matters

A buddy of mine has been doing SEO as long as I have, and we recently talked through how we each measure success now. We landed on the same hierarchy independently, which tells me it’s real:

1. Conversions. The top KPI, full stop. Revenue, leads, signups, whatever the business counts as money.
2. Clicks. Real humans arriving at the site, measured in Search Console.
3. Page rank. Your ranking positions for the keywords targeted. Rankings matter, but only as a means to clicks.
4. Search volume. Dead last. A directional input, never a goal.

Read that order again, because the entire industry is built backwards. Agencies sell search volume projections. Reports celebrate rankings. Dashboards glow with traffic curves. And meanwhile the client is asking the only question that matters: did any of this make us money?

We’re not here for vanity metrics. We’re here to measure what matters. Every step of the workflow below runs through that filter.

This hierarchy also explains why AI Overviews hurt less than the panic headlines suggest, if your strategy is built right. AI Overviews are eating informational clicks at the top of the funnel. They are not eating the searcher who types “best probiotic for women over 50” with a credit card in hand. Targeted high intent search queries mean more than volume ever did, and the sites feeling the most pain right now are the ones that built their entire traffic base on high-volume informational content. That model is over.

Why Competitor Keyword Analysis Still Matters

Filtered through the conversions-first lens, here’s what this analysis actually buys you:

It reveals the keyword gaps that convert. A keyword gap analysis shows you keywords your competitors rank for that you don’t. The gold is not the biggest gap. It’s the commercial gap: the bottom-funnel search terms a competitor is quietly converting on while you chase blog traffic.

It exposes content gaps worth filling. Not every content gap deserves content anymore. But when competitors cover a topic that maps to your product and carries real purchase intent, identifying that gap tells you exactly what to create to meet your target audience at the moment they’re ready to act.

It uncovers the language of buyers. Your customers don’t use your internal vocabulary. Competitor data shows you the relevant search terms real people type into Google search when they’re close to a decision.

It tracks market trends before they’re obvious. Monitoring competitors’ keyword strategies over time helps you adapt to market changes. When a competitor starts ranking for a new cluster of search queries, the market is telling you where demand is moving.

It benchmarks reality. Comparing your search rankings against organic competitors tells you where you actually stand in the search engine results pages, not where you think you stand.

Step One: Identify Your Actual SEO Competitors

This is where most people get it wrong on day one.

Your business competitors and your SEO competitors are not the same list. A supplement brand thinks its competitor is the other supplement brand at the trade show. But when you look at the search results for their target keywords, the sites actually outranking them are Healthline, a Reddit thread, and a niche blog run by one person.

Those are the organic competitors. Those are the competitors’ websites you analyze.

My process:

1. Take your 10 to 15 most important target keywords, weighted toward commercial intent
2. Search each one and note who appears in the top 10 of the search engine results
3. Any domain that shows up repeatedly across multiple search queries is an SEO competitor

You can shortcut this in Ahrefs with the Organic Competitors report inside Site Explorer. Drop in your own website and it shows which domains overlap with your keyword rankings the most. I still recommend manual searches anyway, because in 2026 the SERP itself is evidence. Note which queries trigger an AI Overview and which don’t. A keyword where the AI Overview answers everything above the fold is worth less than the tools will ever admit.

One more thing to check while you’re in there: who is getting cited inside the AI Overviews for your money keywords. That’s a new competitor list layered on top of the old one, and it matters for the AEO section coming up.

Step Two: Pull Their Organic Keywords

Take each competitor’s domain and run it through your keyword research tool. Mine is Ahrefs, and I’ll explain why in the tools section.

In Ahrefs, I drop the competitor’s URL into Site Explorer and open the Organic Keywords report. This shows every keyword the domain ranks for, its position, its estimated search volume, and the estimated organic traffic each keyword drives.

For a typical analysis, I export everything and filter down:

• Positions 1 through 20. If they rank on page one or two, the keyword is winnable and worth evaluating.
• A minimum search volume floor, held loosely. Usually 50 to 100 average monthly searches. In B2B or high-ticket niches I go lower, because a keyword with 30 searches a month and buying intent can outearn a keyword with 30,000 and none. Remember the hierarchy: volume is fourth.
• Branded terms removed. You don’t want to rank for their brand name. Strip those immediately. This is the “cleaning” phase everyone skips.

Repeat for each competitor. I usually analyze three to five competitors’ domains per project. More than that and you’re drowning in duplicate data.

Step Three: Run a Keyword Gap Analysis

This is the highest-leverage step in the entire process.

A keyword gap analysis (Ahrefs calls it Content Gap) compares your domain against your competitors and shows you the specific keywords they rank for that you don’t. Ahrefs lets you stack up to four competitors against your own site at once, which is more than enough. If a keyword shows up for three or four competitors and your site is nowhere, that’s not a coincidence. That’s a hole in your strategy.

I sort the gap report two ways:

1. By number of competitors ranking. Keywords where multiple competitors rank are validated topics.
2. By traffic potential. Ahrefs shows the estimated traffic to the top ranking page, which I trust far more than raw search volume. More on that in a minute.

Then I add the 2026 pass: for the top opportunities, I check the live SERP. Heavy AI Overview presence on an informational query pushes it down my list. Clean commercial SERP with shopping intent pushes it up. The gap report tells you where the openings are. The SERP tells you which openings still pay.

Step Four: Score Every Keyword for Business Value

This is the step that turns the whole exercise into revenue, and it’s built directly on the KPI hierarchy.

I assign every keyword a business value score from 0 to 3:

• 3: Someone searching this is ready to buy what my client sells. Bottom of funnel, high commercial intent. These map straight to the top KPI: conversions.
• 2: Strong commercial relevance. The searcher has the problem we solve and is researching solutions.
• 1: Topically related. Useful for topical authority and AI citation presence, unlikely to convert directly.
• 0: Irrelevant. A competitor ranks for it, but it does nothing for this business. Delete it.

Then I prioritize using three inputs together: business value first, search intent second, search volume last. Never volume alone. Never volume first.

A keyword with 200 monthly searches and a business value of 3 beats a keyword with 20,000 searches and a business value of 1 every single time. Higher search volume means more potential traffic, sure. But in a world where AI Overviews are siphoning informational clicks, that potential is more theoretical than ever. Traffic that doesn’t align with what you sell was always just server load. Now you often don’t even get the traffic.

Understanding user intent is the tiebreaker. Before I commit to any particular keyword, I search it myself and look at what Google is rewarding. Product pages? Listicles? In-depth guides? An AI Overview with three citations? The keyword SERP tells you what format to create and whether a click is even available to win.

While I’m in the SERP, I study how the top-ranking competitor content is structured. What questions they answer, what they miss, whether they’re the source the AI Overview is quoting. That last part is the new tell. If a competitor’s page is getting cited by the AI layer, their structure is worth understanding.

The Truth About Search Volume (What the Tools Won’t Tell You)

This is the part of the article I care about most, because almost nobody with real Search Console access across dozens of sites is willing to say it plainly. And it’s exactly why search volume sits fourth in my KPI hierarchy.

Every keyword tool’s search volume numbers are wrong. The question is how wrong, and in which direction.

I have Search Console access to a lot of websites. Client sites, my own sites, sites I’ve audited. For years I’ve compared tool estimates against real clicks and impressions in 28-day Search Console performance reviews. Here’s what I’ve found:

Ahrefs runs closer to reality. Its monthly search volume estimates, and especially its traffic estimates for ranking pages, land nearer to the actual clicks I see in Search Console than any other tool I’ve tested. Ahrefs claims a 95% prediction rate on search volume, and while I’d never take a vendor’s accuracy claim at face value, my own Search Console comparisons back up that it’s directionally the most honest tool available.

Semrush inflates. I’ll just say it. A keyword that realistically drives 100 clicks a month will show up in Semrush as 1,000. The numbers are consistently bloated, and if you build traffic projections off them, you will overpromise and underdeliver. I’ve watched it happen.

But here’s the mental model I want you to walk away with:

Both tools work as a compass. Neither works as a GPS.

Run this experiment. Look up “dogs” and “cats” in both tools:

• Ahrefs estimates the #1 ranking spot for “dogs” gets around 785,000 in traffic. For “cats,” around 664,000.
• Semrush shows “dogs” at some absurd volume like 1 billion, and “cats” at 500 million.

Neither absolute number is real. But both tools agree on the same directional truth: more people search for dogs than cats. That’s the signal. The tools are accurate on relative user search trends and inaccurate on absolute search volume and actual traffic. And in the AI Overview era, the gap between “estimated volume” and “clicks you’ll actually receive” is wider than it has ever been, because the tools count the searches but the AI intercepts the clicks.

Real traffic is measured in clicks and impressions inside Search Console, full stop. Which brings up the eternal debate: GA4 or Search Console as the source of truth for organic? Most SEOs I’ve worked with across my career agree. Search Console is the true north star. GA4 has its uses, but when I want to know what Google is actually sending to a site, I look at Search Console. Then I follow those clicks into conversions, because clicks are KPI number two and conversions are number one.

Use the tools to compare, rank, and prioritize keywords against each other. Never quote their raw numbers to a client as a traffic promise.

Competitor Keyword Analysis Tools: My Honest Take

Every article on this topic lists eight tools with identical enthusiasm because they’re chasing affiliate commissions. I’m not doing that. Three tools are worth your attention, and here’s what I actually think of each.

Ahrefs (What I Use)

Ahrefs is my only keyword research tool, by choice. Site Explorer for pulling a competitor’s organic keywords, Content Gap for keyword gap analysis, Keywords Explorer for expanding the list, and traffic estimates that hold up best against real Search Console data. Keyword difficulty scores are directionally useful for gauging how hard a SERP is to crack. If you’re going to pay for one SEO tool, this is the one.

Semrush

The most popular alternative, with keyword data from over 140 countries and a keyword gap tool that also compares up to four competitors against your domain. It’s a capable platform and plenty of good SEOs use it. My issue is the one I covered above: the search volume numbers are bloated, and I don’t build strategy on inflated inputs. If you use it, use it as a compass and validate against Search Console.

Google Keyword Planner

Free if you have a Google Ads account, and it pulls average monthly searches straight from Google. The catch: it’s built for paid keywords, it buckets volumes into wide ranges unless you’re actively spending, and it tells you nothing about who ranks organically. Fine as a free supplement. Not a competitor analysis tool on its own.

The honest summary: pick one primary tool, learn it deeply, and validate everything against Search Console. Tool-hopping is procrastination dressed up as diligence.

AEO and GEO: Same Game, New Scoreboard

Let me save you from a lot of LinkedIn noise. AEO (answer engine optimization) and GEO (generative engine optimization) are SEO evolving under a new name. The fundamentals did not change. Clear structure, genuine expertise, content that actually answers the question, and authority signals like backlinks. What changed is where the answer gets delivered.

When someone asks ChatGPT or Claude for the best option in your category, those models are drawing on content and citations from across the web. When Google’s AI Overview summarizes a topic, it cites sources. Your competitor keyword analysis now has a second dimension: not just who ranks in the classic search results, but who gets cited and recommended by the AI layer.

Practically, here’s what I do differently:

Structure content to be quotable. Direct answers near the top, clear headings that mirror real search queries, specific facts and numbers the models can lift and attribute. The same structure that wins featured snippets wins AI citations.

Cover the bottom of the funnel with real substance. AI assistants are increasingly where comparison and recommendation questions get asked. If your competitor is the answer the model gives and you’re not mentioned, that’s a keyword gap you’ll never see in a ranking report.

Track brand visibility inside the AI tools themselves. I run recurring checks on how brands show up across ChatGPT, Claude, and Perplexity for their money queries. It’s the same competitive research instinct, pointed at a new surface. Ask the AI tools your own target keywords and see who they name. That’s your AEO competitor set.

Don’t abandon classic SEO to chase this. The models learn from the same web Google crawls. The sites winning AI citations are overwhelmingly the sites that were already doing SEO right. AEO is not a replacement discipline. It’s the same discipline with a second scoreboard.

Turning the Analysis Into a Content Strategy

A keyword list is not a strategy. Here’s how I convert the analysis into output, in priority order.

Start at the bottom of the funnel. Business value 3 keywords first, always. In the old model you could afford to build top-of-funnel traffic and nurture it down. With AI Overviews absorbing informational clicks, the informational layer converts worse than ever. Win the high intent search queries first, then expand upward selectively.

Build topic clusters, not one-off posts. I stopped building content strategies around individual keywords years ago. I group the prioritized keywords into clusters: one pillar page targeting the head term, supporting articles targeting related search queries, all internally linked. Google ranks topical authority, AI models cite topical authorities, and clusters are how you build it. Bulk blogging against isolated keyword volume is obsolete. It was dying before AI Overviews. It’s dead now.

Fill content gaps with genuinely better content. The commercial topics your competitors own and you’ve ignored become your content calendar. Create content that fills the information gap for real, not a reworded version of what already ranks. The AI layer makes this non-negotiable, because derivative content gets summarized and skipped while original insight gets cited.

Optimize existing content for missed keywords. The fastest win in SEO and almost everyone skips it. The gap analysis often reveals keywords where you already have a page sitting on page two. Reoptimizing that page, updating the title tag and meta description, expanding sections to match search intent, will move rankings faster than any new article. I’ve used this exact workflow to lift pages from position 15 to the top five in weeks. And because it targets pages with existing traction, it moves clicks and conversions, not just rankings.

Learn from their structure, then beat it. Where a competitor ranks with thin content, go deeper. Where they rank with a wall of text, go clearer. The goal is not to match the SERP. It’s to give Google, and the AI models reading over Google’s shoulder, an obviously better answer.

Keep Monitoring (This Is Not a One-Time Project)

Effective competitor analysis is ongoing. Rankings shift, competitors publish, algorithms update, AI Overviews expand into new query types, and the way people search keeps migrating into chat interfaces. The competitive landscape you mapped in January is not the one you’re competing in by June.

My cadence: a full competitor keyword analysis quarterly, plus a monthly check on keyword rankings and Search Console clicks for the terms that matter most, plus periodic checks on AI visibility for money queries. Tracking competitor keywords over time is how you spot emerging trends early, catch a competitor’s new content push before it buries you, and keep your keyword strategy pointed at demand that still converts.

And measure in the right order. Did conversions move? Did clicks move? Then rankings, then everything else. If a report leads with search volume, someone is selling you vanity.

The Bottom Line

Competitor keyword analysis is the fastest way to find keywords with proven demand, because your competitors already did the validation for you. But the era of turning that analysis into a hundred blog posts and a hockey stick traffic chart is over. AI Overviews and tools like ChatGPT and Claude have compressed top of funnel traffic, and the winners now are the sites that target fewer, higher intent keywords and convert them.

The workflow: identify your real SEO competitors, pull their organic keywords in Ahrefs, run a keyword gap analysis, score everything for business value and intent, and build clusters around what survives. Treat every search volume number as a compass rather than a fact. Let Search Console be your source of truth for clicks, and let conversions be your definition of success.

Conversions first. Clicks second. Rankings third. Volume last. Measure what matters, and the strategy builds itself.

If you’d rather have someone who’s done this a few hundred times run it for you, that’s literally what I do. Book a call and I’ll show you exactly where your competitors are beating you, where the AI layer is quoting them instead of you, and where they’re exposed.

FAQ

What is the difference between competitor keyword analysis and competitive analysis?

Competitive analysis is the broad exercise of studying everything your competitors do, from pricing to positioning to marketing channels. Competitive keyword analysis is one slice of that: a focused look at the SEO keywords and search terms your search competitors rank for. In my workflow, it’s the highest ROI slice, because it’s built on hard ranking data instead of assumptions.

What is the best way to analyze competitor keywords?

Start with your top competitors, meaning the three to five domains that repeatedly outrank you for commercial terms, and analyze competitor keywords through a tool like Ahrefs Site Explorer. Export their organic keywords, remove branded terms, then filter by intent and business value before you look at volume. You gain valuable insights only when you clean and score the data. A raw export is noise. A scored shortlist is a strategy.

How do I find competitor keywords without paid tools?

You can discover keywords manually by searching your most important terms and studying who ranks, then reading their titles, headings, and page structure to see the keywords targeted on each page. Google Keyword Planner is also free for anyone with a Google Ads account, and it will show you average monthly searches for relevant keywords. It won’t tell you what your competitors rank for, though. For that, you need a tool like Ahrefs.

Are there budget tools for competitive keyword research?

Yes, though I don’t personally use them. KWFinder provides access to a database of around 200 million keywords with a simple interface, and SE Ranking offers a free version with limited data access plus rank tracking. Both are workable entry points for competitive keyword research on a small site. Just know their competitor data is thinner than Ahrefs, and whatever tool you pick, validate the numbers against Search Console.

How do I identify keywords that are actually worth targeting?

Identify keywords by working backwards from revenue, not forwards from volume. The most valuable keywords are the ones where the searcher is close to a decision your business can fulfill, even if the monthly search numbers look small. Score every candidate 0 to 3 for business value, check the live SERP for intent and AI Overview presence, and cut anything that only produces traffic without a path to conversion.

Should I target all the keywords my competitors rank for?

No, and this is one of the biggest mistakes I see. Pulling all the keywords from a competitor’s domain gives you raw material, not a plan. A good chunk of what any site ranks for is branded, irrelevant, or low value, and in the AI Overview era, a lot of the informational keywords won’t send clicks anyway. Score everything for business value and search intent first, then invest your seo efforts only in the keywords that can realistically drive conversions or build topical authority.

How is competitor keyword analysis different for SEO vs PPC?

The process is similar but the data source changes. For search engine optimization, you analyze the organic keywords driving your competitors’ search rankings. For paid, you analyze ppc keywords, meaning the terms competitors actively bid on in Google Ads. The lists rarely match, because paid teams chase conversion terms while organic rankings accumulate across the whole funnel. I run them as separate exercises and compare afterward, since a keyword a competitor pays for month after month is usually a keyword that converts.

What is organic traffic research?

Organic traffic research is the practice of studying where a website’s search traffic actually comes from: which keywords drive clicks, which pages capture them, and how that compares to competitors. I run it in two layers. Ahrefs for the competitive view of any domain’s estimated traffic, and Search Console for the ground truth on my own sites. Comparing the two is where the valuable insights live, because the gap between estimated and real traffic tells you which keywords are being intercepted by AI Overviews and which still deliver clicks.

How long does it take to see results from competitor keyword analysis?

The analysis itself takes me a few hours per competitor. Results depend on what you do with it. Reoptimizing existing pages for missed relevant keywords can move search rankings in a few weeks. Net-new content built around gap keywords typically takes three to six months to mature, depending on your site’s authority and how competitive the SERP is. Either way, judge the results by conversions and Search Console clicks, not by ranking screenshots.