The short version: you find three or four real competitors, pull their ranking keywords from a tool or a free workaround, then sort that list by what they rank for that you don’t, not the other way round. The mistake almost everyone makes is treating the whole keyword list as a shopping catalogue instead of asking which of those rankings bring the competitor business. Most of them don’t.
Pick the right competitors first, not the biggest ones
Before you touch a single tool, work out who you’re researching. I see this go wrong constantly. A client will say “let’s look at what Amazon ranks for” when they sell handmade candles from a workshop in Devon. Amazon isn’t your competitor. Your competitor is the other three candle makers who show up on page one of Google for “soy candles gift set UK” and who you’d lose a sale to.
I ask clients to write down five businesses first, then check each one against three questions:
- Do they sell roughly the same thing at a similar price point?
- Do they show up when you search your own target keywords, not just your brand name?
- Would a customer realistically compare you to them before buying?
If a business fails two of those three, drop it. This step alone saves hours later, because every hour you spend analysing the wrong competitor’s keywords is an hour wasted on someone else’s audience.
The three-tab method I use on every client project
This is the actual sequence, not a theory. I keep three browser tabs open and work through them in order.
Tab one: their top pages
Pull the competitor’s top-performing pages by traffic, not by keyword count. A tool like Ahrefs, Semrush or even the free Ubersuggest tier will show you this under “top pages.” I want to know which single pages are doing the heavy lifting. Nine times out of ten it’s two or three pages carrying 40 percent of the site’s organic traffic. That tells you where to point your energy.
Tab two: their full keyword list, filtered
Export the full list of keywords they rank for, then immediately filter out anything ranking below position 20 and anything with under 10 searches a month. This cuts a typical list from several thousand rows down to something workable, often under 300. If you want a proper walk-through of this exact filtering process, I wrote a longer breakdown in how to do competitor keyword research for your niche, which goes step by step rather than just listing tool names.
Tab three: the gap report
Now compare your own domain against two or three competitors at once using a content gap or keyword gap tool. This shows you keywords they all rank for that you don’t touch at all. This is the tab that changes what you write next, not tab two.
A real example, with the numbers
I did this for a small accountancy firm outside Manchester in 2025. Three partners, no in-house marketing person, ranking for their own name and not much else. We picked three competitors, all similar-sized local firms who came up for “small business accountant Manchester” and “self assessment help Manchester.”
The keyword export across those three competitors gave us 1,140 keywords combined before filtering. After cutting anything below position 20 and under 10 monthly searches, we were left with 267. The gap report between our client and the three competitors showed 34 keywords that all three ranked for and our client ranked for zero of. Things like “sole trader tax deadline calculator,” “how much to pay an accountant self employed UK,” and “CIS scheme accountant Manchester.”
We wrote four pages targeting the highest-intent terms from that list of 34. Within five months, two of those pages were on page one, and one of them, the CIS scheme page, brought in three new client enquiries in its first quarter live. That’s the whole point of this exercise. Not a spreadsheet with 1,000 rows. Four pages built from 34 keywords that mattered.
What the data won’t tell you, and why that matters more than the tool you pick
Here’s the part most guides on this topic quietly skip over. A big chunk of the keywords your competitor ranks for are not deliberate. They’re accidental. An old blog post from 2019 that happened to match a search trend nobody planned for. A page that ranks for a long-tail phrase the competitor’s writer never intended to target. If you export a competitor’s keyword list and treat every single row as a strategic decision they made on purpose, you’ll waste weeks chasing rankings that never brought that business a single customer either.
I’ve had clients get excited about a competitor ranking for some obscure keyword with 90 monthly searches, convinced it must be a secret goldmine, when really it’s just a page that got lucky and sits there doing nothing for the competitor’s revenue. The fix is simple but people skip it: cross-check any keyword you’re excited about against that competitor’s actual page. Read it. Does the page look built around that keyword, with a clear structure, internal links, and updates over time? Or does it read like a page written for something else entirely that Google happened to match anyway? If it’s the second one, don’t copy it.
This is also where a rank tracker earns its keep, because you can watch whether a competitor’s position on a given keyword is stable or wobbling. If they’ve been bouncing between position 8 and position 19 for six months, that’s not a keyword worth building a whole content strategy around. I go into this, including what tracking tools can’t show you, in how an SEO rank tracker helps you monitor keyword positions.
You don’t need a paid tool to start
People assume competitor keyword research needs a £99 a month subscription before you can begin. It doesn’t. Google’s own “People also ask” boxes and “related searches” at the bottom of a results page, when searched using a competitor’s brand name plus your industry terms, will surface real questions their audience is asking. Free tools like Google Search Console (if you have access to a shared client account, which happens more than people admit), Answer The Public’s limited free tier, and even just reading a competitor’s own site navigation menu will get you a workable starting list.
There are also legitimate ways to look at what’s happening on a competitor’s site without breaking any rules or spending anything, things like checking their sitemap.xml directly in a browser, looking at their blog category pages to see how they’ve grouped content, or using the Wayback Machine to see how their top pages have changed over time. I cover the full set of these methods in what are the legal ways to spy on competitors’ websites, and there’s a separate walk-through specifically on pulling their ranking keywords without paying for anything in how to see which keywords your competitors rank for without paying.
Turning the list into something that beats them
Once you’ve got your filtered gap list, the job isn’t to write the same page they wrote. It’s to write a better one. Look at what their top page for a given keyword is missing. Is it thin, under 600 words, with no examples? Is it from 2021 and never updated? Is it missing a clear answer near the top of the page? Build yours to fix exactly that gap.
If you’re producing a lot of these pages at once, which is often the fastest way to close a big gap quickly, this is where having a proper content system, whether that’s a small team or an AI-assisted workflow, changes how fast you can move. I’ve written about setting this up at scale in building an AI content team to rapidly outperform your competitors, and if you want the free-tool version of the whole keyword research process rather than the competitor-specific one, that’s covered separately in how to do keyword research for free without expensive tools.
One more thing worth saying plainly: if you’re doing this for a business with real revenue on the line and you don’t have the time to sit through 300 rows of exported data every quarter, it’s worth getting an outside pair of eyes on it rather than guessing. That’s a fair reason to bring in someone who does this daily rather than trying to teach yourself from scratch while also running the business.
How often to redo this
Competitor rankings shift. A quarterly check is enough for most small businesses, monthly if you’re in a fast-moving space like local trades or ecommerce where a competitor might launch a new product line and suddenly rank for 40 new terms overnight. Set a calendar reminder. Nobody does this consistently unless it’s scheduled, including me, and I’ve missed a quarter or two myself when client work piled up.
Frequently asked questions
How many competitors should I research at once?
Three to five is the right range. Fewer than three and you won’t spot patterns in the gap report. More than five and the list gets so long you never act on any of it. Pick the competitors you’d lose a customer to, not the biggest brand in your industry.
Is it legal to research a competitor’s keywords?
Yes. You’re looking at publicly available search results and publicly available pages, the same way any customer or journalist could. There’s a clear line between this and anything involving hacking, scraping login-protected data, or misrepresenting who you are, and none of that is needed here.
What’s the biggest mistake people make with competitor keyword lists?
Treating every keyword on the list as equally valuable. A lot of what a competitor ranks for is accidental, low-intent, or wobbling in position rather than stable. Check the actual page behind any keyword that excites you before you build a strategy around it.
Do I need a paid tool like Ahrefs or Semrush to do this ?
No, though it speeds things up considerably. You can build a workable competitor keyword list using free tools, Google’s own search features, and manual page checks. It takes longer but gets you 60 to 70 percent of the way there for free.
Related reading: How Do You Assess a Content Discovery Advertising Platform Before Spending Money and How Do You Choose the Best Hashtags for Your Instagram Posts in 2026.
If you want the full breakdown, here is everything I know about SEO.
