Most businesses look at their own rankings and stop there. That misses the main point. The pages taking your best traffic are often not the companies you compete with day to day. They are the sites Google has decided answer the search better right now. If you want more qualified visits, competitor keyword research is how you find those pages, see why they win, and decide which gaps are worth fixing.

For a small business, this is not a quarterly spreadsheet exercise. It is a way to stop guessing. You use it when leads from search have flattened, when a rival seems to appear everywhere, when your service pages are not ranking outside your brand name, or when AI answers keep naming other businesses instead of you.

What competitor keyword research is really for

The practical purpose is simple. We want to know which searches bring buyers, which rival pages capture those searches, and what we need to change on our own site to compete.

That is different from collecting a long list of keywords with search volume beside them. Volume is not the same as value. A search for a broad topic can bring curiosity clicks. A search for a specific service, product model, city, or delivery area often brings the enquiry.

Used properly, competitor keyword research helps answer questions like these:

  • Which service pages should we improve first?
  • Which locations are we invisible in?
  • Which topics are buyers searching before they contact a supplier?
  • Are rivals winning because they have better pages, stronger links, better local signals, or simply the right page type?
  • Are we missing commercial pages entirely and trying to rank blog posts for buying searches?

For a small business, the right time to do this is usually one of four moments.

First, before rebuilding a website. If you migrate to WordPress, Shopify, Webflow or Ghost without knowing which pages actually matter in search, you can spend months polishing the wrong content.

Second, when expanding into a new market or language. This matters across the EU because search behaviour changes by country and by language. A German buyer may search differently from a buyer in Ireland, even for the same service. If your site is not in English, translated keyword lists are usually a bad guide. We cover that in more detail in our piece on multilingual SEO that brings the right buyers in.

Third, when leads are fine from referrals but weak from search. That usually means your commercial pages are not matching what buyers type into Google.

Fourth, when agencies need a quick, checkable way to set priorities for a client. A proper gap analysis gives a reason for each page recommendation.

Who your real competitors are in Google

Your real competitors in Google are the domains that appear for the searches you need to win. Some will be direct business rivals. Some will not.

That distinction matters because search results are crowded by three other groups.

The first is directories and marketplaces. Think of local trade directories, software comparison sites, property portals, industry listings, and B2B marketplaces. They often rank because they have strong domain authority, many location pages, and structured data at scale.

The second is informational publishers. These are blogs, magazines, trade associations, and “best of” list sites. They can block your commercial pages for top-of-funnel searches such as “how much does X cost” or “best Y for small business”.

The third is Google’s own result features. Local packs, map results, product carousels, People Also Ask, AI Overviews where available, and video results can push traditional organic listings down the page.

So how do we identify real search competitors?

Start with a short list of your money terms. Not 200 phrases. Ten to twenty is enough. Use combinations of:

  • Main service or product
  • Key locations
  • Important qualifiers such as emergency, same-day, certified, wholesale, repair, installation
  • Language variants buyers actually use

Search them manually in an incognito window. Set the right country. If relevant, set the right language too. For local services, search from the target area or use a location setting. Then note:

  • Which domains appear repeatedly
  • Which page types rank, service page, category page, article, directory profile, comparison page
  • Which result features appear before organic listings
  • Whether local pack results dominate

Do not assume a business is a serious SEO rival because you know them in the market. If they do not appear for your priority searches, they are not your search competitor for this exercise.

Also, do not dismiss directories and publishers because you cannot “beat” them across the board. You do not need to. You need to know where they block commercial pages and where a well-built service page can still rank under or alongside them.

For local businesses, search competitors can vary city by city. A chain may dominate in one region and disappear in another. In the UK and Ireland, local pack visibility can differ sharply even between nearby towns. Across the EU, that fragmentation is often stronger because language and regional phrasing vary more.

Which keywords are worth comparing first

Start with intent, not volume.

The first batch to compare should be the searches most likely to lead to revenue in the next few months. That usually means terms tied to a clear offer, a place, and a buyer need.

A practical priority order looks like this:

  1. Core service or product terms
  2. Service plus location terms
  3. High-intent problem terms
  4. Category and comparison terms close to purchase
  5. Informational terms that support commercial pages

Examples help.

If we sell accounting software in Poland, “księgowość online dla małych firm” may matter more than a broader accounting term with larger volume.

If we run a roofing business in Germany, “Dachdecker Berlin Notdienst” is likely worth more than a generic roofing advice query.

If we provide B2B manufacturing in the Baltics, the key terms may differ by language, buyer segment, and export market. One language version can bring research traffic while another brings actual enquiries.

This is why competitor keyword research should begin with buckets, not a single export from an SEO tool. Build groups such as:

  • By service line
  • By product category
  • By location
  • By language
  • By stage of intent

Then score each keyword group against four simple questions:

  • Does this search describe something we actually sell?
  • Would a visitor from this search be close to buying?
  • Can we serve the location shown in the query?
  • Do we already have a page that deserves to rank?

If the answer to the last question is no, that is often your first useful gap.

Do not let search volume overrule business logic. A phrase with modest volume but strong intent can outperform a large informational term for years.

A simple process to find keyword gaps

You do not need an enterprise workflow. You need a repeatable one.

Step 1. Build your shortlist

List your most important services, products, locations and language versions. Turn these into 20 to 50 priority searches. Keep brand terms separate from non-brand terms.

Step 2. Check your own data in Search Console

Open Search Console and look at Queries and Pages for the last 3 to 6 months. Filter to non-brand terms. Look for:

  • Queries with good impressions but weak average position
  • Queries where the wrong page ranks
  • Pages with clicks but poor click-through rate
  • Location or service variants you expected to see but do not

Search Console will not show everything, but it tells you where Google already associates your site with a topic. That is often the fastest route to improvement.

Step 3. Search Google manually

For your top terms, inspect the first page. Note the ranking domains and page types. If the top results are all service pages and you only have a blog article, the gap is not subtle. You need the right page format.

If the results are mixed, note what wins the top three spots. Often one pattern stands out:

  • Local service pages with reviews and clear service areas
  • Category pages with strong filters
  • Articles answering a specific pre-sales question
  • Directory pages with many business listings

Step 4. Use an SEO tool for overlap and gaps

Any solid SEO platform can help here. Export the keywords your domain ranks for and compare them with two to five recurring competitor domains. We are looking for three types of gaps:

  • Missing pages, where rivals rank and you have no relevant page
  • Weak pages, where you rank but far below stronger competitor pages
  • Missing formats, where rivals win with a page type you do not have

Do not compare yourself with ten giant domains at once. Pick the nearest realistic rivals in search results for each keyword group.

Step 5. Map every gap to a URL

This step is where many audits become useless. A keyword gap by itself is not an action. Map each gap to one of these outcomes:

  • Update an existing service or product page
  • Create a new commercial page
  • Create a supporting article
  • Strengthen a location page
  • Improve internal links to an important page
  • Decide not to pursue it

Now the work becomes manageable.

Step 6. Check whether authority is the blocker

If your page is relevant and reasonably good but still stuck, the issue may be authority rather than copy. Look at the links pointing to the ranking pages and domains. We have written before about how to run a link building campaign that holds up, because weak authority often gets mistaken for a content problem.

Step 7. Watch AI answers as well as search

A growing share of discovery now happens in ChatGPT, Gemini, Perplexity, Claude and Grok. The same gaps often show up there in a different form. If your competitors are named in buyer questions and you are not, that is a visibility problem even if your Google rankings look acceptable. We track that directly in Seonis, alongside search performance, because many businesses now need both.

How to judge whether a gap is worth pursuing

Not every gap deserves a new page.

A useful way to judge a gap is to score it on four factors.

1. Business value

Will this search bring the kind of customer we want? A keyword tied to a profitable service in a location we serve scores high. A broad term that attracts students, job seekers or DIY researchers scores low.

2. Realistic difficulty

Look at the actual results, not just a tool score.

Ask:

  • Are the top results mostly businesses like us, or giant publishers?
  • Is the ranking page type something we can produce properly?
  • Do the top pages have strong local signals, reviews, links or brand searches that we lack?
  • Is the result page dominated by maps, directories or product features?

If the top ten are all national comparison sites and government resources, that is a warning. If several independent businesses rank with ordinary service pages, the gap may be realistic.

3. Local relevance

For local and regional businesses, relevance is not optional. A perfect page for “warehouse cleaning” may still fail if the search is really “warehouse cleaning Leeds” and your page never mentions Leeds, local case studies, or service coverage.

This is where country differences matter. In some EU markets, users often search in the local language plus an English product term. In others, they use strongly local phrasing. We should not assume the UK pattern applies elsewhere.

4. Effort required

Some gaps need a title tag and a better internal link. Others need a new landing page, fresh photography, testimonials, schema, and links. Be honest about the effort.

A good rule is this. Prioritise gaps where:

  • The search has clear buying intent
  • You already have some relevance or impressions
  • The required page is close to what you already offer
  • The authority gap is not extreme

For agencies, this is the point where a recommendation becomes commercially useful. Instead of saying “create more content”, we can say “improve three service pages, add two city pages, and publish one pricing explainer because the current gaps are close to revenue and achievable”.

What to do after you find the gaps

Turn the findings into a short action plan. Not a 60-row backlog no one touches.

We usually group actions into four buckets.

1. Fix weak money pages first

If an existing service or product page is almost there, improve that before writing new articles.

Typical fixes include:

  • Rewrite the title and H1 to match the target intent
  • Add missing service details, delivery areas, pricing approach or FAQs
  • Include trust signals such as certifications, case studies or review snippets
  • Clarify the call to action
  • Add internal links from relevant articles and category pages

This work often beats creating ten new blog posts.

2. Create missing commercial pages

If competitors rank with dedicated pages for a service, product type or location that you do not have, build the page. Keep it specific. One strong page for a real offer is better than a thin page for every town in the country.

For local businesses, avoid near-duplicate location pages. If a page exists for a city, it should contain something useful about that service in that place.

3. Add supporting content where it helps a commercial page

Not every gap should be answered with a sales page. Some searches need an article, guide or comparison piece that supports a later conversion.

Good supporting content often covers:

  • Cost questions
  • Timelines
  • Process questions
  • Comparisons
  • Suitability for different business types

The point is not traffic for its own sake. The point is to help the main commercial pages rank and convert.

4. Improve internal linking and authority

Once pages exist, connect them properly. Link from articles to service pages with natural anchor text. Link between related services. Make sure important pages are not buried.

If authority is still the issue, plan link acquisition carefully. For agencies handling local-language sites, this is one reason some use our partner programme for agencies, especially when they need native-language publishing and a repeatable authority workflow without building a large in-house SEO team.

5. Track outcomes in search and AI answers

After changes go live, monitor:

  • Impressions and clicks in Search Console
  • Ranking movement for the target pages
  • Enquiries or sales from those pages
  • Whether AI systems begin to cite or name your business more often

This last point is easy to ignore and increasingly costly. A business can improve rankings yet still be absent from AI answers that summarise the market for buyers. We track whether ChatGPT, Gemini, Perplexity, Claude and Grok mention the business for commercial prompts, because that is now part of visibility, not a side issue.

If you want this process to run consistently, Seonis was built for exactly that kind of work on non-English websites. We research keywords in the site’s own language, publish natively rather than translating, check each piece through a language gate, and report in the owner’s language. Our plans, backlink credits and discounts set out what is included.

The main thing is not the tool. It is the discipline. Competitor keyword research works when it leads to page decisions, not when it ends as a ranking report. Find the rivals taking your best searches. Work out which pages deserve to exist. Improve the pages closest to revenue first. Then measure whether search engines and AI systems start choosing you more often.