How to Do Keyword Research for a Local Business
This is a genuinely different exercise rather than a variation on the standard method. The volumes involved are small enough that the tools stop being reliable exactly where a local business needs them most, so you have to work another way.
A Different Exercise Entirely
The standard method sorts a list by numbers and filters it down. Local research cannot work that way, because the numbers attached to local searches are the least reliable figures any tool produces. What replaces them is commercial logic, what customers actually ask you and eventually your own search reporting.
What stays the same. The two filters.
Winnable and worth winning still decide everything. The evidence you use to apply them is what changes.
What changes completely. The evidence.
No dependable volume, no meaningful difficulty score and a very short list of genuinely relevant searches. Working without those is the skill this page teaches.
The advantage nobody mentions. Smaller field.
A local business is competing against a handful of firms rather than the entire internet. That is a contest you can actually win, which is why specificity matters so much.
What most local sites get wrong. Two things.
Building a page for every town within reach. Then ignoring the searches that never mention a place. Blocks six and eight deal with both.
Why it suits a small business. The work is finite.
A local firm has a handful of services and a defined area. That produces a short list of subjects rather than an endless one, which means the exercise can actually be completed rather than merely started.
What sits outside this page. Your profile.
Map results are decided largely by your business profile rather than by your website. Our Google Business Profile guide covers that side properly.
The Tools Stop Working Here
Keyword tools are least reliable precisely where local businesses need them most. Estimating a large national search from sampled behaviour is achievable. Estimating a specific service in one town has almost nothing to work from, so the figures produced are close to meaningless.
Why small numbers break estimation. No pattern.
A model needs enough activity to infer from. Below a certain scale there is nothing to infer, so the output is a placeholder rather than a finding.
What that looks like on screen. Empty columns.
Searches shown as having no volume. Or grouped into a lowest band alongside completely different searches. Both hide whether anybody is looking.
The decision that follows. A page not built.
Somebody sees the figure, concludes there is no demand and skips it. That is the actual cost of trusting the column. We see it constantly.
What we know instead. The phone rings.
A business receiving enquiries for a service in a town has proof of demand. That evidence is better than any estimate and it is already in the building.
Why the low figure can still be worth it. Concentration.
Few searchers, almost all of them wanting the job done nearby. That converts at a rate no broad national term approaches.
The rule we apply. Never delete on volume.
A search with an empty column and obvious commercial sense stays on the list. Our guide to tools covers why those numbers are estimates in the first place.
Service Plus Place Is Only The Start
Everybody starts with the service and the town joined together. That pattern is real and people do search that way. It is also the smallest part of local demand. Treating it as the whole exercise is why so many local sites consist of a dozen town pages and nothing else.
Why it is the obvious one. It is visible.
It appears in every tool, it reads as commercially obvious and it maps neatly onto a page each. Convenient, which is not the same as complete.
What it captures. Deliberate area searchers.
People specifically checking whether somebody covers where they live. Genuine and valuable, especially at the edges of a service area.
What it misses. Everybody else.
The larger group who search without naming anywhere, because the search engine already knows where they are. Block six covers those.
The variations worth having. Beyond the town.
Neighbourhoods, nearby villages, the informal name locals use for an area and occasionally a postcode district. All of them get typed and few of them appear in tools.
The pattern people forget. Place before service.
Plenty of people type the town first and the service second. It reads oddly written down and it is genuinely how a good many searches arrive, so a page built rigidly around one word order misses the rest.
The qualifier that changes everything. Urgency.
Emergency, today, out of hours or weekend attached to a service and a place is the highest intent search a trade receives. Those deserve their own attention.
Where the pages themselves get built. Carefully.
Block eight covers it. our landing pages guide covers where the line sits.
Which Places Are Actually Realistic
A business cannot appear prominently for a town it has no connection to. Distance from where you are based matters. It matters more for some kinds of result than others, so the list of places worth pursuing is usually shorter than the list a business would like.
What we will not give you. A radius.
No mileage, no travel time and no rule about how far is too far. It depends on the trade, the density of competitors and the value of the job, so any figure would be invented.
What decides it instead. Three things.
Whether you would actually travel there for the work. Whether the job value justifies the journey. And whether anybody there has heard of you.
The evidence you already have. Past jobs.
Where have your customers actually been, over several years. That map is the truthful answer to which places are realistic. It is sitting in your invoices.
Why the ambition fails. No connection.
Pursuing a city an hour away where you have never worked means competing against firms based there, with nothing to show for the place. The effort produces nothing.
The order to work in. Outward.
Your own town first, then the places you genuinely work in most, then the edges. Building the far ones first is how businesses waste a year.
What to do about the edges. Be specific.
A page about a place you occasionally reach should say what you do there and be plain about it. That is more convincing than pretending to be local.
Near Me, And What It Actually Means
You do not target the phrase near me. Businesses put it in their copy, which does nothing, because what decides those results is where the searcher is standing rather than which words appear on a page.
What the phrase means to a search engine. Use my location.
It is an instruction rather than a subject. The searcher is saying to use where they are, so the answer is calculated from position rather than from text.
Why copy does not help. Nothing to match.
Writing electrician near me on your page matches nobody, since the phrase resolves to a place before anything is compared. It also reads badly to the person who arrives.
What actually decides those results. Mostly your profile.
Proximity to the searcher, your business profile and your general standing locally. Our Google Business Profile material covers that work.
What the website contributes. Consistency.
Clear, consistent information about where you are and what you do. That supports the profile rather than replacing it.
Why it still matters to research. Volume of use.
An enormous amount of local searching happens this way, so understanding that it sits outside your copy changes where you spend effort.
The related family. Nearby and open now.
Same mechanism, same conclusion. All of them are location instructions rather than phrases to write into a page.
The Searches That Never Name A Place
Somebody standing in their kitchen types boiler repair. No town, no near me, nothing local in the words at all. They still get local results, because the search engine already knows where they are. This is the largest part of local demand and it is invisible in most keyword research.
Why they get treated as local. The need is local.
A search for a service somebody needs performed at their property is understood as local regardless of phrasing. Nobody wants a boiler engineer two hundred miles away.
Why businesses miss them. They look national.
In a keyword list, boiler repair looks like an enormous national term nobody local could win. In practice each searcher is being shown a local set of results.
What that changes. Your service pages matter more.
Your main service pages are competing locally whether or not they mention a town. That makes them more valuable than the town pages everybody obsesses over.
The practical consequence. Build the service page properly.
A thorough page about the service itself, written for somebody with that problem, does more work than five thin location pages.
How to check this yourself. Your own reporting.
Look at what your site actually appears for. Searches with no place name in them will be there. That is the demonstration.
What this does to the plan. Reorders it.
Services first, then places. Most local sites do it the other way round and wonder why the town pages do nothing.
What Customers Actually Call It
Local language is genuinely local. What a job is called varies by region, by age and by whether somebody has dealt with the problem before. No tool captures that reliably. The business already has the answer.
Where the variation appears. Everywhere.
The name of the job, the name of the part, the name of the area and the informal name for a district that no map uses. All of them get typed.
The best source. Your own phone.
Write down how callers describe the problem, in their words, for a fortnight. That list is better local research than any subscription produces.
What to watch for. Regional names.
Trades have terms that shift across the country. Using the one your customers use, rather than the one the trade body prefers, is what gets you found.
The area name problem. Boundaries disagree.
People describe where they live by a name that may not match the official designation. Both get searched and only one appears on a map.
How to check a variation. The search box.
Type it and see what the suggestions offer. Those are drawn from what people actually search, which settles the question in seconds.
The generational split. Worth noticing.
Older customers frequently use the term that was standard when they first encountered the job. Younger ones use whatever a search engine suggested to them. Both are your customers and both phrasings deserve to be on the page somewhere.
What to do with the findings. Use their word.
Lead with the customer term and introduce the correct one afterwards, which is the same rule as everywhere else in this series.
Turning It Into Pages Without Doorways
The temptation at the end of local research is a page for every place on the list. Resist it. A location page needs content only somebody who works there could write. If you cannot produce that, you do not have a page.
What a real location page contains. Specifics.
Jobs you have done there, the property types and what they mean for the work, access or parking conditions, how quickly you can reach it and anything locally particular about the job.
What a doorway page contains. A swapped word.
The same paragraphs with the town changed. Anybody reading two of them together sees it immediately. It is treated as a problem rather than a technique.
The test before building. Write three sentences.
Three things about working in that place that would not be true of anywhere else. If they will not come, that page should not exist yet.
What to do instead for thin places. One page, many names.
A single areas covered page listing everywhere you serve is legitimate and useful. It is also considerably better than a dozen empty pages.
The order of work. Services, then places.
Because of block six. The service pages compete locally already, so they earn their keep before any town page does.
What to build first among places. Where you already work.
You have material for those. The places you wish you worked are the hardest pages to write and the least likely to succeed.
What to do with the areas page. Keep it current.
An areas covered page listing everywhere you serve should match reality rather than ambition. A list including places you would refuse to travel to costs you nothing in search terms and costs you a wasted phone call every so often.
Where the rest sits. The hub.
All twenty-two guides are on the keyword research guide.
A tool showing
no volume is not
evidence that
nobody searched.
Small numbers are the hardest thing to estimate, so the searches that actually produce work for a local firm are described worst of all. Somebody looks at an empty column, concludes there is no demand and does not build the page. Meanwhile the phone rings every week for exactly that job in exactly that town.
How we research a local business:
Before building a location page, write three things about working there that would not be true anywhere else. If they will not come, the page should not exist yet.
Every guide.
One subject.
What the words mean, what the numbers in a keyword tool actually are, how to find the searches worth targeting, how to turn a list into a publishing plan and what to do when a page you relied on slips.
