Keyword research for local businesses.

Keyword research is the part of SEO everyone nods along to and almost nobody actually does. It gets skipped because it sounds academic, like homework you do before the real work starts. In practice it is the opposite. It is the decision that determines which pages you build, what you write on them, how you describe your services in your Google Business Profile, and ultimately whether the phone rings.

Get it right and everything downstream gets easier. Get it wrong and you can build a beautiful website, publish content every month, and still watch competitors take the calls — because they are answering the searches your customers actually type, and you are answering the ones you assumed they type.

For a local business the stakes are higher than for a national brand, and the odds are better. Higher, because you are not chasing a wide audience; you are chasing the few hundred people within driving distance who need what you do this month. Better, because local keyword competition is dramatically softer. You are not trying to outrank Amazon. You are trying to outrank four or five businesses in your own market, most of whom have never done this work at all.

What the search data actually says

Before the how, the scale. A handful of numbers get repeated constantly in this corner of marketing, so here they are with their actual provenance — which in one case is not what you would expect.

The famous one: 46% of Google searches have local intent. That comes from a remark by a Google representative at an industry event in 2018. Google has never published the data behind it or defined what counts as local intent, and nothing more recent has replaced it. It gets quoted as gospel. Treat it as a directional statement from Google rather than a measurement.

Better documented, if older: in a 2016 diary study of 1,000 US smartphone users, Google found that 76% of people who searched on a phone for something nearby visited a related business within a day, and 28% of those searches ended in a purchase. Someone searching “plumber near me” at eleven at night is not running a research project. They have a burst pipe.

The payoff for showing up is measurable. SOCi’s 2022 analysis of more than 300,000 Google listings found that listings in the three-pack earned 126% more search traffic, and 93% more conversion actions like calls and direction requests, than listings ranked four through ten.

And the reason to be optimistic rather than anxious: most of your competitors are not doing this. In a ReviewTrackers survey of more than 500 companies published in 2021, 58% said they did not optimize for local search at all, another 31% said they would like to, and only 30% had a plan in place. Those figures are a few years old. The gap they describe is still the first thing we see in most audits.

One more, by way of demonstration. You will see confident figures for “near me” search volume — 200 million a month globally, 7.1 million in the US. Both trace back to a single blog post that cites two vendor homepages rather than any published research, and they contradict each other by a factor of nearly thirty. That is the state of most statistics in this field, which is a good reason to check the date and the source on anything you are about to build a strategy around.

Why search volume is the wrong metric

The most common keyword research mistake is chasing the biggest number in the spreadsheet. It feels logical — more searches should mean more customers — and for a local business it is almost always the wrong move.

Two numbers get quoted constantly here, and both are usually quoted without the part that matters.

Backlinko’s analysis of 306 million US keywords found 91.8% of them get between 1 and 100 searches a month. Ahrefs reports that about 93% of the keywords in its US database — roughly 2.3 billion of them — get fewer than ten searches a month. Both are true. Both describe how many distinct phrases exist, not where the searching happens. In Backlinko’s own data those millions of low-volume terms account for just 3.3% of total search volume.

So the honest version is not “there is hidden traffic in the long tail.” Ahrefs tested that directly and found the opposite: terms their tool scored at ten searches or fewer delivered about eleven impressions a month, roughly what the tool predicted, with fewer than 1% producing more than a hundred. Low volume usually means low volume.

The case for specific phrases in a local market is different, and it is stronger:

• Your whole market is a small number. You do not need thousands of searches. You need the few dozen people within driving distance who need this now, and a page that is unmistakably the right answer for them.
• Intent is concentrated in the specific phrases. “Plumber” is muddy — directories, aggregators, someone writing a school report. “Emergency water heater replacement in Hammond LA” is a person with a wallet out.
• Competition thins out fast. Nobody has written the better page for the specific version of the question, because the volume looks too small to bother with.
• One page collects many variations. In a 2017 study of three million searches, Ahrefs found the average page ranking #1 also ranked in the top ten for nearly a thousand other keywords, with a median closer to 400. That study is old enough to treat as direction rather than arithmetic, but the pattern holds: build the page around a topic and it picks up phrasings you never typed.

How to do the research

Here is a process that works for a single-location business and needs no expensive software to start.

1. List what you actually sell, in customer language. Not internal terminology — the words a customer would use. A vet clinic offers “wellness exams,” but customers search “puppy shots near me.” An HVAC company performs “condenser coil replacement,” but customers search “AC not blowing cold air.” Write down every service, then translate each one into the phrasing a non-expert would use.
2. Mine Google itself. The most underused free research tool there is. Start typing your service into Google and read every autocomplete suggestion — those are real queries. Scroll to the People Also Ask box and note every question. Then read the related searches at the bottom. Do that for each core service and you will have more genuine keyword data than most paid tools hand you.
3. Check your own data. Your Google Business Profile shows the search terms people used to find you. Google Search Console shows every query that produced an impression for your site, including the ones you rank on page three for. That second report is the gold: it shows terms you are already close on, where a little focused work moves you into range.
4. Study whoever is ranking above you. Search your main service plus your city and look at the top three organic results and the three Map Pack listings. What pages do they have? What services are listed on their profiles? What questions do they answer? You are not copying them, you are finding the gaps you can fill better.
5. Sort by intent, not volume. Put every keyword in one of three buckets. Buying intent means ready now — “emergency,” “near me,” “open now,” “cost,” “quote,” “best.” Research intent means comparing options and buying later. Informational intent means they want to know something and may never buy. Build for buying intent first, research second, informational last.

Mapping keywords to pages, where most local sites fall apart

Research is only half the job. The other half is deciding where each keyword lives, and this is where most local websites lose.

The typical small business site has four pages: Home, About, Services, Contact. The Services page is one long list with a sentence about each offering. That structure asks a single page to rank for eight different services, and it will rank well for none of them.

The fix is one page per meaningful service. If you offer drain cleaning, water heater installation, repiping and slab leak repair, that is four pages, each targeting its own cluster and each thorough enough to genuinely answer what that customer needs to know. Specifics matter: real process detail, real considerations, real pricing factors. A page that says “we offer expert drain cleaning” ranks nowhere. A page explaining what causes recurring clogs, what hydro jetting actually involves and when a camera inspection is worth paying for can rank, because it is the better answer.

Then geography. If you serve several towns, each significant one can justify its own page — but only if the page is genuinely about that place. A city page that is a find-and-replace of another with the name swapped is a doorway page, and Google has penalized those for years. Real local pages reference actual neighborhoods, local conditions, work you have done in that area, and the specific reasons the job differs there. (Rankings change with where the searcher is standing, which is why this matters more than it looks.)

Once the structure is right, put your keywords where they carry weight: the page title, the H1, the first paragraph, at least one subheading, the image alt text, the meta description, and naturally through the body. Naturally is the operative word. Keyword stuffing has been ineffective for a decade and it looks unprofessional to the humans you are trying to convert.

Feed your Google Business Profile the same research

For a local business, your Google Business Profile is arguably more important than your website, and your keyword research should feed it directly.

Start with categories. Your primary category is one of the strongest local ranking signals you control, and most businesses pick one and stop. Use the most specific one available, then add every genuinely relevant secondary category — each one widens the range of searches you are eligible for. Then build out the services section with the research in hand. Most profiles leave it empty, which is the equivalent of refusing to tell Google what you sell.

Your business description has room for keyword-informed copy too. Localo’s analysis of more than two million Google Business Profiles found profiles in the top three average just under 70 words of description, more than two hundred reviews and upwards of 250 images. Those are correlations with visibility rather than proven ranking factors — established businesses have more of everything — but as a picture of what a competitive profile looks like, it is a useful benchmark.

Reviews deserve a mention in any keyword conversation, because customers describe your services in their own reviews using the exact language other customers search with. A review that says “fixed our AC on the hottest day of July” is doing keyword work you could never write yourself, and it carries credibility no marketing copy can match. BrightLocal’s 2026 survey found 97% of consumers read reviews for local businesses.

• Use the most specific primary category available, plus every secondary category that genuinely applies.
• List every service individually, each with a real description in the language customers search with.
• Write a substantive business description rather than a single line.
• Post consistently. It signals an active profile, and most competitors have never posted once.
• Build a steady flow of reviews — customers describe your work in the exact words other customers search for.

What changes now that AI answers come first

One real change to how this work should be done: a growing share of local searches now get answered before anyone clicks. SparkToro’s June 2026 analysis of Similarweb clickstream data put the US zero-click rate at 68%, and BrightLocal found 45% of consumers have used AI tools to find a local business, up from 6% a year earlier.

It is worth being precise about where that hits you, because the headline number gets misused. Whitespark’s 2025 study of 540 local queries found AI Overviews on 68% of them against 39% for the local pack — but split by intent, the picture inverts. For plain local-intent searches, the kind where someone wants a business to call, the local pack appeared 93% of the time and an AI Overview only 15%. For informational queries it flipped, with AI Overviews on 92%.

So the Map Pack still owns “plumber near me.” What AI has taken is the question that comes before it — and that question is where your keyword research now has to reach. BrightEdge reported that 25% of queries of eight words or more triggered an AI Overview as of January 2025, though it published no sample size or method, so treat it as a vendor observation rather than a measurement.

Practically, capture full questions rather than service phrases. Not “AC repair Hammond” but “why is my AC blowing warm air” and “how much does it cost to replace an AC unit in Louisiana.” Put those questions in as headings and answer each one directly in the first sentence underneath. That format is what both featured snippets and AI answers pull from. And write with specifics: real numbers, real process, real detail. Vague marketing language gives these systems nothing to extract.

Five mistakes that cost local businesses calls

These come up constantly in the audits we run, and every one is fixable.

1. Chasing volume over intent. The most expensive mistake on the list. Ten visitors searching a buying phrase beat a thousand searching an informational one.
2. Ignoring your own data. Search Console is free and shows exactly which queries already produce impressions for your site. Plenty of businesses pay for keyword tools while never opening the report showing the terms they are already close on.
3. Using jargon customers never type. If you describe your service the way you would to a colleague rather than to a customer, you are optimizing for a conversation nobody is having in the search bar.
4. Cramming everything onto one services page. One page cannot rank for eight services. Give each meaningful service its own page, with enough depth to be the best answer available.
5. Treating it as a one-time project. Markets shift, competitors appear, seasons change. A quarterly review keeps the keyword set current without it becoming a job.

How to tell whether it is working

Traffic is a poor primary metric for a local business. You can double your traffic with informational content and not get a single extra call. Measure the things that connect to revenue instead.

Track your Map Pack visibility across your whole service area rather than a single ranking number. A local business does not have one ranking — it ranks differently depending on where the searcher is standing, and grid-based tracking shows you where you appear and where you do not. Then watch calls and direction requests from your Google Business Profile, calls and form submissions from the website, and impressions and clicks by individual query in Search Console. That last report shows movement on specific terms over time, which is how you tell whether the work is landing or whether a page needs rewriting.

Give it time. Local SEO compounds rather than spikes. Foundation work lands in the first month or two, movement on lower-competition terms usually shows in the 60 to 90 day window, and meaningful traction on primary keywords typically arrives around months five and six. Anyone promising faster than that is overselling.

Frequently asked questions

Why is keyword research important for local businesses?

Because the phrases people actually type decide which pages you need. In a 2016 Google diary study, 76% of people who searched on a phone for something nearby visited a related business within a day, and SOCi’s 2022 analysis of more than 300,000 listings found three-pack listings earned 126% more search traffic than listings ranked four through ten. Keyword research identifies the exact phrases those buyers use, so your pages and your Google Business Profile get built around the searches that produce calls rather than the ones that only produce traffic.

How often should I update my keywords?

Review the set quarterly, and revisit it any time you add a service, expand your service area, or notice a drop in calls. Quarterly catches seasonal shifts and new competitors without turning keyword research into a full-time job.

Should I target high-volume or long-tail keywords?

For local businesses, specific wins — though not for the reason usually given. Backlinko found 91.8% of keywords get 1 to 100 searches a month, and Ahrefs puts about 93% of its US database under ten searches a month, but those terms carry a small share of total search volume and Ahrefs’ own test found no hidden traffic in them. The reason to target them locally is intent and competition: your entire market is a small number of people anyway, and nobody has bothered to write the better page for the specific question.

Do I need a paid keyword research tool?

Not to start. Google autocomplete, People Also Ask, the related searches at the bottom of the page, your Google Business Profile search terms report and Google Search Console will surface most of what a single-location business needs. Paid tools add competitive data and volume estimates, which matter more once you are fighting for a crowded service term.

How many keywords should one page target?

One topic per page, not one keyword. Ahrefs’ 2017 study of three million searches found the average page ranking #1 also ranked in the top ten for nearly a thousand other keywords, median around 400. Build the page around a clear topic and intent and it will collect variations you never explicitly targeted.

Does keyword research still matter with AI Overviews and ChatGPT?

It matters more, but the target has shifted toward questions. Whitespark’s 2025 research found AI Overviews on 68% of local business queries against 39% for the local pack — though for plain local-intent searches the local pack still appeared 93% of the time and an AI Overview only 15%. The shift is in the research question that comes before the call, so map the full questions customers ask rather than two-word service phrases.

How do I know if my keyword research is working?

Track Map Pack visibility across your service area rather than a single ranking number, plus calls and form submissions from your Google Business Profile, and impressions and clicks by query in Search Console. Traffic alone is a poor measure for a local business, because ten visitors searching a buying phrase are worth more than a thousand searching something informational.

Where to start

Keyword research is the foundation everything else sits on: your pages, your Google Business Profile, your content, and your visibility in both the Map Pack and AI answers. Do it well and every other piece of marketing you buy works harder. Skip it and you are guessing.

If you would rather not do it yourself, it is part of the Growth plan, and our prices are published rather than quoted on a call.