Most keyword advice tells a small business to hunt for the biggest search numbers. That instinct is exactly how a small site ends up invisible: you pour months into a term you'll never outrank, while the queries you could have won sit untouched. High volume isn't opportunity if a wall of national brands stands between you and it.
Here's the takeaway up front: for a small business, keyword research is a filtering job, not a discovery job. You're after the short set of queries where two things are true — the searcher wants what you sell, and the results are weak enough that a site your size can realistically appear. Volume is the last thing you check, not the first. This guide is a repeatable, free-tool process for finding those keywords.
Start with intent, not search volume
Every search has an intent behind it, and matching it is the whole game. There are four broad types:
- Informational — learning something ("how does drain cleaning work").
- Navigational — looking for a specific brand or site.
- Commercial — comparing before buying ("best accountant for freelancers").
- Transactional — ready to act now ("emergency plumber near me").
A big brand can chase all four. A small business with a handful of pages should weight heavily toward commercial and transactional queries — the people closest to spending — plus the questions buyers ask right before they choose.
The fastest way to read intent is to see what already ranks. Search the term: if page one is service and product pages, it's a buying query; if it's guides and blog posts, it's informational. Match your page type to what Google already rewards — a blog post won't rank where every winner is a service page.
Build a seed list from your customers' own words
Don't open a tool yet. Start with language — the words your customers use, which are rarely your internal jargon. Pull your seed list from real sources:
- Your services and products, phrased the way a customer would say them out loud.
- Sales calls, emails, and DMs — the questions people ask before they commit. Every objection is a keyword.
- Reviews, yours and competitors'. How people describe the problem and the result is gold-standard phrasing.
- Google autocomplete and "People also ask." Start typing a core service and note every suggestion — that's Google reporting real phrasings for free.
- Google Search Console, if your site exists. Its Performance report lists queries you already earn impressions for — your highest-signal terms, often sitting on page two, one good page away from page one.
Aim for 30 to 50 seed phrases, grouped by the service each maps to.
Expand and check demand with free tools
You don't need a paid subscription for serious keyword research — paid tools save time at scale but don't change the method. Use these, noting why each earns its place:
- Google autocomplete, "Related searches," and "People also ask" — free, and it's Google surfacing real variations. Best for long-tail phrasings you'd never guess.
- Google Search Console — free, and it's your own data, the highest-signal source there is. Use it to spot near-miss queries a small push could lift onto page one.
- Google Keyword Planner (free with a Google Ads account) — volume ranges and related terms. Ranges are broad, but broad is enough to separate "some demand" from "none."
- Google Trends — free; use it for direction (rising, seasonal, steady) and to compare two candidates, not for absolute numbers.
- The free tier of an SEO tool — a few daily lookups for a rough difficulty score to sort by. Handy, not required.
Keep the low-volume, specific terms too — for a small business, those are often the point.
Judge whether you can actually rank (the winnability filter)
This is the step most guides skip, and the one that decides whether your effort pays off. Before committing to a keyword, assess its difficulty by reading the results yourself: open an incognito window, search the term, and study page one.
Winnable signals:
- Forums, Reddit, or Quora are ranking — no strong dedicated page owns the query.
- Top pages are thin, generic, or years out of date, and don't quite answer the question.
- Other small or local businesses appear, not just national names.
Hard signals:
- Page one is wall-to-wall major brands, big publishers, or large comparison sites.
- The ranking pages are deep, current, and thorough.
- A featured snippet or knowledge panel is locked up by an authority.
Target queries where at least one page-one result is beatable with a more specific, more current page. A small site doesn't win on authority — it wins on specificity and freshness. That's why long-tail terms are easier: fewer strong pages compete for an exact query.
Favor long-tail and local modifiers
Long-tail keywords are longer, more specific searches — usually four or more words, like "gluten free birthday cake delivery [your city]." Each has lower volume, but they convert better, rank far easier, and add up to steady, qualified traffic.
Watch winnability and intent climb as a term gets more specific:
- "cake" — huge volume, no clear intent, unwinnable.
- "birthday cake" — still broad, still fiercely competitive.
- "custom birthday cakes [your city]" — winnable, local, ready to buy.
- "gluten free birthday cake delivery [your city]" — very winnable, very high intent, and yours because almost no one has written the perfect page for it.
For any business with a service area, local modifiers — a city, neighborhood, or "near me" — are your highest-return terms, because you compete against nearby businesses, not the whole internet. Ten specific local terms, each bringing a few ready-to-buy visitors, out-earn one head term you never crack.
Map keywords to pages — one primary per page
A keyword list isn't a plan until it's mapped to pages. Group your filtered terms into clusters that share intent, and turn each cluster into one page:
- Give each page one primary keyword plus a few close variants — don't build two pages chasing the same term, or they'll compete (keyword cannibalization) and both lose.
- Match page type to intent: transactional and commercial terms go on service, product, or landing pages; informational questions become blog posts that link to those money pages.
- Sequence by winnability — publish the most winnable, highest-intent pages first so you see results while harder terms build.
Choosing keywords is only half the job; the pages then have to get found. If you've built pages and they still aren't appearing, work through why your business isn't showing up on Google before assuming your keywords were wrong.
Your keyword research checklist
Run every keyword through this before it earns a page:
- Seed it from a real customer source, not a guess.
- Confirm intent from the page types already ranking.
- Check demand with a free tool — enough to justify a page.
- Read the results and judge if a page-one spot is beatable.
- Prefer the long-tail, local version when a head term is out of reach.
- Assign one primary keyword per page, matched to intent.
- Prioritize winnable, high-intent terms first.
Clear every step and the term earns a page. Fail at intent or winnability and drop it — however tempting the volume.
FAQ
What is keyword research and why does it matter for a small business?
Keyword research is finding the exact phrases people type when they want what you offer, then choosing which to build pages around. It matters because your time and pages are limited: the right terms bring in people ready to buy instead of traffic that never converts.
How do I do keyword research for free?
Start with your customers' words from calls and reviews, expand them with Google autocomplete and "People also ask," check rough demand in Keyword Planner and Google Trends, and mine Google Search Console for queries you already rank for. That free stack covers the whole method.
What are long-tail keywords, and why do they matter more for small businesses?
Long-tail keywords are longer, more specific searches (often four-plus words) with lower volume but higher intent and far less competition. They matter more for small businesses because you can realistically rank for them and the searchers are closer to buying.
How do I know if I can rank for a keyword?
Search it in an incognito window and read page one. If you see forums, thin or dated pages, or other small businesses, it's winnable with a better, more specific page. If page one is all major brands and deep, current content, treat it as hard and target a longer, more local variant.
How many keywords should I target on one page?
One primary keyword per page, plus a handful of close variants that share the same intent. Building two pages around the same term makes them compete with each other, so cluster related keywords onto one strong page instead of splitting them.
How is keyword research different for local businesses?
Lead with location modifiers — a city, neighborhood, or "near me" — because you then compete only against nearby rivals, which makes those terms both winnable and high-intent. Pair the modifier with the specific service ("emergency electrician [your city]") to catch searchers ready to call.
Next step
Keyword research for a small business is about discipline, not budget. Build a seed list from how customers describe their problem, filter hard by intent and winnability, favor the specific and local over the broad, and give each keyword its own page. Then every page you publish is a bet you can win — not a guess.
Want help turning a keyword list into a page plan that ranks? Talk to the team at Build Mind — we help small businesses get found by the people already searching for what they sell.
