WordPress,
Shopify,
Wix, Webflow and Ghost. Any other site, hosted on your own domain →Fifty keywords in a spreadsheet is where small business SEO usually stalls. Finding them is the easy half, and we have covered it in detail in how to find keywords your customers actually search for. The harder half is choosing, because a business that can publish one post a week gets twelve articles a quarter, and every keyword you pick means another you leave on the shelf.
This article covers how to choose keywords for small business sites in that exact situation: a long list, a short publishing capacity, and no budget for paid research tools. The method takes an hour and needs nothing more than a browser.
How to choose keywords for a small business: intent beats volume
The single biggest mistake is sorting the list by search volume and starting at the top. Volume tells you how many people type a phrase. It tells you nothing about whether any of them will ever pay you.
A keyword searched thirty times a month by people who need your service this week is worth more than one searched three thousand times a month by students, hobbyists and people in other countries. "emergency boiler repair leeds" might bring you two paying jobs a month. "how does a boiler work" might bring you two thousand visitors and nothing else.
Rule of thumb: a 30-searches-a-month keyword with buying intent beats a 3,000-searches-a-month informational one. Score intent first, volume second, always.
A simple scoring method: intent times reachability
Give every keyword on your list two scores.
Intent, scored 1 to 3. Ask what the searcher does next if your page answers them perfectly.
- 3: they contact a business or buy something. Phrases with "cost", "price", "near me", "hire", "repair", a town name, or a product name usually land here.
- 2: they are comparing options or diagnosing a problem you solve. "xero vs quickbooks for landlords", "why does my server keep dropping connections".
- 1: they want an answer and will leave. "what is a p60", "how do heat pumps work".
Reachability, scored 1 to 3. Can a site like yours realistically reach page one? That is a judgement about difficulty, and you can make it for free, which is the next section.
Multiply the two scores. Anything scoring 6 or 9 goes on the shortlist. Anything scoring 1 or 2 gets cut, whatever its volume. The 3s and 4s are your reserve list for later quarters. Intent-1 keywords are not worthless, they can support a pillar and cluster plan once the money keywords are covered, but they should never be where a small site starts.
Free difficulty checks: just look at page one
Paid tools sell you a difficulty score out of a hundred. You can get a more honest answer by searching the keyword yourself, in a private browsing window so your history does not skew the results, and reading page one like a competitor.
- Check who ranks. If page one is national brands, government sites and Wikipedia, score reachability 1 and move on. If it is other small businesses like yours, score it 3. A mix scores 2.
- Check the quality. Forums, Reddit threads, thin directory listings and pages that only half-answer the question are a green light. Google is ranking them because nothing better exists. Write the something better.
- Check the match. If the ranking pages answer a slightly different question, there is a gap. Someone searching "payroll for one employee" who finds only general payroll guides is being underserved.
- Check for any small site at all. One independent business on page one proves the door is open. None at all, for a commercial term, usually means the term is harder than it looks.
Ten minutes of this across your top candidates beats any difficulty metric, because you are seeing exactly what Google has chosen to show, not a model's guess about it.
Local modifier or national term?
If you serve a geographic area, the local version of a keyword nearly always wins on both scores. "it support bristol" has a fraction of the volume of "it support", but the searcher is closer to buying and page one is other regional firms rather than global brands. Pick the local version whenever you actually serve that area.
Pick the national term only when you genuinely sell nationally, an online shop or a remote service, or when the local version has so little search that even ranking first would bring almost nobody. Even then, the long-tail national phrase usually beats the head term: "accountant for amazon sellers" is winnable, "accountant" is not. Trades businesses weighing this up may also want our guide to local SEO for tradesmen.
Which free tool serves which step
- Google Keyword Planner (free with a Google Ads account, no spend required): rough volume tiers for your list. The ranges are broad, but you only need to know whether a phrase is searched tens, hundreds or thousands of times a month.
- AnswerThePublic: expanding a seed term into question phrasings during the finding stage, and checking you have not missed an obvious high-intent variant before you cut.
- Google Trends: checking whether a keyword is seasonal or declining before you commit a quarter to it. A term that spikes every January should be published in November, not June.
- Keywords Everywhere (a low-cost browser extension): volume and related terms shown directly on the results page while you do the page-one difficulty check, which saves switching between tabs.
The worked shortlist: 50 keywords cut to a quarter's plan of 12
Here is the method applied to a fictional IT support firm in Bristol that started with 50 candidates. After scoring, 31 were cut for intent of 1 or a page one full of national brands, and 7 went to the reserve list. These 12 made the first quarter, ordered highest score first.
| Keyword | Intent | Page one today | Why it made the cut |
|---|---|---|---|
| it support for small business bristol | 3 | Local firms, one directory | Buyer keyword, beatable competition |
| outsourced it support cost uk | 3 | Thin agency pages | Pricing intent, weak answers ranking |
| managed it services bristol | 3 | Local firms | Core service term, local page one |
| cyber essentials certification help | 3 | Mixed, some forums | High-value buyers, gap in quality |
| office 365 migration for small business | 3 | US-focused guides | UK angle missing entirely |
| it support contract vs pay as you go | 2 | Two thin blog posts | Comparison stage, easy to outdo |
| how much does it support cost per user | 3 | Vague agency pages | Nobody gives a straight answer |
| server keeps disconnecting from network | 2 | Forums and Reddit | Problem we fix, forums ranking |
| should a 10 person company have a server | 2 | Old forum threads | Buying-decision question, stale results |
| it support for accountants | 3 | One national firm, directories | Niche vertical, low competition |
| signs you need managed it support | 2 | Thin listicles | Early buyer stage, weak field |
| backup vs disaster recovery small business | 2 | Enterprise vendors | Small business angle unserved |
Notice what is absent: "what is cyber security" and "best laptops 2025" both had ten times the volume of anything in the table. Both scored intent 1. Both were cut without regret.
Worked check: "it support for accountants" showed under 50 monthly searches in Keyword Planner. It stayed in, because every one of those searchers is a firm looking to hire, and page one is directories.
If you want to see the full journey from a blank page to a list like this, our start-to-finish keyword research example walks through the finding stage in the same detail.
Where Helio SEO fits in
This scoring and page-one checking is exactly what Helio SEO does when it builds a monthly content plan for a site, and the reasoning behind each keyword choice is shown in the dashboard rather than hidden. If you enjoy this part of the work, the method above needs nothing but an hour and a browser. If you would rather it simply happened while you run the business, that is the part we automate.
