How does a small business do SEO?

Almost all SEO advice is written for teams. It assumes someone can publish weekly, run outreach campaigns, and maintain a content calendar. Handed to a business owner with four hours a month, that advice is not merely too big — it is misleading, because at that scale a different set of activities is optimal.

Scarcity does not just reduce how much you do. It changes what is worth doing at all.

The constraint changes the strategy

A team publishing twelve pages a month can afford for eight to fail, because the four that work carry the programme. That is a portfolio strategy, and it is entirely rational at that volume.

You cannot run a portfolio with four hours. Every hour has to have a reasonable expectation of return on its own, which rules out most of what SEO teams do and rules in a narrow set of activities with unusually reliable payoffs.

Three principles follow.

Improve rather than create. A page that already gets impressions has completed the slow parts — discovery, indexation, initial evaluation. Improving it is a short push. A new page starts from zero and needs months. At low volume, always take the short push.

Prefer work that cannot fail quietly. Rewriting a title on a page with 4,000 monthly impressions has a measurable outcome within weeks. Publishing an article has an outcome you will not be able to read for a quarter. When you get four hours, take the measurable one.

Do the free things completely before paying for anything. Indexation, titles, internal links, business profile, sales-call questions. Most small businesses have never finished this list, and it is the list with the best returns on any budget.

Three advantages you actually have

Small businesses are usually told they are at a disadvantage in search. Mostly true, with three real exceptions.

You know what customers actually ask. You have spoken to them. You know the phrasing, the objections, the thing everyone gets confused about. Large competitors commission keyword research to approximate this from data, and it produces a worse answer than the one already in your head. The questions you hear weekly are search queries, phrased the way people search, with buying intent already attached.

You can be specific in ways large companies cannot. A national competitor writes for the broadest possible audience because that is what justifies the cost of the page. That leaves every specific case uncovered — the awkward constraint, the unusual configuration, the situation that applies to a hundred people a month. You can write for those, and they are winnable precisely because nobody with a content budget can justify writing them.

You can decide in a day. A change that takes a large organisation two months of approvals takes you an afternoon. Where results are dated and the question has moved, speed genuinely wins the position.

Notice all three point the same way: narrower and faster, rather than more.

The first month, hour by hour

Roughly ten to fifteen hours, concentrated, producing most of the value you will get in the first year.

Hours 1–2: find out what is true. Set up Google Search Console if it is not already. Check the Pages report — how many URLs are indexed against how many you have. Check which queries you already appear for. Most owners are surprised in both directions: some pages missing entirely, some queries producing impressions nobody knew about.

Hours 3–4: fix indexation. Whatever the first check turned up. Unindexed pages are wasted assets and this is the highest-leverage repair available.

Hours 5–6: rewrite titles on your top ten pages by impressions. Target phrase near the front, under sixty characters, written as a promise rather than a label. This is the fastest measurable change in SEO and most small sites have never done it deliberately.

Hours 7–8: write down the twenty questions customers ask. From memory, from the inbox, from what the sales conversation keeps returning to. In their words, not yours. This is your keyword research and it is better than a purchased list.

Hours 9–10: fill out your business profile completely. If you serve anywhere physical, this outranks almost everything else on this list for local queries. Complete every field — categories, service areas, hours, services, real photographs — and set up a genuine process for asking satisfied customers for reviews. Review volume and recency are among the strongest local ranking inputs, and they are almost entirely within your control while being nearly impossible for a national competitor to replicate in your town.

Hours 11–15: answer the three best questions from hour seven. One page each, answer in the first hundred words, structured for scanning. Not articles — answers.

The ongoing four hours

A monthly rhythm that fits in an afternoon.

Done for twelve months, that is twelve improved pages, twelve new answers and twenty-four reviews. It outperforms most of what agencies deliver on a small retainer, because almost all of that time goes to work with a direct expected return rather than to reporting.

What to ignore entirely

Permission to skip things, which is the part usually missing from small business SEO advice.

• Publishing on a schedule. Frequency is not a ranking factor. Publish when there is a question worth owning.
• Chasing perfect page speed scores. Get off the genuinely slow end — compress images, remove unused scripts — then stop. Beyond that, returns collapse.
• Link building campaigns. At this scale, outreach is a poor use of hours. Get listed in genuine industry directories and partner pages, then let links follow from being useful.
• Most audit tool warnings. The 340 missing alt attributes are not why you are not ranking.
• Broad head terms. The two-word phrase for your category is held by competitors with budgets. It is not this year’s work.
• Social media as an SEO activity. It has value; ranking is not it.

The unifying reason: each of these has either a low expected return or a return you cannot measure, and neither is affordable when hours are the scarce input.

How to tell it is working

The hardest part of doing this yourself is not the work. It is knowing, four months in and with nothing obvious happening, whether to continue.

Three numbers answer it, all free, all in Search Console, and all of which move before traffic does.

Impressions. The count of times you appeared in results at all. This moves first — often within weeks — because appearing at position 40 counts. Rising impressions with flat clicks is not failure; it is the normal shape of early progress and the single most misread signal in SEO.

Average position on the queries you targeted. Filter to the specific questions you wrote pages for. Movement from 34 to 19 produces no visitors and is real progress, and it is the clearest evidence that the work is landing before any commercial result appears.

Count of queries in the top ten. The one to report to anyone else, because it is hard to game and it correlates with revenue better than raw traffic does.

Check them monthly, decide quarterly. The mistake almost everyone makes at this scale is checking weekly, seeing noise, and concluding it is not working — then changing approach, which resets the clock and guarantees the next quarter looks the same.

A rough expectation for a small site doing the four-hour rhythm honestly: impressions moving by week six, average position improving by month three, first meaningful non-branded clicks around month four to five, and a clear trend by month six. If all three numbers are flat at month four, something is structurally wrong — usually that the target queries were never winnable, or the pages are not indexed. Both are checkable in an hour, and both are covered in why a site is not ranking.

When to pay someone

Three situations where money genuinely beats time.

A migration or redesign. The highest-risk event in SEO. Changing URLs without correct redirects can remove most organic traffic within weeks, and it is hard to reverse. Paying an experienced person for two days of oversight here is the single best-value expenditure available.

A traffic collapse you cannot diagnose. If traffic halved and the obvious checks came back clean, buy a diagnosis rather than a retainer. It is a bounded piece of work with a clear deliverable.

Genuine technical complexity. JavaScript rendering problems, faceted navigation, international targeting. These need specialist knowledge and are hard to learn on the job.

Everything else on this page you can do yourself, better, because you know the business. The decision framework is in do I need an SEO agency.

One addition worth making room for. When someone asks an assistant to recommend a business like yours, it composes an answer from what it retrieves — and for small and local businesses that increasingly includes review content, directory listings and community discussion rather than your own site. It is worth knowing what those answers currently say about you, because it is a conversation happening about your business that produces no traffic, no notification and no trace in any report you read.

Frequently asked questions