SEO Basics for Small Businesses

For a small business, SEO is decided by the order you work in, not by a clever trick. First your site has to be findable at all: served over HTTPS, not accidentally set to noindex, every important page linked from the homepage, readable on a phone. Then comes the part that makes the difference: you collect the questions people ask before they buy, group them by the need behind them, and build one page per need that answers that question in the first paragraph. After that, if you work locally, a verified Google Business Profile with current, non-contradictory details. Only then measurement, free of charge through Google Search Console. My working rule for this: five pages that each answer a real customer question in full will carry you further than ten technically polished pages that answer nothing. Google sets the expectation plainly itself: "There are no secrets here that'll automatically rank your site first in Google (sorry!)" Nothing further down this order ever replaces what is missing further up.

I run my business as a solo freelancer with a workforce of AI employees built on Claude, and my own website is where I worked this order out. How it is built is on Hiring AI Employees, where AI is worth anything in a small business is in Bringing AI to a Small Business. This post is about the order, not the toolbox: which tool measures what, and what it costs, is in SEO Audit Tools Compared.

The order that matters for a small business

Four stages, each one assumes the one before it. Start at stage four and you will measure a problem that lives at stage two.

Google says this about page experience: "Google Search always seeks to show the most relevant content, even if the page experience is sub-par." A fast page without an answer does not become an answer, and the best answer is useless if the page cannot be indexed.

Stage 1: Being findable at all

Good news: on a small, technically simple website, this stage can often be handled in one focused pass. After that, regular checks are enough instead of a permanent rebuild. Google itself writes that there is usually nothing to do: "You usually don't need to do anything except publish your site on the web."

Six points, covered once:

1. HTTPS, and a single address. Not with and without www at the same time: one variant, with the other redirecting to it.
2. No accidental noindex, no blocking robots.txt. The most common serious mistake happens at relaunch: the staging site's block goes live with it.
3. Every important page is quick to reach from the homepage. At most two clicks is my rule of thumb, not a Google requirement. The core of it: internal links help Google find and understand important pages, and an orphan page gives away that advantage.
4. A sitemap that is correct. Not necessarily complete, but without URLs that return 404 or carry a noindex.
5. The mobile view is the main view. Does the first screen of your service page on a phone say what you offer and who it is for?
6. Loading time should be decent, not perfect. Google calls page experience more than one signal: "There is no single signal. Our core ranking systems look at a variety of signals that align with overall page experience." On small websites, oversized original images are a common and quick place to check first.

A small business rarely needs more. The full technical run with tools is in How to Run an SEO Audit With AI.

Stage 2: The page that answers the question people ask before buying

This is where the biggest leverage sits, and where the fewest people work: technical items can be ticked off, content has to be thought up. Yet the raw material is there: every question you get asked again and again is raw material for a page.

First collect. Take your last twenty enquiries and note what those people wanted to know before they said yes or no. Not what you enjoy explaining. It is usually the same set:

• Roughly what does this cost, and what moves the price?
• How does it work, and what do I have to prepare?
• Do you work in my region, or for my industry?
• What happens if it does not work out?
• How are you different from the obvious alternative?

Then group. Not every wording deserves its own page. "What does this cost?" and "What moves the price?" are the same need; "What does this cost?" and "Do you work in my region?" are two. Answer every wording separately and you get overlapping half pages instead of strong answers.

Then build. One need, one page. The main question goes into the title and the heading the way a human would ask it, and the first paragraph answers it in full. Below that comes what carries the answer, including the related questions from the same group: a price range with the factors that move it, the process in steps, an honest note on limits. At the end, exactly one next step. Google poses the same standard as a check question: "Does the content provide a substantial, complete, or comprehensive description of the topic?"

And the counter-check. The same page asks: "Is the content primarily made to attract visits from search engines?" A page that exists only so that a term appears on it is something your reader notices first. What you would happily send to a real prospect by email, on the other hand, is almost always a good page for search too.

That is the working rule of this post, a heuristic and not a measured ratio: ten pages that only describe how wonderful the company is answer no question at all. Five that each prepare a buying decision answer five.

Stage 3: Showing up locally

If your customers come from the area, the Google Business Profile is the second biggest lever, and it costs nothing but care; what counts in the profile in detail and how to handle reviews has its own post. Google names three factors: relevance ("how well a Business Profile matches what someone is searching for"), distance, and prominence ("how well-known a business is"). Distance is not yours to influence, the other two are.

1. Verify the profile. That shows Google you are authorised to represent the business and makes an appearance more likely. Without verification the rest does not start.
2. Fill in everything. Category, address, opening hours, services. Google on this: "Businesses with complete and accurate info are more likely to show up in local search results."
3. Keep the details free of contradictions. Same name, correct address, reachable phone number, current opening hours, on your website as in the profile. Google nowhere demands character-identical spelling: an abbreviation or a different valid phone format is not a ranking problem. What hurts is contradictions between your legal page and your contact page.
4. Collect reviews and reply to them. On reviews themselves Google is clear: "More reviews and positive ratings can help your business's local ranking." On replies it only says that they show you value the feedback: evidenced towards readers, not listed as a ranking factor.

And the line that settles a good share of the offers in your inbox: "There's no way to request or pay for a better local ranking on Google." Anyone selling you a better spot on the map is selling something the owner of the map rules out.

Stage 4: Measuring instead of guessing

Without measurement you argue about taste, with measurement about numbers. The tool is free and it is called Google Search Console; setup and the numbers people misread are in Setting Up Google Search Console.

For a small business, three looks a month are enough.

On top of that, one rule that pays off more than any optimisation: every measure gets a date for its follow-up measurement, because an audit without a second measurement is an opinion with tables. And because more and more people ask their question in an AI chat, a second measurement setup next to it is worth having: whether your pages appear as a source in AI answers is its own metric and does not follow from your Google ranking. More on that in What Is Generative Engine Optimization?.

What you can safely leave alone

The most honest part of any SEO basics piece is the list of what you do not need. Every point here is stated, in substance, in Google's own starter guide or its spam policies.

That last point eats the most time: rewrite your titles every three days and you never get a clean measurement, because changes take hours to months according to Google. And if somebody promises you a position: "No one can guarantee a #1 ranking on Google."

The employee who keeps to this order

Knowing the order is easy, keeping to it is where it falls apart: after two weeks you slide back into single fixes, because small measurable things deliver a sense of progress faster than the fifth answer page. That is why this is a role here, not a good intention.

Sebastian, my AI employee for SEO and GEO, gets a URL and a goal and measures with real tools instead of from memory: Lighthouse and the PageSpeed Insights API, linkinator for broken links, advertools for crawls, Playwright CLI for fetches and screenshots, plus Search Console and Bing Webmaster Tools where access exists. Every raw output lands in a dated job folder, every finding names where it came from, and the result is a prioritised plan. He changes nothing on a live website and submits nothing in anyone else's account: he measures, I decide. Other people's servers get at most one request every three to five seconds, my house rule and not a universally safe limit: robots.txt and the target site's published limits come first.

The prioritising is the actual benefit: of thirty findings from a tool, twenty-five do not matter at your size, and the work is naming the three that fit your stage. Trends without evidence get named as trends without evidence.

If you would rather build a role like that yourself, the path is in Creating an AI Employee: How to Start, and how the roles work together here is in My AI Workforce. You do not have to think the first version through alone: put it up for discussion in the community or bring it to one of the calls, where people have already walked the same path.

Frequently asked questions

Is SEO worth it for a small business at all?

The basics almost always are, because they cost little time and keep working: reachable pages, a verified business profile, pages with real answers. What rarely pays off is competing for broad terms. The difference is which questions you pick.

Can I do SEO myself, without an agency?

Yes, the basics are built for that. Search Console, Bing Webmaster Tools, Lighthouse and PageSpeed Insights are free, and nobody can do the content work for you: only you know what customers ask before they buy. What is usually missing is the routine, not the tool; the four criteria that decide for or against an agency are written up separately.

How long does SEO take to work?

It depends on the change, and nobody can tell you honestly in advance. Google on this: "Some changes might take effect in a few hours, others could take several months." A technical fix such as a removed noindex often works quickly, but it depends on when Google fetches the page again.

What is the most common SEO mistake on small websites?

A website that describes how good the company is but has not one page answering a concrete question people ask before buying. Close behind, technically: a noindex that went live with the relaunch. The first costs you enquiries, the second everything.

Where to go from here

Take half an hour and write down the questions from your last twenty enquiries. Sort them by the need behind them until about five groups are left, then check whether each group has a page that answers its main question in the first paragraph. Usually three or four are missing: that is your plan for the next few weeks.

Alongside that, set up Search Console so the measurement is running while you write. The rest follows from the numbers. When you no longer want to do this yourself, the role gets more interesting than the tool: how my AI employees are built is on Hiring AI Employees, and the ready-made templates plus the people who use them are in my Community.