Most SEO advice is a list of tactics with no sequence, which is why so much of it gets done in the wrong order. This is the sequence that works, and what is safe to ignore until the previous step is finished.
Key takeaways
- SEO has a dependency order: crawlability, then intent, then content, then links. Work done out of order mostly gets wasted.
- Ranking for a term nobody searches with buying intent is a cost, not a win, pick topics by what the searcher wants to do next.
- Most small sites lose more traffic to duplicate, thin and orphaned pages than they ever gain from new ones.
- Expect three to six months before compounding starts. Anything faster is usually paid traffic wearing a disguise.
On this page
What SEO actually is
SEO is the work of making it obvious to a search engine which questions your site deserves to answer. Not tricking it, and not stuffing words into pages. Every durable tactic reduces to one of three things: letting the engine reach the page, letting it understand the page, and giving it a reason to trust the page.
That framing matters because it tells you what to ignore. If a tactic does not make a page more reachable, more understandable or more trustworthy, it is decoration. A great deal of what gets sold as SEO is decoration.
The order the work has to happen in
The four stages depend on each other, so doing them out of order wastes the work. Publishing excellent content on a site the crawler cannot reach is the most common and most expensive version of this mistake.
- Reachable. The page returns 200, is not blocked in robots.txt, is not
noindex, and is linked from somewhere a crawler goes. - Understandable. One clear topic per page, a title that names it, headings that structure it, and internal links that place it in context.
- Worth ranking. The page answers the question better, faster or more specifically than the ones currently ranking.
- Trusted. Other credible sites reference it, and the site as a whole has a track record on the topic.
Stage four is the one everyone wants to start at, because buying links feels like progress. It is also the only stage that cannot compensate for failures in the first three. A trusted domain with an unreachable page still ranks nothing.
Choosing what to rank for
Pick topics by what the searcher intends to do next, not by search volume. A term with 200 searches a month from people ready to buy is worth more than one with 20,000 from people writing a school essay.
The practical test is to search the term yourself and look at what is already ranking. The results tell you what the engine believes the intent is, and you do not get to argue with it. If page one is all free tools, a sales page will not rank there no matter how good it is.
| Intent | The searcher wants | What ranks |
|---|---|---|
| Informational | To understand something | Guides, explainers, definitions |
| Commercial | To compare options | Comparisons, reviews, alternatives |
| Transactional | To buy or sign up | Product, pricing and category pages |
| Navigational | To reach a specific site | That brand’s own pages |
What to change on the page
One page, one job. Pages that try to rank for several unrelated things usually rank for none of them, and they split your internal links between competing targets.
The title and the first heading
The title tag is the promise shown in the results; the H1 is the promise kept on the page. They should agree. If a visitor cannot tell within one line that they landed in the right place, the engine cannot either.
Internal links
Internal linking is the most under-used lever available to a small site, because it is free and entirely under your control. Every important page should be reachable in two or three clicks from the homepage and linked from related articles using words that describe the destination.
Pruning, which nobody wants to do
Deleting or consolidating weak pages usually helps more than adding new ones. Thin, duplicated and orphaned pages dilute the signals of the pages you care about. If a page has had no impressions in a year and answers nothing a customer asks, it is costing you.
How to know it is working
Watch impressions before clicks, and clicks before revenue. Those three move in that order, usually weeks apart, and confusing them makes people abandon work that was about to pay off.
Google Search Console is enough for the first year. Look at impressions by query to see what the engine is starting to associate you with, average position to see movement long before it produces traffic, and click-through rate to find pages that rank but do not earn the click: usually a title problem, not a ranking problem.
If a page is gaining impressions and losing position, you are competing for the right topic and losing on quality. That is a much better problem than silence.
Serhii Yelbaiev, Well Web Marketing
If you would rather start from a diagnosis than a guess, our SEO and AI-search audit works through the first three stages on your actual site and hands back the fixes in priority order.
Frequently asked questions
On a site with some history, three to six months before compounding becomes visible, and closer to nine on a new domain. The first signal is impressions rising for terms you did not previously appear for at all.
No. You need to answer the questions your buyers ask. For many small businesses that is a dozen well-made service and location pages, not a weekly blog nobody reads.
Yes, but the goal widens. Answer engines still crawl and rank pages before summarising them, so the foundations carry over, what changes is that you also need passages that can be quoted on their own.
Do the first two stages yourself if you can; they are mostly decisions about your own business. Bring help in when the constraint becomes execution volume or technical debt.
Sources and method
Based on our own client work across small and medium sites, measured in Google Search Console. Intent categories follow the standard informational / commercial / transactional / navigational split used across the search industry. Last reviewed February 2026.