What is SEO? A plain explanation, without the jargon

What a search engine actually does

Three steps, and every SEO decision maps back to one of them.

It crawls. Automated software requests your pages the way a browser would. If it cannot reach a page, or the page returns nothing readable, that page does not exist as far as search is concerned.

It indexes. The engine decides whether the page is worth storing and what it is about. Not every crawled page is stored. Thin, duplicated, or contradictory pages get dropped here, silently.

It ranks. When someone searches, the engine picks from what it stored and orders the results. This is the only step most people think about, and it is the one you have least direct control over.

The order matters. A page that cannot be crawled will never be indexed, and a page that is not indexed can never rank, no matter how good it is. Most sites with disappointing search performance have a problem in step one or two while spending all their effort on step three.

The three parts of the job

Technical. Can the engine reach your pages, read them, and understand their structure? This covers crawling, indexation, site speed, internal linking, correct handling of duplicate and multilingual versions, and whether your content exists in the HTML the server sends rather than being drawn in afterwards by scripts. This part is unglamorous and it is where the largest, cheapest wins usually sit.

Content. Do your pages answer what people actually type? Not what you call your product internally. What a buyer types when they do not yet know your name. That gap is usually wider than owners expect, and in Arabic it is wider still, because the same intent gets typed in Modern Standard Arabic, in dialect, and in Latin letters.

Authority. Do other sites reference yours? Search engines treat a link, a mention, or a citation as evidence that something is worth surfacing. This is the slowest part, the easiest to fake badly, and the one where bad advice does the most damage.

What has changed, and what has not

The visible change is that a growing share of searches end in an answer rather than a list of links. Someone asks an assistant, gets three names, and contacts one of them. Fourth place in that answer does not exist.

The unglamorous truth is that this did not invalidate the fundamentals. It raised the price of failing them. An assistant answers by reading sources it can fetch and parse. If your content is not in the server’s HTML, or your robots file blocks the crawler, you are absent from both the classic results and the generated answer, for the same reason.

So the work did not change as much as the stakes did. Being readable, being specific, and being referenced elsewhere still describe the job.

What to ignore

Guaranteed rankings. Nobody sells positions in Google, and nobody can guarantee a citation inside an assistant’s answer. These systems do not publish their logic and do not sell placement. Anyone promising a specific position is either misunderstanding their own product or counting on you not checking.

Keyword density targets. Writing to hit a percentage produces text that reads as though it were written to hit a percentage. Write for the person, use the words they use.

Bought link packages. Links acquired in bulk from sites that exist to sell links are the single most common way an owner damages a domain while believing they are investing in it.

Anything measured only in tasks. “Twelve blog posts published” is not a result. Whether enquiries went up is a result. If your reporting cannot connect the two, you are paying for activity.

Where to start

Start with diagnosis rather than execution, because the alternative is spending six months on the wrong third of the problem.

You can begin yourself, right now: open your homepage, view source, and search it for the text of your service. If the words are not written there, search engines and AI tools cannot find them either, and that is the first problem worth fixing, before anything else.

If it surfaces something worth going deeper on, the Digital Presence Audit covers every surface growth can come from: technical faults, what your market actually searches in Arabic and English, where competitors beat you on every measurable channel, whether AI answers cite you, and why the visitors you already have do not enquire. It ends in a plan ordered by impact rather than a list of faults.

That ordering is the whole point. Any tool can hand you a long list of faults; what you are paying for is knowing which handful of them to start on tomorrow, and which ones you can leave alone without worrying.