What technical SEO is and what to check first

Technical SEO is what lets Google find, crawl, and store the URLs you want to rank. Crawling is not indexing, and indexing is not ranking. robots.txt does not deindex a page, noindex is not read if the bot cannot enter, and a canonical is a hint, not an order. If the page that sells is excluded, the next step is not to write more text.

What technical SEO is

Technical SEO is what lets a search engine find a website, crawl the right URLs, and store the ones you want to compete. It is not writing the article or choosing the keyword. If the copy is good and the URL is set to noindex, blocked, or canonicalized to another page, the copy will not rank. Technical work opens that door. On-page SEO fixes what is inside once the door is open.

“What is technical SEO” is an informational search, smaller than “on-page SEO,” and it explains half of the “I published and I do not show up” cases. This page is the map: crawl, index, canonical, redirects, and speed, in that order. The manual for each piece already lives in other articles. Here you decide which one to open first.

Crawling, indexing, and ranking are not the same

Crawling means the bot entered the URL. Indexing means Google stored it and can show it. Ranking means Google shows it for a specific query, ahead of others. A URL can be crawled and not indexed. It can be indexed and not rank. It cannot rank in a stable way if it is not indexed.

That is why the first screen of a technical review is the Pages report in Search Console, not a health score from a tool. There you see whether the URL is indexed, excluded, blocked, or discovered and not yet stored. Each state has a different cause. Treating all of them as “Google does not like me” leads to a reconsideration request that does not apply.

Robots, noindex, and canonical: three levers people mix up

They are three different instructions. Mixing them is the most expensive technical mistake in a redesign.

robots.txt tells the bot what it should not crawl. It does not remove a URL from the index if it was already indexed: it stops the bot from seeing it again. A blocked URL can still appear if someone links to it, sometimes without a clear title. Do not use robots.txt to “deindex.”
noindex tells Google it may crawl the URL but must not store it. It is for thank-you pages, internal search results, and drafts that were published by mistake. For it to work, the bot has to be able to enter. If you also block the URL in robots.txt, the bot may never read the noindex.
canonical says which URL is the main one when several are similar. It is a strong hint, not an absolute order. If the canonical points at a page you do not want to rank, you are giving that page the signal. If every product page canonicalizes to the homepage “just in case,” the homepage receives a contradictory message and the product pages never compete.

The healthy combination on a service site is simple. Pages that sell are crawled, indexed, and canonicalized to themselves. Pages that should not appear carry noindex and remain crawlable. robots.txt is reserved for what you do not want the bot to spend time walking, not for hiding something that was indexed badly.

Sitemaps, redirects, and mobile

A sitemap is a list of URLs you would like Google to know about. It does not index. It does not fix a noindex. It should not include pages you have marked to stay out, URLs that redirect, or test URLs. If the sitemap and the index disagree, what Google finds by crawling wins, not the file you uploaded.

A 301 redirect says the old URL has moved permanently. It is the piece of a migration: every URL that had impressions needs its own destination, one by one, not a mass jump to the homepage. A 302 says the change is temporary and does not consolidate the signal the same way. Chains, where A redirects to B and B to C, slow the crawl and sometimes lose the destination. The procedure is in how to migrate without losing rankings.

Speed and mobile matter when the URL is already indexed and the experience is poor. Starting with a cache plugin while the homepage is noindex wastes the afternoon. When it is time to look at load metrics, the place is Core Web Vitals. On a site with few pages, crawl budget is almost never why you do not appear. On a store whose filters generate thousands of URLs, it is. The difference is in what crawl budget is.

The order of a technical review

List the URLs that must rank. Service, contact, and the guides that should already be in the index. Without the list, you end up auditing the whole blog and forgetting the page that sells.
See whether they are indexed. URL inspection in Search Console for a sample, plus the Pages report. Write down the exclusion reason as Google states it, not a tool’s interpretation.
If they are not indexed, find the lever. noindex, canonical, a block, or duplicate content. Do not request indexing in bulk until the cause is fixed. Google will exclude them again.
If they are indexed and do not rank, leave technical SEO. The next step is intent and on-page work, not another cache setting.
If they rank and the page fails on a phone, look at speed. Then, yes. With the URL in the index and a query that already shows impressions.
After a template or domain change, repeat step 2. That is when a staging noindex slips into production. You do not need a monthly technical audit if the structure has not changed and pages are still entering the index.

Symptoms that look technical and are not

“I do not show up for my keyword” can mean the results are full of maps, videos, or pages in another format. No canonical changes that. “I lost impressions” can be a brand query people stopped searching, not a penalty. Compare the queries report before you touch the server.

“The tool flags 40 errors” mixes a duplicate H1 on a thank-you page with a blocked homepage. Sort by whether the issue blocks indexing or ranking. Note the rest, and do not turn it into the project of the month. That priority is the same as in an SEO audit: a few fixes, each with a URL and a way to check they are done.

What to write down so the same failure does not return

Which environments may carry noindex, and which one is production, which must not.
Who publishes, and who checks after publishing that the new URL is allowed to be indexed.
The redirect map, if there is a migration, stored somewhere other than one person’s memory.
A sitemap that lists only what you want in the index, regenerated when a URL is deleted, not only when one is created.

Technical SEO is not filed away. You review it when you publish, when you migrate, or when indexed pages drop without you having deleted content. If the index is healthy and the page still does not show, the next step is not another plugin. It is the content and, once there is a page worth citing, links. Keeping that order is the work of an SEO agency, not a green score in a browser extension.

Technical symptom, likely cause, and next step

The reason Search Console writes down matters more than a tool’s score. The table turns the symptom into the first fix, not into a list of plugins.

What you see	Likely cause	Next step	Do not
Excluded by noindex	A template or staging site went live	Remove noindex and inspect again	Also block it in robots.txt
Blocked by robots.txt	A rule that is too broad	Open crawling for that path	Use robots.txt to deindex
Canonical to another URL	The hint points at the wrong page	Self-canonical if this is the main URL	Canonicalize every page to the homepage
Crawled, not indexed	Duplicate or nothing new	Merge or improve the URL that should remain	Request indexing in a loop
Redirect chain	A migration in hops	One 301 straight to the final URL	Send everything to the homepage
Indexed and slow on mobile	A heavy template	Measure Core Web Vitals on that URL	Start with speed if it is not in the index

Conclusion

Review technical SEO when you publish, when you migrate, or when indexed pages drop without you deleting content. Once the index is healthy, what does not rank is solved in the content, not with another cache plugin.

Frequently asked questions about technical SEO

What is technical SEO?

It is the set of conditions that let a search engine crawl a site, understand which URL to store, and show it properly. It includes crawling, indexing, redirects, canonicals, and speed. It does not include writing the article or choosing the keyword. That is on-page SEO.

What do you check first in technical SEO?

Whether the pages you want to sell are indexed, and whether the ones you do not want, such as filters or thank-you pages, are out. Then whether canonicals point at the wrong URL. Speed matters when the page is already indexed and the phone experience is poor. Starting with speed while the URL is noindex wastes the time.

Is technical SEO only for large websites?

No. A small site also drops out if robots.txt blocks the right folder, if everything is noindex after a redesign, or if a migration does not redirect. Crawl budget, by contrast, is almost never the problem on a site with few pages. Do not mix the two.

How often do you repeat it?

When you change domain, template, or URL structure, and when Search Console shows a drop in indexed pages that does not match deleted content. You do not need a monthly technical audit if nothing structural changed and pages are still entering the index.

¿Necesitas un equipo de Marketing?

Hablemos y hagamos crecer tu negocio