Technical SEO services, built into the site and checked every week

Who needs technical SEO services?

Technical SEO is for a site that has grown, moved or changed hands, when nobody is sure what search engines can still reach. Pages that used to bring in calls stop showing up. Search Console lists hundreds of addresses it won’t index. A redesign went live and the old links now end on error pages. We’re based in Valencia, California, and the technical work is done remotely, for businesses across the US. What the technical work has to handle grows with the site:

• Established local businesses: a site of a few service pages and a contact page, where one wrong setting can hide the pages that bring in calls.
• Multi-location and regional businesses: a page for each location, each needing its own markup, its own place in the sitemap and links from the pages around it.
• National businesses: large sites built from templates, where crawling, rendering and duplicate pages decide how much of the site search engines ever see.

If your site is already crawled and indexed cleanly, technical work may not be where your next dollar should go, and we’ll say so.

What is technical SEO?

Technical SEO is the part of SEO that deals with how search engines reach and process a site, before anyone weighs what its pages say. Google’s guide to JavaScript and search describes three phases: crawling, where Googlebot fetches a page; rendering, where it runs the page’s scripts; and indexing, where it stores what it found. Technical work covers what can go wrong in each one: the server’s response codes, redirects, robots.txt rules and noindex tags, sitemaps, canonical tags, rendering, structured data, HTTPS and page speed.

Search Console is the first place to look. The Page indexing report says which pages made it into the index and why the others didn’t, URL Inspection shows how Google saw one address, and the Core Web Vitals report shows speed from real visits. The words on each page are a separate job, covered on our on-page SEO page. The two meet in internal links and structured data.

What does a technical SEO audit check?

A technical SEO audit works through six areas:

• Status codes and redirects: every address returns the right response, and every redirect reaches a live page in one hop, with no chains and no loops.
• Robots rules and noindex tags: what robots.txt blocks and which pages carry a noindex tag, with a reason for each. Google’s guide says robots.txt is there to manage crawler traffic, and that a blocked address can still appear in results; a noindex tag is what keeps a page out.
• Sitemaps: Google says a sitemap helps it crawl a site more efficiently, and that a site of about 500 pages or fewer, linked well from page to page, may not need one. A sitemap makes no promise that a page gets crawled or indexed, so the audit compares it with what the Page indexing report shows.
• Canonical tags: which version of each page Google treats as the main one. Google weighs redirects and canonical tags as strong signals and a listing in the sitemap as a weak one, and it prefers HTTPS pages over HTTP ones.
• Rendering: whether the words and links on each page are in the HTML the server sends, or appear only after scripts run in the browser.
• Page experience: Core Web Vitals from real visits on phones and desktops, HTTPS across the site, and pages that work at phone width.

With us, the audit comes first in the monthly SEO work, and the fixes follow from what it finds. Our page on SEO audit services covers what an audit report and a monthly report should hold, and SEO pricing sets out what US agency price pages and guides show for the monthly work at each level.

Which technical SEO issues matter most?

The ones that hide the pages that bring in work, roughly in this order:

1. A stray noindex tag or robots.txt block: such as one left behind by an update or a staging copy. A noindex tag takes a page out of the index on its own.
2. Broken paths: links and redirects that end on an error page, which a redesign can leave behind.
3. Split versions: the same page at two addresses, such as with and without www, or on HTTP and HTTPS. Google picks one to show, and it leans toward the HTTPS version.
4. Words that need scripts: text and links that appear only after JavaScript runs. Google can process JavaScript, but rendering waits in a queue, and other crawlers may not run scripts at all.
5. Slow pages on phones: Core Web Vitals outside Google’s thresholds on the pages people reach from search.

Crawl budget is a separate question. Google wrote its crawl budget guide mainly for three kinds of site: those with a million or more pages that change about once a week, those with ten thousand or more that change daily, and those where Search Console lists a large share of the addresses as discovered but not yet indexed. It says a site whose pages are crawled the day they’re published doesn’t need the guide.

Why does technical SEO matter for AI search too?

Because the AI features in Google Search draw on the same crawling and indexing. Google’s guide to its generative AI features says the way Search finds and processes pages is still the core of how its AI systems reach a site’s content, and that the existing technical SEO practices are still worth doing. By the guide’s own list, three things make a page eligible for AI Overviews or AI Mode: Google has indexed it, it qualifies to show in Search with a snippet, and the site’s Search Console setting includes it in Search generative AI features. Even then, Google is clear that being eligible is no promise of being shown.

That third condition is a setting. By August 31, 2026, Search Console had given every site a Search generative AI control, switched to include unless someone changes it. Switch it to exclude and the site’s pages are kept from AI Overviews, AI Mode and Discover’s AI features alike, so if your site never appears in them, look under Settings first.

The same guide says no new file or special markup is needed. On the sites we build, robots.txt names the AI crawlers allowed to read the site, and an llms.txt file sits beside it. Google Search ignores llms.txt. Google calls it fine to keep one for other services that read it, and says it neither helps nor harms a site in Google Search. How AI answers mention a business is its own piece of work, covered under generative engine optimization.

Can you do technical SEO on a site you didn’t build?

Yes, by rebuilding it in our own system first. We build on Astro or Next.js, and we don’t take on SEO inside a WordPress, Shopify, Wix or Squarespace site. Technical SEO works in the code and on the server: response codes, redirects, rendering and how the sitemap is made. So a site built elsewhere is rebuilt in our system, keeping its page addresses, with a permanent redirect for any address that has to change. Our WordPress SEO and Shopify SEO pages cover those two moves.

On the sites we build, new and changed pages are announced through IndexNow, which Microsoft Bing, Naver, Seznam.cz, Yandex and Yep support. Google isn’t on that list, so for Google the monthly work sends re-index requests through Search Console.

What happens to your URLs when a site moves?

Google’s guidance on moving a site with URL changes sets out the steps:

• A map from old to new: each old address paired with its new one.
• Permanent redirects on the server: Google advises server-side permanent redirects, kept for as long as possible and generally at least a year.
• Redirects tested: Google says to check that each old address now leads to the right new one.
• Links and sitemaps updated: internal links point straight at the new addresses, and the new sitemap goes to Search Console.
• Change of Address: for a move to a new domain, Google asks for a Change of Address in Search Console, filed for the old site.

Google also says to expect rankings to fluctuate for a while, and that a medium-sized site can take a few weeks or more before Google shows the new addresses in place of the old ones, with larger sites taking longer. When we rebuild a site, the redirect map and the permanent redirects are part of the build. Our page on website migration covers the move in more depth.

Does page speed affect rankings?

It counts, and relevance counts more. Core Web Vitals are among the signals Google’s ranking systems use, and Google gives a threshold for each: the largest piece of content should load within 2.5 seconds, the page should respond to a tap or click in under 200 milliseconds, and its layout shift score should stay under 0.1. Google also says it seeks to show the most relevant page even when that page’s experience is poor, that there is no single page experience signal, and that good results in Search Console’s Core Web Vitals report are no promise that a page will rank at the top of Google’s results. Chasing a perfect score for SEO alone, it adds, may not be the best use of your time.

Speed starts in the build, since a slow template slows every page made from it. After launch, Search Console’s Core Web Vitals report shows how real visits fare.

Does JavaScript hurt SEO?

It can make SEO harder. Google says it can process content in JavaScript as long as nothing blocks it, and its guide to generative AI features calls SEO on sites built with JavaScript frameworks generally more complex. Pages wait in a render queue before Google runs their scripts, which can take a few seconds or longer, and Google’s JavaScript guide says server-side rendering or pre-rendering is still a great idea, because it makes a site faster for people and crawlers and not every crawler runs JavaScript.

Recent sites we build run on Astro. Astro’s documentation says websites run faster when they render HTML on the server, and it lists zero JavaScript by default among its features. So the words and links search engines need are in the HTML the server sends.

What does structured data do?

Structured data is code that describes a page in terms search engines can read without guessing: the business, its services, its locations, and its questions and answers. Google says it uses structured data to understand a page’s content, recommends the JSON-LD format, and offers the Rich Results Test to check it. Two of Google’s rules are worth knowing before anyone adds it: mark up only what visitors can see on the page, and treat a rich result as something markup makes a page eligible for, with no promise that it will show.

The sites we build carry structured data, and adding or correcting it on a site’s existing pages is part of the monthly SEO work.

What should you ask a technical SEO company?

Whether you hire a technical SEO agency, a larger company or a freelancer, Google’s guide to hiring an SEO suggests questions for the interview. Here are five of them, with our answers:

• Do you follow the Google Search Essentials? We earn links and never pay for them, and we don’t sell link counts. Google’s spam policies list links bought or sold for ranking as link spam.
• What results do you expect, and in what timeframe? We set targets from the research and report against them, without promising a ranking or a date. Google’s SEO starter guide allows anything from a few hours to several months for a change to take effect.
• How do you measure your success? By leads first: the one-page report we send on the last day of each month opens with them.
• How long have you been in business? Midnite Systems was formed in 2023. The co-founders are Nicolas Perez, our CEO, and Noah Alvarez, our CCO.
• How can I expect to communicate with you? First comes a 15-minute call, with Nicolas or Noah. After that come the monthly report and a review each quarter.

Google adds that your SEO should ask about your business: what makes it unique, who your competitors are, how search results can help and how your customers find you. If your site has moved, grown or slipped in search and you want to know what search engines can still reach, book a 15-minute call.