Specialization

Technical SEO Specialist

Technical SEO is the plumbing that decides whether search bots and AI bots can find, read, and trust your pages. This path is the role at scale: what to master from Entry to Senior, whether you run a big distributor catalog, a multi-location home-services site, or a dental group with several offices.

For: Technical SEOs and SEO engineers on large or multi-location sites
Proficiency: Entry → Senior
Duration: Self-paced

On this path

Entry
01 · Find every page that should rank, and make it reachable
02 · See which bots actually visit you
03 · Check that AI bots can reach your pages
Mid
04 · Keep junk pages from wasting crawl budget
05 · Make sure a bot can read the page
06 · Be fast, especially on phones
07 · Mark up the page so engines understand it
Senior
08 · Don’t lose your rankings in a site rebuild
Need this done?

8 skills

Reviewed June 2026

Before this path

Know this first: SEO Specialist.

This path is for people who do the technical side of SEO. You already know the basics. Here is what the job becomes when a site gets big or spreads across many locations, in the order you would learn it, from Entry to Senior. The examples span the kinds of businesses we work with: industrial e-commerce, home services, and dental. Each skill ends with a check you can run on a real site.

Entry — run the checks

Skills 01–03

By the end you can audit a large or multi-location site, read its server logs, and prove a search or AI bot can reach every page that should rank.

01

Find every page that should rank, and make it reachable

Make sure bots can actually reach every page you want found, and that none are orphaned (stranded with no links pointing to them).

Why it matters

Big sites hide pages from bots without meaning to, and the cause changes by business. A distributor has hundreds of thousands of product pages built from one shared template (you fix the template, not pages one by one). A home-services company has a page for each city and service. A dental group has a page for each office and treatment. If nothing links to those pages, bots never find them, and no amount of good writing will rank them.

In the field

A 60-location roofing site had a page for every city, but the only links to them sat in a dropdown that loads with JavaScript (code the bot does not run). Bots saw the homepage and almost nothing else.

Edge cases

Pages reachable only through a JavaScript menu or a search box a bot cannot use.
Key pages buried five clicks deep with no direct link.
Old pages that lost their internal links in a redesign and went orphan.

Proficient when

You can list the pages you want found, then prove a bot can reach each one by following links from the homepage.

02

See which bots actually visit you

Read your server logs (the record your site keeps of every visit) to see which bots reach which pages.

Why it matters

Logs show what is really happening, not what you hope. Sort the search bots (Google, Bing) from the AI crawlers (GPTBot, PerplexityBot, ClaudeBot: the bots that fetch pages for AI answers). Bots do not visit evenly. They hit some pages over and over and never find others. That tells you which pages are reachable and which are orphaned.

In the field

On a dental group’s site, the logs showed Googlebot crawling the homepage every day but never reaching the individual office pages. That is a reach problem, not a content problem.

Edge cases

Logs wiped or trimmed too soon to show the long tail of pages.
A CDN (a cache that sits in front of your site) serving saved copies, so your own logs miss real bot visits.
A fake bot spoofing GPTBot. Verify it against GPTBot’s published IP ranges.

Proficient when

You can pull a week of logs, group the visits by bot, and say which pages get crawled, which are thin, and which are orphaned.

See AI crawler

03

Check that AI bots can reach your pages

Confirm the AI crawlers can actually load your important pages. This is usually the first thing you hand a client.

Why it matters

Your robots.txt file (the note at your site root that tells bots what they may visit) is not the whole story. A WAF or bot protection (the security layer in front of your site) can turn a bot away before it ever reads that note. The best page in the world is invisible if the AI bot gets stopped at the door.

In the field

A roofing company’s pages stopped showing up in Perplexity answers. A WAF rule set to block "all bots" was also blocking the AI crawlers. Allowing those specific bots back in brought the citations back.

Edge cases

Your robots.txt allows the bot, but your WAF blocks it. The two disagree.
You block the bot that trains AI models (GPTBot) but also block the one that fetches live answers (OAI-SearchBot).
A rule that only blocks data-center traffic, which is exactly where bots live.

Proficient when

You can load one important page as a normal visitor, then again as an AI bot, and explain any difference in what comes back.

See AI crawler

Mid — own the whole site

Skills 04–07

By the end you own how the whole site gets found, read, and indexed: no crawl budget wasted on junk, pages bots can actually read, fast on phones, and marked up so engines understand them.

04

Keep junk pages from wasting crawl budget

Stop the site from creating thousands of near-empty pages that waste a bot’s crawl budget (the limited time it spends on your site).

Why it matters

Every business finds a way to waste crawl budget. A catalog’s filters (faceted navigation) multiply into millions of near-empty URLs. A home-services site spins up thin, near-duplicate city pages (doorway pages). A dental group copies the same service page onto every office. Bots burn their time on the junk instead of the pages that win customers.

In the field

One catalog category quietly created about 40,000 filter URLs a bot could reach, against about 1,800 real products. On a home-services site, the same waste showed up as 300 near-identical "roof repair in [city]" pages with nothing local on them.

Edge cases

A canonical tag (the tag that names the original page) pointing a filter page at itself instead of the clean category.
Pages set to "noindex" (hidden from Google) that bots still waste time walking through.
Near-duplicate city or office pages that add no real local information.

Proficient when

You can take one section, list every URL a bot can reach, and decide which few are worth keeping.

05

Make sure a bot can read the page

Put your key content in HTML the bot reads on arrival, not behind JavaScript it has to run first. (Serving the finished HTML up front is called server-side rendering.)

Why it matters

Bots read the page as it arrives, and many do not run JavaScript (the code that fills in details a moment later). So if your specs, prices, booking widget, or office hours only appear after that code runs, most AI crawlers see a blank page.

In the field

A dental group’s booking widget and office hours only loaded with JavaScript. Turn JavaScript off and the page was nearly empty. That blank version is what the bot sees.

Edge cases

Later pages of a list (page 2, 3, 4) that only load with JavaScript, so bots never reach them.
Content pulled in by a third-party widget the bot cannot load.
Prices or details behind a login, invisible to any bot.

Proficient when

You can turn off JavaScript on an important page and still read the content that matters.

06

Learn anytime

Be fast, especially on phones

Keep pages fast on a mid-range phone, where most local and home-services searches happen. The yardstick is Core Web Vitals (Google’s page-speed and stability scores).

Why it matters

Speed is both a ranking factor and a sales factor. It bites hardest where people search on a phone in the moment: a homeowner looking for a roofer, a patient looking for a dentist nearby. A slow page loses the click before it even loads.

In the field

A roofing site’s autoplay hero video pushed the largest element’s load time past five seconds on a phone connection. Rankings and form fills both dropped.

Edge cases

Huge unoptimized hero images or autoplay video.
Third-party chat and tracking scripts freezing the page while it loads.
A fast desktop score hiding a slow phone score.

Proficient when

You can run a page through a mobile speed test and name the top two things slowing it down.

07

Mark up the page so engines understand it

Add structured data (a hidden, machine-readable summary of the page) so search and AI engines know what the page is.

Why it matters

Structured data tells an engine "this is a product / a local business / a service / a review." It feeds rich results (stars, prices, hours) and helps AI answers quote you correctly. The right type depends on the business: Product for a catalog item, LocalBusiness and Service for a contractor or dental office, Review for ratings.

In the field

A multi-location dental group added LocalBusiness and Review markup for each office. Their offices started showing star ratings and hours in results, and AI answers began naming the right location.

Edge cases

Markup that does not match what is visible on the page (against Google’s rules).
One generic Organization block instead of a LocalBusiness block per location.
Fake or self-serving review markup, which gets penalized.

Proficient when

You can pick the right schema type for a page and validate it with no errors.

Senior — protect it through change

Skill 08

By the end you can carry the whole technical setup through a site rebuild without losing rankings, and you set the standard the rest of the team follows.

08

Don’t lose your rankings in a site rebuild

Carry your rankings safely through a replatform (a full site rebuild on a new platform). It is the riskiest moment in the job.

Why it matters

A rebuild can undo years of work in a weekend. URLs change. Redirects (the rules that send an old URL to its new home) get lost. Pages go back to needing JavaScript to load. A fresh WAF setup quietly blocks the AI crawlers again.

In the field

A home-services company relaunched on a new platform and 1,000 city and service URLs changed with no redirects. Traffic cratered. The new WAF default also blocked the AI crawlers, so they stopped visiting.

Edge cases

A "noindex" (hide-from-Google) setting left on from the staging site (the test copy) that ships to the live site.
Redirects matched one-to-one but missing the filter and search-result URLs.
The new platform hiding content behind JavaScript again, undoing your earlier fix.

Proficient when

Before any replatform, you save the top 1,000 crawled URLs and their status codes, and you can prove the live site matches afterward on redirects, page loading, and bot access.

For buyers

Need this done?

A technical SEO specialist makes a big site crawlable and indexable, then checks that AI crawlers (GPTBot, PerplexityBot, ClaudeBot) can even reach the pages — the precondition for any GEO work paying off. The shape repeats across verticals: a distributor’s faceted catalog, a contractor’s hundreds of location pages, a dental group’s per-office duplicates.

Signs you need this work

You rank fine on some pages but suspect AI engines never see the rest of the site at all
Your filters, location pages, or per-office templates spin up far more indexable URLs than you have real products or services
Your key content lives in a JS app, in PDFs, or behind a login or portal that anonymous crawlers can’t reach
You’re about to replatform or migrate and don’t want to lose indexation, or crawler access, in the move

Demand for this work is lumpy. It’s project-shaped — an audit, a replatform, a faceted-nav or location-page cleanup — with monitoring in between, not a steady full-time load. For most sub-enterprise businesses that means buying it as a fixed-scope project or a light retainer, not a headcount line.

A full-time hire only makes sense where an SEO team already exists for the specialist to plug into. A lone technical SEO with no content or GEO counterpart is a mishire. If you hire one search person total, make it a hybrid who also owns content and GEO (see the GEO Specialist and SEO Specialist paths).

This is exactly the kind of work Sale Solution takes on as a project — the technical and GEO groundwork behind getting an industrial, home-services, or dental site cited in AI answers.

Cost reality

ZipRecruiter puts the technical SEO specialist average at about $81K (roughly $28–57/hr). Because the work is lumpy, a project or retainer usually beats a fully-loaded full-time hire for sub-enterprise businesses. (Source: docs/strategy/career-path/03-roles.md §4.2, citing ZipRecruiter salary pages.)

Where this sits