A broken site can't rank — and most sites are quietly broken. My fixed-scope technical audits surface crawl, index, and performance issues, ranked by revenue impact, in a format your developers can execute without translation.
Get a Quote View Case StudiesSix years of compounding SEO for an Atlanta-area plumber — including a full domain migration that didn't drop a single ranking.
A technical SEO consultant makes sure search engines can reach your pages, understand them, and serve them quickly enough that people stay. Everything else in search engine optimization sits on top of that. You can publish the best page on the internet, and if Googlebot cannot crawl and index it, it does not exist.
The job splits into four questions, asked in order. Can a crawler get to the page? Once there, can it render and index the page? Once indexed, is this the version of the page you want to rank, or one of six near-duplicates? And once it ranks, does the page load fast enough that the click turns into a session instead of a bounce?
Most sites fail somewhere in the first two. I have inherited sites with thousands of crawl errors, orphaned category pages that no internal link ever pointed at, and staging environments quietly indexed alongside production. None of that shows up in a content audit. All of it caps what your content strategy can ever return.
This is why technical seo audits vary so wildly in value. What separates technical seo consultants from a crawl tool is the second half of the work: deciding what to fix first, and writing the fix so a developer can ship it without a translation layer. A crawler will hand you 4,000 issues. Roughly nine of them are costing you money this quarter. My job is to tell you which nine, prove it with data, and hand your team a ticket they can close.
Crawlability is the first gate. Googlebot arrives with a finite budget for your site, follows what it can reach, and leaves. If that budget is spent on faceted navigation, session-ID parameters, and paginated archives nobody searches for, your money pages get crawled less often and updates take longer to register.
The controls here are unglamorous and they matter more than almost anything else on this page. A robots.txt file that blocks the right directories and none of the wrong ones. An xml sitemap that lists canonical, indexable, 200-status URLs and nothing else, because a sitemap full of redirects and 404s teaches Google to trust it less. Clean url structures that stay readable and stable. Redirects that resolve in one hop instead of chaining through four.
Google publishes the baseline itself in its Search Essentials technical requirements, and it is shorter than most people expect: Googlebot is not blocked, the page returns a successful status code, and the content is indexable. Sites fail those three more often than they fail anything sophisticated.
Diagnosis runs through Google Search Console first, because it reports what Google actually did rather than what a third-party crawler predicts. The Page Indexing report is where indexing issues surface as categories instead of guesses: discovered but not crawled, crawled but not indexed, duplicate without user-selected canonical. Each of those has a different cause and a different fix. I pair that with a full crawl to catch broken links, redirect chains, and orphan pages, then reconcile the two lists. Where they disagree is usually where the real problem is hiding.
On the Plumb Smart engagement, the previous vendor had left thousands of technical errors impairing crawlability. Clearing those crawl errors came before a single new page or link. Top-3 rankings went from 3 keywords to 19.
Most duplicate content is accidental. It comes from filters, sort orders, tracking parameters, print views, trailing slashes, uppercase variants, and a staging subdomain someone forgot to password-protect. The page is not plagiarised. It is simply reachable at nine addresses, and Google has to pick one.
Canonicalization is how you make that choice instead of letting the algorithm make it for you. Canonical tags are a hint rather than a directive, which is the part teams get wrong: if your canonical tag points one way while your internal linking, xml sitemap, and redirects all point another, Google follows the majority of the signals and ignores the tag. Consistency across every signal is what makes canonicalization stick.
The practical fix list is short. Pick one url structure and enforce it at the server. Redirect the variants with a single permanent hop. Make sure paginated and filtered views either self-canonicalize or get blocked, depending on whether anyone searches for them. Keep metadata unique on anything you actually want indexed, because two hundred pages sharing one title tag is a duplicate content signal even when the body copy differs.
Getting this right also protects the work you have already paid for. Every consolidated duplicate returns crawl budget to pages that convert, and every resolved canonical conflict stops splitting link equity across addresses that should have been one.
Core Web Vitals are Google's attempt to measure user experience with numbers a machine can read. There are three, and the thresholds are public: Largest Contentful Paint at 2.5 seconds or less, Interaction to Next Paint at 200 milliseconds or less, and Cumulative Layout Shift at 0.1 or less, each assessed at the 75th percentile of real page loads. INP replaced First Input Delay as the responsiveness metric in 2024, which caught a lot of sites that had optimized for the old measure.
Two things are worth being honest about. First, Google has said page experience is one of many signals and will not outrank genuinely better content. Second, site speed still moves revenue whether or not it moves rankings, because slow page load times raise bounce rates and depress conversion on exactly the traffic you worked hardest to earn. That is why page speed optimization belongs in the same conversation as funnel analysis and CRO rather than in a purely technical silo.
Field data beats lab data every time. A perfect Lighthouse score on a developer laptop tells you almost nothing about a customer on a four-year-old Android phone on a bad connection. I work from Chrome User Experience Report data in Search Console, find the templates failing at the 75th percentile, and fix templates rather than individual URLs. One template fix usually clears thousands of pages at once.
At CBT Baltimore, failing Core Web Vitals, broken links, and canonical problems were resolved before the content program scaled. Organic MQLs grew 271% as of October 2024, and 900% since the engagement began in January 2022.
Javascript seo is where modern sites quietly lose. A React or Vue front end can look perfect to a human and return an almost empty document to a crawler, because the content arrives only after the browser executes the bundle. Google does render JavaScript, but rendering happens in a second pass with its own queue, and anything that fails, times out, or depends on a user interaction may never make it into the index.
Google's own JavaScript SEO basics documentation is the reference here, and the practical tests are simple. Fetch the raw HTML and check whether your headings, body copy, and internal links exist before javascript rendering. Use the URL Inspection tool to see the rendered DOM Google actually produced. Confirm that links are real anchor elements with href attributes rather than click handlers, because a crawler cannot click.
Where content matters for search, server-side rendering or static generation removes the risk entirely. Where a rewrite is not on the table, the fallback is making sure the critical content, canonical tags, and metadata are present in the initial HTML response even if the interactive layer hydrates later.
Site architecture is the part of technical seo that behaves most like strategy. The way pages nest, link, and cluster tells search engines which pages you consider important and what each one is about. Get it right and authority flows from your strongest pages into the ones you need to rank. Get it wrong and your best content sits four clicks deep behind a filter.
The working rules are boring and effective. Keep important pages within three clicks of the home page. Give every indexable page at least one internal link from a page that is itself crawled regularly. Use descriptive anchor text that matches how people actually search, because anchor text is one of the clearest relevance signals you fully control. Fix broken links rather than letting them accumulate, since each one wastes crawl budget and strands whatever it pointed at.
Site structure and on-page seo are not separable in practice. A clean hierarchy makes on-page optimization simpler, because each template has one job and one keyword theme instead of competing with three siblings. That is also what makes topical clusters work: the architecture does half the job the content would otherwise have to do alone.
This is the layer where technical work and full-stack SEO consulting meet. Keyword research decides which clusters deserve to exist. Architecture decides whether search engines can tell.
Structured data is how you state facts about a page in a format machines parse without inference. Schema markup does not directly raise rankings. What it does is make your page eligible for richer treatment in results, and increasingly it gives AI systems a clean set of claims to quote rather than a paragraph to interpret.
Useful structured data implementation is narrower than most vendors suggest. Organization and Person markup so your entity resolves consistently. Article and FAQPage where the content genuinely matches. Product, Review, and Breadcrumb where they apply. LocalBusiness where a physical location matters, which overlaps with local SEO work. Marking up things that are not on the page is a manual action waiting to happen.
The failure I see most often is markup that exists but never renders. A plugin stores the JSON-LD in a database field the theme does not print, validation passes on a copied snippet, and nobody checks the live page. Every schema deployment I run ends with fetching the production URL and confirming the markup is in the source, not in the CMS.
Some technical problems only appear at scale or during change. A website migration is the highest-risk event in search, because url structures, templates, and internal linking all change at once and every mistake compounds. The discipline there is its own practice, covered on the website migration SEO page, and it is the single most common reason a company calls a consultant after the damage is done.
International seo brings hreflang, which is the most error-prone tag in the specification. Return links must be reciprocal, language and region codes have to be valid, every variant must reference itself, and the whole set has to agree with your canonical tags. One missing return link can invalidate a cluster silently.
Large catalogs bring crawl budget arithmetic. When a site has a million URLs and Google crawls forty thousand a day, the question stops being whether a page is crawlable and becomes how often it gets recrawled. Log-file analysis answers that with evidence: which templates Googlebot actually spends its requests on, versus which ones you assumed it did. That work drives most of the e-commerce SEO and SaaS SEO engagements I take.
AI search visibility runs on the same plumbing. ChatGPT, Perplexity, and AI Overviews cite pages their crawlers can fetch, parse, and attribute. A page that is slow, blocked, rendered entirely in JavaScript, or duplicated across six URLs is as invisible to an answer engine as it is to a classic crawler, and often more so, because these systems retrieve fewer sources per query and have less patience for ambiguity.
Two things change at the margin. Retrieval favors clearly bounded, self-contained passages, so headings that state the question and paragraphs that answer it completely outperform prose that only makes sense in sequence. And attribution favors entities a machine can resolve, which puts structured data and consistent naming on the critical path rather than the nice-to-have list.
The strategic layer of that work sits with a GEO consultant. The technical layer is the same audit described on this page, run with an extra question added: can a retrieval system get a clean, quotable answer out of this page without executing your front end?
Technical seo audits are only worth what gets shipped from them, so mine are scoped to produce a work queue rather than a document. Most technical seo audits fail for the same reason: they catalogue problems thoroughly and rank them not at all, so the register lands in a shared drive and nothing changes. The sequence below is consistent across engagements.
Search Console, analytics, and rank data first, so every later finding can be sized in sessions and revenue instead of severity labels. Which templates earn money decides which problems are urgent.
A full crawl plus rendered-DOM comparison plus server logs where volume justifies it. Three views of the same site, because each one catches what the others miss.
What you published, versus what is indexed, versus what earns impressions. The gaps between those three sets are where the recoverable traffic is.
Every finding scored on effort against impact, ordered so the top of the list is what a developer should start Monday. No 90-page PDF that never gets executed.
Each fix written with the affected URLs, the expected before and after, and the acceptance test. Your team ships without needing to interpret SEO vocabulary.
Fixes verified against the live site, then tracked in a dashboard you own so the effect of each change is visible. Built alongside the GA4 and Looker Studio work where it applies.
Technical seo services that end at the register are half a service. I stay through implementation, either alongside your developers or hands-on where access allows, and the audit is not closed until the fixes are verified on the live site.
All three models work. They fail in different ways, and the honest comparison is worth more than a pitch. Most companies weighing this are really choosing between a technical seo agency retainer and one senior practitioner, so start with what the site actually needs rather than with what is easiest to buy.
A technical seo agency gives you bench depth, tooling, and coverage when someone leaves. The tradeoff is that the senior technical seo experts who won the account are rarely the people in your tickets six months later, and technical work is exactly where that substitution costs the most. If you go this route, ask who runs the crawl, ask to meet them, and put their name in the contract. That question separates a strong technical seo company from a resold audit template faster than any case study will.
An in-house hire gives you context no outsider accumulates and availability no retainer buys. The catch is that a single specialist has no one to check their work, and technical seo rewards a second opinion. This model works best once there is enough recurring engineering work to keep the role busy.
An independent consultant gives you the senior person directly, with no account layer and no incentive to expand scope. The limit is capacity. I take a small number of engagements, which is the point, and it means I am the wrong answer for a company that needs five workstreams running at once. Where the need is leadership rather than execution, fractional SEO leadership is usually the better structure.
Whichever you choose, evaluate on the same three things: can they show you a prioritized register from a real engagement, can they write a fix a developer would accept without rewriting, and will they stay to verify it shipped. A good consultant and a good technical seo agency will both pass that test. Anyone who cannot is selling a crawl export.
Technical work is easy to do and easy to hide behind. The defense is deciding what should move before touching anything.
Leading indicators come first, because they respond within weeks: pages crawled per day, average response time in Search Console, valid indexed pages against submitted, Core Web Vitals passing at the 75th percentile, and the count of open crawl errors trending down. If those do not move, nothing downstream will.
Lagging indicators follow: impressions on the templates you fixed, clicks, ranking distribution, and eventually revenue. Isolate the effect by template rather than sitewide, because a sitewide chart during an active seo strategy blends technical fixes with content marketing, link building, digital pr, and whatever algorithm updates landed that month. Template-level analysis is how you tell them apart.
Two habits keep the reporting honest. Annotate every deploy so a traffic change can be traced to a cause instead of a theory. And report technical wins in the same dashboard as content, cro, and paid media, so the business sees one number and not four teams claiming the same lift.
None of this makes technical work the whole seo strategy. It makes the rest of the strategy pay. Link building returns more when the pages earning those links are indexed and fast. Digital pr returns more when the entity being cited resolves cleanly. Content and paid media both convert better on templates that load inside the Core Web Vitals thresholds. The technical layer is a multiplier on spend you are already committing.
Interested in what that looks like on your site? Get in touch and I will tell you what I would look at first, before you commit to anything.
A prioritized issues register scored on effort against impact, annotated crawl data, rendered-DOM comparisons, Core Web Vitals diagnostics from field data, schema validation, and a dev-ready implementation brief for each fix, plus a walkthrough call. The register is ordered so your team can start at the top and work down.
Yes, either alongside your dev team or hands-on where access allows. I never audit and vanish. Technical seo services that stop at the document leave the value on the table, because the register is worth nothing until the tickets close and the fixes are verified on the live site.
Because content only performs if search engines can crawl and index it, choose the right version of it, and load it fast enough that visitors stay. Technical seo is the layer that makes everything else in search engine optimization possible. Good content behind a blocked directory, a broken canonical, or a JavaScript payload Googlebot times out on earns nothing.
Crawl and log-file analysis, comfort reading rendered HTML and server responses, working knowledge of how modern front ends build pages, structured data, and enough engineering fluency to write a fix a developer will accept without rewriting it. Just as important is the judgment to rank 4,000 crawl findings down to the handful that affect revenue this quarter.
It depends on site size, platform complexity, and whether the engagement ends at the audit or continues through implementation. A fixed-scope audit on a small marketing site and a log-file-driven review of a million-URL catalog are different projects. I scope and quote per engagement rather than publishing a rate card, and the quote states exactly what is included before you commit.
A technical seo agency is a team that sells crawl, index, performance, and architecture work as a service, usually with account management layered on top. Evaluating one comes down to three questions: who personally runs the crawl and writes the fixes, can they show a prioritized register from a real engagement, and will they verify the work shipped. Ask to meet the practitioner, not the pitch team, and get that name into the contract.
Both models work. A technical seo agency gives you bench depth and coverage; an independent consultant gives you the senior person directly with no substitution risk. Where you need several workstreams running at once, an agency wins. Where you need one experienced set of eyes on a site that has never had one, technical seo experts working directly with your developers usually move faster.
By removing the constraints before adding effort. That usually means clearing crawl errors so budget reaches pages that convert, resolving canonicalization so link equity stops splitting, fixing the templates failing Core Web Vitals rather than individual URLs, and making sure content survives javascript rendering. Those changes compound, because every later investment in content or links lands on a site that can actually use it.
Crawl and index signals move within days to a few weeks: pages crawled per day, valid indexed pages, and open crawl errors respond quickly. Ranking and traffic effects follow over one to three months as Google recrawls and reassesses the affected templates. Migration recovery and large-catalog crawl budget changes sit at the longer end of that range.
Free 30-minute discovery call. Bring your goals and pain points — I'll bring the plan.
Request a Consultation See the Numbers