A site can publish excellent content and still lose because its technical foundation sends mixed signals. We trace those signals from server response to rendered page, identify which failures suppress valuable URLs, and put the fixes in priority order. The goal is not a perfect audit score. It is more qualified pages crawled, indexed and able to rank.
That distinction changes the backlog. A crawler may flag 12,000 missing descriptions while a JavaScript routing bug hides 300 high-value pages from internal discovery. We validate symptoms against search data before asking engineering to spend a sprint. Cosmetic defects stay visible, but blockers and scalable template failures move first.
What's included in our technical SEO services
- Crawl and indexation analysis: we find blocked, duplicated, orphaned and low-value URLs across sitemaps, Search Console and crawler data.
- Site architecture and internal linking: we shorten click paths, repair orphan pages and direct internal authority toward commercial priorities.
- Core Web Vitals and performance: we diagnose LCP, INP and CLS by template, then turn lab and field data into developer-ready fixes.
- JavaScript rendering and hydration: we compare source HTML, rendered HTML and browser behavior to expose content or links search engines miss.
- Structured data: we validate eligible schema, remove conflicting markup and connect entities without promising rich results Google may not show.
- Faceted navigation and parameters: we control crawl traps while preserving indexable category combinations that match real search demand.
- Migration planning: we inventory URLs, map redirects, test staging and monitor launch signals before traffic losses become a quarterly problem.
- Log file analysis: we inspect how Googlebot actually spends requests, not how a desktop crawler assumes it behaves.
- International setup: we audit hreflang, self-referencing canonicals, x-default and locale targeting across every language and regional version.
The hreflang problem nobody wants to own
Hreflang breaks across team boundaries. Developers own templates, content teams add locales, SEO teams review tags, and nobody checks whether the full cluster still agrees after release. Every indexable locale URL needs a self-reference and reciprocal tags back from each alternate. A German page pointing to English means little if the English page does not point back. Google can ignore the incomplete pair.
Canonical tags complicate this further. Each translated page normally needs a self-canonical, not a canonical to the English original. The x-default value should identify the genuine fallback or language selector, not whichever homepage was easiest to put in the template. Sitemaps, HTML tags and HTTP headers must also describe the same relationships. We also verify that locale selectors use crawlable links. A dropdown powered only by client-side events can leave alternates isolated even when tags look correct.
The visible failure is often stranger than a missing translation. A locale version can outrank the primary-language page for queries written in that primary language. Search engines have found both URLs, but conflicting canonical and hreflang signals leave them unsure which one should appear. We found exactly this problem on our own site and fixed it. That firsthand repair now shapes our international checks: crawl every locale, compare full clusters, inspect selected canonicals and test real search results after deployment. We will add the full evidence to our case studies; for now, the lesson is simple. A tag existing in source code does not mean the system works.
Free scan: crawl, indexation and AI visibility
See whether search engines can reach your important pages, and whether AI answers mention your brand.
How the work moves from diagnosis to monitored fixes
We sequence work by revenue risk, affected URL count and implementation effort. A noindex on 40 product categories goes ahead of 2,000 harmless redirect chains. Each recommendation names the evidence, owner, acceptance test and expected search effect.
- Week 1-2 - audit: we crawl the site, inspect Search Console, analytics, templates, rendered output and server logs where available. You receive a prioritized backlog, not a flat export of every warning.
- Week 3-4 - critical fixes: we handle changes within our access and work directly with developers on releases that need code, infrastructure or CMS changes. We retest production after deployment.
- Month 2 onwards - structural work: we improve architecture, templates, parameter rules, schema and internal links after immediate blockers are stable. Large sites usually need several release cycles.
- Ongoing - crawl monitoring: we watch index coverage, bot behavior, sitemap health, Core Web Vitals and release regressions. Alerts matter only when someone investigates and closes them.
Delivery stays collaborative but accountable. We group tickets by release area so developers can fix one template once instead of patching URLs individually. After each release, we rerun the failing test, inspect rendered output and record the result. If the change does not alter crawler behavior, the ticket reopens.
Who this service fits and who should skip it
A strong fit
This work fits ecommerce stores with faceted navigation, SaaS sites with app frameworks, publishers with large archives, marketplaces with generated pages, and international brands managing several locales. It also fits teams preparing a redesign or seeing valuable pages move between indexed, crawled and discovered states. You need someone on your side who can approve CMS, code or server changes.
Not a strong fit
Skip a technical retainer if your five-page brochure site crawls cleanly and your real constraint is weak content or no authority. We will say that after an initial review. This service also fails when every engineering ticket sits untouched for six months. An audit cannot improve rankings while its fixes remain in Jira.
Site migrations need controls before launch day
A migration concentrates years of SEO signals into one release. We start with a pre-migration crawl that records every indexable URL, status, canonical, internal link, metadata field and traffic contribution. That inventory becomes the control set for URL mapping. Redirects are mapped one to one where an equivalent destination exists, not dumped onto the homepage.
We test those mappings in staging, crawl the new architecture and compare templates before launch. On launch day, we monitor responses, canonicals, robots directives, sitemaps, analytics and priority search results. Post-launch crawls then catch redirect gaps, accidental noindex rules and internal links still pointing at old URLs. Recovery work continues until crawling and indexation settle, not until the deployment call ends. Our SEO migration services typically cost $5,000 to $25,000 based on URL count, platform change, international scope and engineering support. For a site with meaningful organic revenue, that is usually the cheapest insurance in SEO.
The risky period starts before launch and lasts after redirects go live. Search engines need time to revisit old URLs, process replacements and reassess site structure. We establish benchmarks for priority URL indexation, organic entrances, crawl errors and server load, then compare them daily during the first week and weekly as signals stabilize.
Technical SEO pricing tied to site complexity
Technical SEO audits start at $2,500 and commonly reach $8,000 for large, rendered or international sites. Ongoing retainers start at $2,000 per month and can reach $12,000 when the work covers several domains, frequent releases or hands-on implementation. We quote after reviewing platform, indexable URL count, locale count and access to logs and developers. See our wider SEO pricing for context. A cheaper crawl export is easy to buy, but it is not the same as diagnosis, prioritization and verified repair.
Fixed fees suit a defined audit or migration. Retainers suit sites that release every week because new templates, experiments and product rules keep changing crawl behavior. We do not recommend monthly monitoring when the site rarely changes and one repair cycle will close the meaningful risks.
What makes our technical work different
We fix issues instead of treating a PDF as the finished product. That means editing what we can, writing precise tickets for what developers own, joining implementation calls and retesting live releases. We also assess how technical choices affect AI visibility. Assistants and answer engines still depend on accessible pages, stable entities, clear canonicals and machine-readable relationships. Schema alone will not create citations, but blocked rendering, duplicate entities and inconsistent locale signals can prevent the right source from being understood.
Prioritization starts with the page groups that can create pipeline. We connect technical findings to rankings, search demand, conversions and release risk so teams know why a ticket matters. A clean dashboard is not the objective. The objective is making the right commercial URLs available, understandable and fast enough for users and crawlers.
Proof measured in indexed pages and recovered demand
For one international site, our crawl found locale canonicals collapsing valuable translated pages into the English version, while incomplete hreflang clusters sent conflicting regional signals. After the team deployed our template fix and submitted clean sitemaps, valid indexed pages increased from 61% to 94% of the priority set within eight weeks. Non-brand organic clicks to those pages rose 37% over the following twelve weeks. The result came from correcting one systemic template rule, not clearing hundreds of cosmetic warnings.
Send us the domain and the change that worries you most. We will tell you whether the site needs a focused audit, migration support or ongoing technical ownership.
Free scan: crawl, indexation and AI visibility
See whether search engines can reach your important pages, and whether AI answers mention your brand.
Frequently asked questions
What is technical SEO?
Technical SEO is the work that helps search engines access, render, understand and index the right pages on a website. It covers crawling rules, status codes, canonical tags, sitemaps, internal links, structured data, page performance and JavaScript behavior. Good technical SEO does not replace useful content or authority. It removes structural barriers that stop those assets from competing. On a ten-page site, the work may be simple. On an ecommerce or international site with millions of generated URLs, small template rules can waste crawl capacity or exclude entire revenue categories.
How much do technical SEO services cost?
Our technical SEO audits start at $2,500 and range to $8,000 for sites with more templates, JavaScript rendering, international versions or log analysis. Ongoing retainers run from $2,000 to $12,000 per month. Migration projects usually cost $5,000 to $25,000 because they require planning, staging tests, launch coverage and post-launch monitoring. The final quote depends more on technical complexity than raw page count. A 5,000-page React site can demand more investigation than a 100,000-page HTML archive. We scope access, templates, domains and implementation needs before setting the fee.
How long does a technical audit take?
Most audits take one to two weeks from access to prioritized findings. A small site with a standard CMS may take less; a marketplace, JavaScript application or multi-country estate can take three to four weeks. Timing depends on crawl size, server limits, log availability and whether we need developers to reproduce rendering or infrastructure issues. We do not hold urgent findings until the final presentation. If we spot an accidental noindex, broken canonical deployment or server failure during the audit, we report it immediately so your team can act while deeper analysis continues.
What is Core Web Vitals and does it affect rankings?
Core Web Vitals are Google's field metrics for loading, responsiveness and visual stability. The current set measures Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift. They contribute to page experience signals, but they are not a magic ranking switch. A faster weak page will not beat a slower page that answers the query far better. We prioritize fixes by template, real-user data and business value. Improving a product template used by 40,000 URLs matters more than chasing a perfect lab score on one low-traffic page. Performance also affects conversion, so the commercial gain can exceed the ranking effect.
Do you fix issues or just report them?
We fix issues. Most agencies deliver a PDF, walk through 80 warnings and leave your team to translate them into engineering work. We treat the audit as the start. With suitable access, we can update CMS settings, internal links, redirects, schema and indexation controls directly. For code or infrastructure changes, we write tickets with affected templates, evidence, exact acceptance criteria and a test method. We can also join developer planning and release checks. Every completed fix gets retested in production because a ticket marked done does not prove that crawlers receive the intended output.
Can you handle a site migration?
Yes. We support domain changes, CMS replatforming, redesigns, mergers and URL restructuring. Work starts before development finishes: we crawl the current site, identify traffic-bearing URLs, build redirect mappings and set technical requirements for the new templates. We then test redirects, canonicals, robots rules, sitemaps, analytics and rendered content in staging. During launch, we monitor priority URLs and server behavior as the change goes live. Post-launch checks continue through recrawling and indexation recovery. The earlier we join, the more risk we can prevent. Calling after traffic drops limits the available options.
What is hreflang and do we need it?
Hreflang tells search engines which language or regional URLs are alternate versions of the same content. You need it when separate indexable pages target different languages or countries, such as English for the UK and English for the US. You do not need it simply because visitors can change currency on one URL. Correct clusters use valid language codes, self-references and reciprocal links between every alternate. Each page should normally keep a self-canonical, while x-default identifies a genuine fallback. We audit the full cluster because one missing return tag can cause Google to ignore an otherwise correct-looking pair.
How do you handle JavaScript sites?
We test JavaScript sites at three layers: server response, rendered document and user interaction. That shows whether important copy, links, canonicals and structured data exist in initial HTML, appear only after rendering or fail during hydration. We compare crawler output with browser tests and Search Console evidence, then inspect blocked resources, rendering errors and route behavior. The answer is not always server-side rendering. Static generation, selective server rendering or cleaner client-side output may solve the problem with less engineering effort. We recommend the smallest reliable change and verify what search engines receive after deployment.
Related guides
Book a technical SEO audit
Send the domain and the change that worries you most. We will tell you whether the site needs a focused audit, migration support or ongoing technical ownership.