Technical SEO audit work is the least glamorous line item in a B2B marketing plan and the one most likely to be the reason nothing else is working. Content calendars get approved. Link campaigns get budget. Meanwhile the site quietly serves a 200 status code on pages that should not exist, blocks the crawler from the pages that should, and takes six seconds to render a product page on a mid-range laptop over office Wi-Fi. None of that shows up in a content report. All of it shows up in the rankings.
B2B sites earn this problem honestly. They accumulate a decade of campaign landing pages, a documentation subdomain owned by engineering, a resource library behind a gate, a careers section on a third-party applicant tracking system, and three generations of URL conventions from three replatforms. Each was a reasonable decision. Together they produce a site that search engine optimization practitioners describe, politely, as complicated. A technical SEO audit is the process of measuring exactly how complicated, and deciding what to do about it in an order that a development team will actually accept.
This technical SEO audit checklist is written for that job specifically. It assumes low search volumes and high deal values, which changes the arithmetic on almost every recommendation. It assumes you share a codebase with a product team whose sprint capacity you do not control. And it assumes you will have to justify each finding to somebody who wants to know what it is worth before they schedule it.
If you are costing this work, our guides to SEO cost in the UK and SEO budget planning cover the money, and our SEO pricing models guide covers how the contract is structured. If a rebuild is on the horizon, start with the website redesign SEO migration checklist instead — the technical SEO audit you need before a migration is a different document.
Table of contents
- What a Technical SEO Audit Actually Finds on a B2B Website
- Scope the Technical SEO Audit Before You Crawl Anything
- Crawlability: Can Search Engines Reach the Pages That Matter?
- Indexation: Is the Right Set of Pages in the Index?
- Architecture and Internal Linking for Long B2B Buying Cycles
- Core Web Vitals and Page Experience on B2B Templates
- Rendering, JavaScript and the Gated-Content Problem
- Structured Data and Readiness for AI Answer Engines
- International, Subdomain and Multi-Site Complexity
- Prioritising Findings: Turning the Technical SEO Audit Into a Roadmap
- Cadence, Tooling and What a Technical SEO Audit Costs
- Frequently Asked Questions About a Technical SEO Audit
- References
What a Technical SEO Audit Actually Finds on a B2B Website
The phrase covers a wide range of work, from a forty-pound automated crawl report to a four-week engagement with log-file analysis. It helps to be precise about what the exercise is for before arguing about tooling.
It is a diagnosis, not a list of errors
A crawler will hand you nine hundred issues. Most of them do not matter. A technical SEO audit is the judgement layer on top: which of those nine hundred findings are costing you qualified traffic today, which will cost you traffic when the site grows, and which are cosmetic. Any deliverable that arrives as an unfiltered software export has skipped the only part that required a human, and you should treat it as raw material rather than as the technical SEO audit itself.
B2B search volume changes what counts as severe
An ecommerce site with a thousand transactions a day can measure a technical fix in a week. A B2B site with eleven enquiries a month cannot. That statistical thinness means the technical SEO audit has to reason from mechanism rather than from observed uplift — you fix a crawl trap because crawl traps demonstrably suppress discovery, not because you can prove the enquiry count moved. It also means a single high-intent page dropping two positions can matter more than a thousand blog URLs being slow.
The site is usually several sites wearing a trench coat
The marketing site runs on one platform. Documentation runs on another, often on a subdomain. The customer portal sits behind a login. The events microsite was built by an agency in 2021 and nobody has the credentials. A serious technical SEO audit inventories all of them, because search engines do not respect your org chart and will happily index the staging environment that engineering forgot had a public IP.
What it should hand back
A finished technical SEO audit produces four artefacts: a prioritised issue register with severity and effort against each row, a set of reproducible tickets a developer can act on without further interpretation, a measurement baseline so the next technical SEO audit can show movement, and a short executive summary explaining the commercial argument. If any of the four is missing, the work will stall at the point where it needs somebody else’s time.
Scope the Technical SEO Audit Before You Crawl Anything
Scoping is where most technical SEO audits are won or lost. Crawl first and you will spend three days processing noise from a subdomain that was never in scope.
Inventory every property, including the ones you forgot
List every hostname the business controls: the main site, www and apex variants, documentation, help centre, status page, blog if it is separately hosted, regional domains, the careers portal, campaign microsites, and any legacy domain still pointed at your servers. Check the DNS records rather than asking around, because the answer you get in a meeting will be incomplete. Each hostname either enters the technical SEO audit scope or gets an explicit note saying why it does not.
Decide the URL universe and freeze it
Agree what counts as a page for the purposes of this technical SEO audit. Do parameterised filter combinations count? Do paginated archives? Do PDF datasheets, which on many B2B sites are the single most-linked asset type? Write the decision down, then export a frozen snapshot of the crawl. The live site will change during the technical SEO audit, and a moving baseline makes every later reconciliation an argument.
Connect the data sources before you start
The crawl is one input of five. You also want twelve months of Search Console performance and page indexing data, twelve months of analytics landing-page data, at least four weeks of raw server logs, and the CMS’s own list of published content. A technical SEO audit built on a crawl alone cannot see what search engines actually requested, which is precisely the evidence that turns a theory into a finding.
Set the questions the technical SEO audit must answer
Scope creep in this work is infinite unless you bound it with questions. Good ones for a B2B site: can search engines reach and index every page we want ranked, are we wasting discovery effort on pages we do not want ranked, is the site fast enough on the devices our buyers actually use, does our content render without JavaScript execution, and is our entity and schema markup good enough to be cited by AI answer engines. Five questions is enough.
| Audit factor | B2B site | Consumer or ecommerce site |
|---|---|---|
| Typical URL count | 2,000 – 15,000 | 50,000 – millions |
| Crawl budget pressure | Rarely the binding constraint | Often the primary constraint |
| Value per ranking page | Very high, very few pages | Low, spread across many |
| Measurable feedback loop | Weak — low conversion volume | Strong — daily transaction data |
| Gated and login-walled content | Extensive, often the best content | Minimal |
| Subdomain sprawl | Common — docs, support, careers | Less common |
| Development capacity for fixes | Shared with product roadmap | Often a dedicated web team |
| Dominant failure mode | Content invisible to crawlers | Duplicate and thin pages at scale |
Crawlability: Can Search Engines Reach the Pages That Matter?
Everything downstream in a technical SEO audit depends on this. A perfect page that cannot be reached is worth nothing, and B2B sites block their best content more often than they realise.
Read robots.txt line by line, not at a glance
Open the file and read every directive rather than trusting a tool’s summary. The classic B2B failures are a Disallow: /resources/ left over from when the resource library was in development, a blanket disallow on /*? that kills legitimate filtered pages, and a staging rule copied into production. The Robots Exclusion Protocol is now a formal standard, and its matching rules are more specific than most teams assume — a longer matching rule wins regardless of order.
Find the crawl traps before they find your budget
Faceted navigation, calendar widgets, session identifiers in URLs, and infinite scroll implementations all generate unbounded URL spaces. On a B2B site the usual culprit is a resource library with four filter dimensions, which multiplies into tens of thousands of near-identical pages. Sort your crawl by URL pattern and look for any pattern with an implausible count. Google’s own guidance on managing crawl budget is written for large sites, but the diagnostic logic applies at any size.
Check the internal link graph, not just the sitemap
A page listed in the XML sitemap but linked from nowhere is an orphan, and orphans are the single most common finding in a B2B technical SEO audit. They happen because campaign pages are built to be reached from an email, and nobody ever adds them to the navigation. Cross-reference the crawl against the sitemap and against the CMS export: URLs in the sitemap but absent from the crawl are orphaned, and the gap is usually larger than anyone expects.
Confirm the sitemap is telling the truth
The XML sitemap should list canonical, indexable, 200-status URLs and nothing else. Most generated sitemaps on older B2B platforms include redirects, noindexed pages and 404s, which teaches search engines to trust the file less. Validate it against the Sitemaps protocol, check that lastmod reflects genuine content changes rather than a nightly rebuild timestamp, and split files that exceed the size limits.
Indexation: Is the Right Set of Pages in the Index?
Reachable is not the same as indexed. The second pass of any technical SEO audit compares what you want indexed against what actually is.
Reconcile three counts, not one
Take the count of indexable URLs from your crawl, the indexed count from the Search Console page indexing report, and the count of URLs earning at least one impression in the last twelve months. Three numbers that should be close and never are. A large gap between crawl and indexed means quality or duplication problems. A large gap between indexed and earning impressions means you have pages the search engine has stored but sees no reason to show.
Work the exclusion reasons individually
The page indexing report groups excluded URLs by reason, and each reason implies a different fix. “Discovered — currently not indexed” points at quality or crawl scheduling. “Crawled — currently not indexed” is usually a thin or duplicate content signal. “Alternate page with proper canonical tag” is normally correct and can be dismissed. “Duplicate without user-selected canonical” means the search engine disagreed with you and picked its own. Sample ten URLs from each bucket rather than reading totals.
Audit canonicals against reality
Canonical tags on B2B sites break in predictable ways: a template hard-codes the homepage as canonical across a whole section, a parameterised URL self-canonicalises when it should point at the clean version, or the canonical points at an HTTP URL after an HTTPS migration. Check that every canonical resolves to a 200-status, indexable, self-canonicalising page. Google’s guidance on consolidating duplicate URLs sets out how the signals interact.
Catch the accidental noindex
Every technical SEO audit should include a direct check of the robots meta tag and the X-Robots-Tag HTTP header across the whole crawl, because a noindex directive delivered in a header is invisible in the page source and routinely missed. On B2B sites it appears when a template used for gated assets is reused for an ungated one, or when a CDN rule intended for staging leaks into production.
| Symptom | Likely cause | How to confirm | Severity |
|---|---|---|---|
| Key page absent from index | Noindex header or robots block | Fetch headers, run URL inspection | Critical |
| Indexed count far exceeds crawl | Parameter or filter explosion | Group indexed URLs by pattern | High |
| Wrong URL ranking for a term | Canonical conflict or duplication | Compare canonicals across the set | High |
| Indexed but zero impressions | Thin content or no demand | Check query data and word count | Medium |
| Pages drop in and out of index | Quality threshold or instability | Track weekly index status | Medium |
| Staging environment indexed | Missing auth on non-production | Site query on the staging host | Critical |
Architecture and Internal Linking for Long B2B Buying Cycles
Structure is where a technical SEO audit stops being a compliance exercise and starts influencing revenue, because B2B architecture has to serve a buying committee rather than a shopper.
Measure click depth from the homepage
Export click depth for every indexable URL. On a well-organised B2B site, every commercially significant page sits within three clicks of the homepage. Pages at depth five or more are receiving almost no internal authority and are usually the ones that pagination buried. If your highest-value solution pages sit deeper than your blog posts, the navigation is arguing against your commercial priorities.
Check that the money pages receive internal links
Count inbound internal links per URL and sort ascending. It is remarkably common to find that the pricing page, the integration pages and the industry solution pages each have three or four internal links while a two-year-old blog post has ninety. Internal linking is the one ranking factor entirely under your control, and correcting this imbalance is usually the cheapest high-impact recommendation a technical SEO audit produces.
Map the hub-and-spoke structure honestly
B2B content tends to accumulate as isolated posts rather than as connected topical clusters. Draw the actual link graph for one topic and compare it to what you intended. Most sites discover their hub page links out to spokes but the spokes never link back, or that three pages compete for the same query with no clear hub at all. Consolidation is often a better finding than creation.
Review the navigation for crawler-visible links
Mega-menus built entirely in JavaScript, links rendered only after a click event, and “load more” buttons that never expose a paginated URL all hide your architecture from crawlers. Test by disabling JavaScript and viewing the rendered source. Any link that disappears is a link that may not be passing signals, which is a structural finding rather than a cosmetic one.
Core Web Vitals and Page Experience on B2B Templates
Speed matters less for B2B rankings than vendors claim and more for conversion than marketers admit. A technical SEO audit should measure it for both reasons.
Use field data first, lab data second
Lab tools give you a reproducible score in controlled conditions. Field data tells you what your actual visitors experienced. Start with the field metrics in Search Console and the Chrome UX Report, identify which URL groups fail, and only then use a lab tool to diagnose why. Auditing in the reverse order produces a long list of theoretical improvements to pages nobody visits. The Core Web Vitals definitions are the reference point for both.
Break the largest contentful paint into its parts
A failing LCP is never one problem. Split it into server response time, resource load delay, resource load duration and render delay, then attack the largest component. On B2B sites the usual offender is a hero image served at full desktop resolution to mobile devices, closely followed by a render-blocking web font requested from a third-party host.
Audit third-party tags as their own workstream
Marketing sites accumulate tags: analytics, two chat widgets, a heatmap tool, an account-based marketing pixel, a review widget and a consent manager that loads before all of them. Each was added by someone with a good reason. Measure their combined cost with the tag manager disabled versus enabled — on a typical B2B site this single comparison produces the largest performance number in the whole technical SEO audit.
Check interaction responsiveness on real devices
Interaction to Next Paint punishes heavy JavaScript on interaction, which on B2B sites means form fields, filter controls and configurator widgets — precisely the elements that sit between a visitor and an enquiry. Test on a mid-range Android device on a throttled connection rather than on the marketing team’s laptops.
Rendering, JavaScript and the Gated-Content Problem
This section is where B2B sites diverge most sharply from consumer sites, and where a generic technical SEO audit template will let you down.
Compare raw source against rendered DOM
Fetch the page without executing JavaScript, then fetch the rendered version, and diff the two. Anything present only in the rendered version depends on successful script execution to be seen. That is not automatically fatal, but it introduces a dependency, and JavaScript SEO failures are silent — the page looks fine to you and is empty to the crawler.
Decide what gating actually costs
Gating a whitepaper behind a form is a legitimate commercial decision. Gating it in a way that leaves the search engine with a form and nothing else means the asset can never rank, never earn a link, and never be cited by an answer engine. The technical SEO audit finding is not “remove the gate” — it is to render a substantial, indexable summary above the form so the page has something to rank with.
Audit the crawl path through login walls
Customer portals, community forums and support ticketing systems frequently sit behind authentication while still exposing thousands of crawlable URLs that all return a login page. Search engines see a large volume of near-identical thin pages. The fix is a robots disallow plus authentication, not a noindex tag on a page that still consumes discovery effort.
Check documentation subdomains separately
Technical documentation is often the highest-quality content a B2B company owns and is frequently on a separate stack with its own defaults. Run the full crawl, indexation and rendering checks against it as a discrete property. Static site generators used for docs commonly ship without canonical tags, without a sitemap, and with client-side routing that hides every internal link.
Structured Data and Readiness for AI Answer Engines
Markup used to be about rich snippets. It is now also about being legible to systems that summarise rather than link, which raises the stakes for B2B brands whose buyers research before they ever visit.
Validate what you already have
Run every template type through a validator and fix errors before adding anything new. Broken markup is worse than absent markup because it can suppress eligibility across a whole page type. Pay particular attention to Organization, Product, FAQ and Article types, and check them against the structured data feature gallery rather than against what a plugin generates by default.
Make the entity unambiguous
For a B2B brand the single highest-value piece of markup is a complete Organization entity on a stable URL: legal name, alternate names, logo, registered address, sameAs links to the company’s authoritative profiles, and a clear description. This is what lets a search engine, and increasingly a language model, connect scattered mentions to one company. Schema.org is the vocabulary; consistency is the actual work.
Structure the pages that answer questions
Content that answers a discrete question with a discrete answer — pricing explanations, comparison pages, glossary entries, implementation timelines — is what answer engines quote. A technical SEO audit should flag pages where the answer exists but is buried in the ninth paragraph, because retrievability and readability are now the same problem. Our SEO services team treats this as a standing workstream rather than a one-off fix.
Confirm your content is reachable without a crawl budget argument
Answer engines and traditional crawlers reach content through the same doors. Every crawlability finding earlier in this checklist is therefore also an AI-visibility finding, which is a useful argument when you need to justify unglamorous infrastructure work to a board that has recently become very interested in AI.
International, Subdomain and Multi-Site Complexity
Any B2B company selling across borders inherits a category of technical SEO audit findings that single-market businesses never see.
Audit hreflang as a bidirectional graph
Hreflang annotations must reciprocate: if the UK page points at the German page, the German page must point back, and both must include a self-reference. Missing return tags are the most common implementation error and cause the entire cluster to be ignored. Check the implementation against Google’s guidance on localised versions, and verify the language and region codes are valid rather than plausible.
Resolve regional duplication honestly
English-language sites for the UK, US and Australia are frequently ninety-five per cent identical. That is acceptable with correct hreflang and is a liability without it. The technical SEO audit finding is to decide, per page type, whether the regional variant carries genuinely different information — pricing, currency, compliance statements, case studies — or whether it should be consolidated to one canonical page.
Decide the subdomain question with evidence
Whether documentation belongs on a subdomain or a subfolder is an old argument. Rather than relitigating it, measure what you have: compare indexation rates, impression growth and internal link flow between the two properties. A technical SEO audit can only recommend a migration if the evidence justifies the cost, and migrating a documentation subdomain is a large project with real risk — plan it with a proper redirect map.
Prioritising Findings: Turning the Technical SEO Audit Into a Roadmap
A technical SEO audit that ends with a list has failed. The deliverable is a sequence somebody will actually work through.
Score every finding on impact and effort
Give each issue a traffic-impact estimate and a development-effort estimate, both on a simple three-point scale, and plot them. High impact with low effort goes first and gets done this month. High impact with high effort becomes a scheduled project with a business case. Low impact with low effort gets batched into a housekeeping ticket. Low impact with high effort is documented and deliberately not done.
Write tickets, not recommendations
“Improve internal linking” is not actionable. “Add a contextual link block to the solution page template linking to the three most relevant case studies, ordered by industry tag” is. The single biggest determinant of whether a technical SEO audit produces results is whether its findings arrive in a form a developer can estimate without a meeting.
Separate template fixes from content fixes
Template-level findings fix hundreds of URLs with one deployment and belong in the development backlog. Content-level findings fix one URL each and belong to the marketing team. Mixing them in a single list guarantees that the cheap wins wait behind the expensive ones. Sorting by owner is more useful than sorting by severity.
Agree the measurement baseline before work starts
Record the numbers you will report against: indexed page count, impressions by page group, the proportion of URLs passing Core Web Vitals, average click depth for commercial pages, and organic enquiries. Without a frozen baseline the next technical SEO audit cannot demonstrate progress, and unmeasured work is the first thing cut when budgets tighten.
| Finding | Traffic impact | Dev effort | Priority band |
|---|---|---|---|
| Noindex on commercial templates | High | Low | Immediate |
| Robots block on resource library | High | Low | Immediate |
| Internal links to money pages | High | Medium | This quarter |
| Faceted navigation crawl trap | Medium | High | Scheduled project |
| Client-side rendered navigation | High | High | Scheduled project |
| Organization schema completion | Medium | Low | This quarter |
| Hero image responsive sizing | Medium | Low | This quarter |
| Alt text gaps on archive pages | Low | Low | Housekeeping |
Cadence, Tooling and What a Technical SEO Audit Costs
The last section of the checklist is the operational one: how often, with what, and for how much.
Set a cadence rather than waiting for a crisis
A full technical SEO audit once a year, a lighter quarterly review of the same metrics, and continuous automated monitoring is a proportionate rhythm for most B2B sites. The exception is any quarter containing a replatform, a redesign, a domain change or a large content migration — each of those earns its own technical SEO audit regardless of when the last one ran.
Monitor continuously between technical SEO audits
Set alerts on the things that break silently: a change to robots.txt, a spike in 404s or 5xx responses, a drop in indexed page count, a Core Web Vitals regression, and any change to the noindex directive on a commercial template. Most of the catastrophic findings in a technical SEO audit are things that could have been caught within hours by an alert rather than within months by an annual review.
Understand what each tool can and cannot see
No single tool does the whole job. A desktop crawler sees structure but not behaviour. Search Console sees the search engine’s opinion but samples heavily. Log files see reality but require processing. A lab tool such as Lighthouse sees one page under synthetic conditions. Combining them is not optional if you want findings rather than exports.
Budget realistically for the work
A technical SEO audit of a small site costs £500 to £1,500 and takes one to two weeks. A mid-size site runs £1,500 to £3,500 over two to four weeks, and genuine enterprise scope reaches £3,500 to £10,000 or more across four to eight weeks. Below roughly £300 you are buying software output with a logo on it. An independent technical SEO audit before you commit to a retainer costs £500 to £3,500 and tells you the real state of your site — the same argument we make in our marketing services engagements, and the reason it is worth doing before you sign anything ongoing.
| Data source | What it catches | What it misses | Typical cost |
|---|---|---|---|
| Desktop crawler | Structure, status codes, directives | Orphans, real crawler behaviour | £150 – £250 per year |
| Search Console | Index status, queries, field vitals | Full URL lists, unsampled data | Free |
| Server log files | Actual crawler requests and frequency | Anything never requested | Storage and analyst time |
| Analytics platform | Pages that earn sessions and enquiries | Pages with zero traffic | Free to £1,000+ per month |
| Lab performance tool | Diagnosable render-path problems | Real user conditions | Free |
| Manual review | Judgement, context, false positives | Scale | Consultant day rate |
Frequently Asked Questions About a Technical SEO Audit
How long does a technical SEO audit take?
One to two weeks for a small B2B site, two to four weeks for a mid-size one, and four to eight weeks at enterprise scale. The crawl itself takes hours; the time goes into reconciling data sources, sampling URLs manually and turning findings into tickets somebody can estimate.
Can we run one ourselves instead of hiring someone?
Partly. The data collection side of a technical SEO audit is well within an in-house team’s reach and the tools are cheap. What is hard to do internally is the prioritisation, because it needs pattern recognition across many sites to know which of nine hundred flagged issues actually matter. A hybrid works well: collect internally, buy the judgement.
How often should a B2B site be audited?
Annually as a full exercise, quarterly as a lighter review, plus an unscheduled technical SEO audit after any replatform, redesign, domain change or major migration. Continuous alerting between technical SEO audits catches the failures that would otherwise sit undetected for months.
What is the most common technical SEO audit finding on B2B sites?
Orphaned and under-linked commercial pages. Campaign landing pages built to be reached from an email, solution pages buried five clicks deep, and case studies with no internal links pointing at them. It is also the cheapest category to fix, which is why it usually leads the roadmap.
Does a technical SEO audit help with AI search visibility?
Yes, and more directly than most people expect. Answer engines reach your content through the same crawling and rendering path as traditional search. Content that a crawler cannot reach or render cannot be summarised or cited, so crawlability and structured data findings are AI-visibility findings by another name.
References
Google Search Central: In-Depth Guide to How Google Search Works
Google Search Central: Introduction to robots.txt
Google Search Central: Large Site Owner’s Guide to Managing Crawl Budget
Google Search Central: Consolidate Duplicate URLs with Canonicals
Google Search Central: Understand JavaScript SEO Basics
Google Search Central: Localized Versions of Your Pages
Google Search Central: Structured Data Markup Feature Gallery
RFC 9309: Robots Exclusion Protocol
Sitemaps XML Format and Protocol
