SEO Audit Service
Most audits are a crawler export with a cover page: hundreds of warnings at equal weight, and you left to work out which of them matter. Ours is written the other way round. The few things holding the site back come first, with the evidence for each, followed by a plain list of what to leave alone.
What is included
- A crawl of the site and a check of what crawlers receive before JavaScript runs
- Indexation reviewed against your own Search Console data
- Internal links, Core Web Vitals and structured data assessed template by template
- Content gaps, overlapping pages and a backlink review
- A comparison against your three closest genuine competitors
On this page
- What is an SEO audit service?
- What does a website SEO audit examine?
- How does an audit check what a search engine can actually see?
- How are Core Web Vitals judged in an audit?
- What do the content, backlink and competitor checks tell you?
- How does an audit change for local, ecommerce and international sites?
- What does the finished SEO audit look like?
- How does an audit decide what to fix first?
- What does an SEO audit not do?
- How do you judge an SEO audit from any supplier?
- What is the quickest test of any audit?
- What makes one SEO audit take more work than another?
- When is an SEO audit the wrong first step?
What is an SEO audit service?
An SEO audit service examines your website the way a search engine meets it, then tells you in writing what is holding it back, in what order to fix it, and what to leave alone. It covers crawling and indexation, rendering, internal links, speed, structured data, content, backlinks and your closest competitors.
That is different from running a crawler and sending the export. A tool can list every page with a missing description; it cannot tell you that one noindex tag left over from staging matters more than the other nine hundred findings put together. The value of an audit is the judgement applied to the data, and the evidence that lets you check the judgement.
It is also a diagnosis, not a treatment. An audit tells you what is wrong, and whether anything improves depends on whether the fixes are made. That is why the wider SEO service puts the audit first and the implementation straight after it, and why an audit on its own is worth buying only when somebody is ready to act on the result.
The point
An audit tells you what is wrong, and whether anything improves depends on whether the fixes are made.
What does a website SEO audit examine?
A proper audit works through eight areas in an order that matters, because the early ones decide whether the later ones are worth the trouble: crawl and indexation, rendering, internal links, Core Web Vitals, structured data, content gaps, backlinks and the competitor set. Analytics is checked alongside, since an audit you cannot measure against is only an opinion.
| Area | What the audit looks at | What a real problem looks like |
|---|---|---|
| Crawl and indexation | Which pages a search engine can reach, which it has indexed, and whether the two lists match the pages you want found. Robots.txt, noindex tags, canonicals, sitemaps, redirects and duplicate URLs, read against the Page indexing report in Search Console. | Commercial pages marked noindex since a staging site went live. A sitemap listing URLs that redirect or return errors. Several URLs for one page, each splitting the signal. |
| Rendering | What a crawler receives before any JavaScript runs, compared with what a browser shows. Key templates are fetched as raw HTML and read. | A service page that looks complete in a browser and arrives as an empty shell with no heading and no body text. It is often invisible to AI and answer-engine crawlers before it troubles Google. |
| Internal links and architecture | Click depth from the homepage, orphaned pages, anchor text, and whether your commercial pages are supported by the pages around them. | A key service page linked only from the footer. Four thin pages competing for one search where a single strong page should own it. |
| Core Web Vitals | Loading, responsiveness and layout stability as real visitors experience them, using field data where there is enough of it and lab tests where there is not. | A slow main image on mobile because it was never resized. A layout that jumps as a cookie banner loads and sends people to the wrong button. |
| Structured data | What markup exists, whether it validates, whether it matches the visible page, and what is missing for the page type. | Markup copied from a template that describes an address the business left years ago. Review markup for content that is not on the page. |
| Content gaps | The buyer questions and decisions that no page of yours answers properly, the pages that overlap, and the pages too thin to compete. | A cost question that buyers ask constantly with no page answering it. A hundred-word service page sitting against a competitor's full guide. |
| Backlinks | Who links to the site, to which pages, from what kind of site, and whether any pattern looks bought rather than earned. | Links won years ago pointing at URLs that now return an error. A cluster of links from placements that were obviously paid for. |
| Competitor set | The sites that actually share your search results, which are often not the businesses you think of as rivals, and what they do that you do not. | A rival list that includes none of the directories and publishers that actually fill the first page for your searches. |
| Analytics and tracking | Whether Google Analytics 4 and Search Console are set up so forms, calls and emails count as enquiries. | Form submissions never recorded as conversions, so no earlier work could ever have been judged. |
How does an audit check what a search engine can actually see?
By looking at the site twice: once as a browser shows it, and once as a crawler receives it before any JavaScript runs. Where the two differ, content may be invisible to some search engines and to most AI crawlers. Search Console then shows what Google has really indexed, which is the check that settles the argument.
The indexation work starts from a list of the pages that should be in search, meaning the ones that earn enquiries, and compares it with what Search Console reports as indexed, excluded and why. The findings are usually unglamorous: a noindex tag left over from staging, canonicals pointing at the wrong page, a sitemap that disagrees with the site, parameters generating duplicate URLs. None of it takes cleverness to fix. It takes someone actually looking.
Internal links come out of the same crawl. The audit maps how deep your important pages sit, which pages nothing links to, and where several weak pages are competing for a search that one strong page should own. If the business depends on the map pack, the Google Business Profile is checked as well. Where a crawl exposes a deeper problem on a large or JavaScript-heavy site, a dedicated technical engagement goes further than a general audit can, and the audit tells you honestly whether you need it.
The point
The audit maps how deep your important pages sit, which pages nothing links to, and where several weak pages are competing for a search that one strong page should own.
How are Core Web Vitals judged in an audit?
Against real-visitor data first. Core Web Vitals measure loading (Largest Contentful Paint), responsiveness (Interaction to Next Paint) and layout stability (Cumulative Layout Shift), and Google's guidance is to assess them at the 75th percentile of page loads, separately for mobile and desktop. Lab tests are used to find causes, not to hand out the verdict.
Lab tests run on one simulated device at one moment; field data records what visitors experienced. A page can pass in the lab and fail in the field because real visitors have older phones and slower connections. A small site may have too little traffic for field data at all, in which case the audit says so and relies on lab tests, labelled as such instead of presented as a pass mark.
Speed findings are also ranked honestly. A slow page that nobody visits and that never ranks is a lower priority than a slow page that earns enquiries. If you want the three measures explained without jargon first, what Core Web Vitals actually measure covers them.
What do the content, backlink and competitor checks tell you?
They tell you what to build and what to protect, not what you will rank for. Content gaps show which buyer questions have no good answer on your site, the backlink review shows where your authority sits and whether any of it is at risk, and the competitor set shows who really occupies your search results.
A content gap here is not a keyword list. It is a comparison between the decisions your buyers go through and the pages that exist: cost, comparison, suitability and “is this right for me” questions are usually the missing ones. The audit looks for the reverse problem too, pages that overlap so badly they compete, because consolidating them often does more than new writing. Producing the new content comes after that decision, not before it. Search-volume figures from any tool are estimates, and a sound audit labels them that way.
The backlink review asks three questions. Which pages attract links, and do those pages still exist and redirect properly? Is the profile consistent with a real business, with trade bodies, directories, customers and local press, or does it contain a pattern of purchased links? And how does it compare with the sites that outrank you? Most sites have a handful of lost links to recover and nothing worse. Disavowing links is for a genuine pattern or a manual action, not for every low-quality link a crawler flags, and we do not buy links.
The comparison is made against your three closest genuine competitors, chosen from the search results rather than from the list in your head. They are frequently not the firms you meet at trade events, and sometimes they are not businesses at all but directories and publishers that have taken the page.
The point
The comparison is made against your three closest genuine competitors, chosen from the search results rather than from the list in your head.
How does an audit change for local, ecommerce and international sites?
The core checks stay the same, but each kind of site adds its own: a local business adds the Google Business Profile and location pages, an online shop adds faceted navigation and product templates, and an international site adds language and country signals. The table shows what is added and where the work usually leads.
| Kind of site | What the audit adds | Where it usually leads |
|---|---|---|
| Single-location or service-area business | Google Business Profile completeness, citation consistency, and whether location pages hold anything that could only be written about that place. | Local search and Google Business Profile work, where the map pack is what produces the phone calls. |
| Online shop | Faceted navigation generating endless URLs, duplicate variants, thin category pages and product templates that repeat the same title at scale. | Ecommerce SEO, and the wider picture for ecommerce and retail businesses. |
| Site serving several countries or languages | Whether hreflang is consistent and reciprocal, whether the URL structure matches the strategy, and whether the right page is shown in the right market. | International SEO, because wrong hreflang is worse than none. |
| Software or JavaScript-framework site | Rendering above all: what the server sends, what a crawler can read, and whether the marketing site and the product are separated sensibly. | Specifications a developer can build from, and the choice between changing the content and changing the application. |
What does the finished SEO audit look like?
A written report you can hand to a developer: a short summary, findings ranked by impact, the evidence for each, the fix described precisely enough to act on, and a plain statement of what was not checked. It ends with a baseline of crawl, vitals and indexation figures, so that the effect of the fixes can be measured instead of assumed.
- A one-page summary of what is holding the site back, in order, written for the person who signs off the work.
- Findings rated high, medium, low or ignore by likely effect on enquiries, not by the severity label a tool assigns.
- Evidence for each finding: the affected URLs, what was observed, and how you can see it for yourself on the live site.
- A fix for each finding, described for a developer: what to change, where, and how to confirm it worked. Implementation by our own developers is available if you want it, but the description does not depend on it.
- A content section covering gaps, overlapping pages and thin pages, and a competitor section showing what the closest genuine rivals do that you do not.
- A backlink section: where your authority sits, what is at risk, and anything that needs attention.
- A list of what was left alone and why, so nobody spends a week on warnings that do not affect search.
- A baseline of crawl data, vitals and indexation status as they stand, and a note of anything the audit could not check, such as server logs or a staging copy.
- A walk-through with your team to agree the order of work and answer the questions a document cannot.
How does an audit decide what to fix first?
By likely effect on enquiries, not by how many pages a problem touches. Anything that stops pages being indexed or read comes first, problems that dilute or misdirect authority come second, and cosmetic warnings come last or not at all. The ranking is a judgement, so the report shows its reasoning for each finding.
| Rating | What it covers | Examples | What to do |
|---|---|---|---|
| High | Pages that cannot be found, indexed or read, or an inability to measure anything. | Noindex on commercial pages. Content drawn only by JavaScript. Canonicals pointing at the wrong URL. Enquiries not tracked. | Fix before anything else is built on top. |
| Medium | Signals that are diluted, misdirected or missing. | Thin pages competing for one search. Service pages nothing links to. Links pointing at deleted pages. Missing structured data. | Fix in the order the roadmap sets. |
| Low | Real but small. | Oversized images on pages with little traffic. Heading order. Long titles that get truncated. | Fix when you are already editing the file. |
| Ignore | Warnings that do not affect search. | The meta keywords tag. Informational notices that a tool flags by default. | Leave alone, and say so in the report. |
What does an SEO audit not do?
It does not fix anything, rank anything or promise a result. An audit is a diagnosis: it tells you what is wrong and what to do about it, while the outcome depends on the fixes being made, on competitors, and on search engines that nobody controls. It is also not a giveaway offered in exchange for your details; it is a piece of work and is scoped as one.
- It does not promise rankings, traffic or enquiries. Google's own guidance is that no one can assure a first position, and an audit that predicts an uplift is selling something.
- It does not make the changes. Findings alter the site only when somebody implements them, which is a separate piece of work, scoped once the findings exist.
- It does not replace a strategy. It shows where you stand; the keyword map and the roadmap that follow are a different job, informed by it.
- It does not recover a penalty. A manual action appears in Search Console, and recovering from one is specific work: removing the cause and then asking Google to review.
- It does not include link building or content production. It tells you whether either is worth doing.
- It is not a tool export with a cover page. A crawl is the raw material; the audit is the decisions made from it.
- It does not certify anything. Accessibility, privacy and legal compliance sit outside an SEO audit.
How do you judge an SEO audit from any supplier?
Judge it on whether you could act on it without the author in the room. A good audit ranks its findings, shows evidence for each, names what it did not check, separates what is wrong from what is merely different, and never predicts a traffic uplift. If the cover page could be swapped for another company's and nothing else changed, it is a tool export.
- Does it rank the findings? Three hundred items at equal weight hand the judgement back to you. Expect a short list that matters and a longer list marked as safe to ignore.
- Is there evidence for each one? You should be able to see the affected URLs and reproduce the problem. “Your site has duplicate content” with no example is an assertion, not a finding.
- Did anyone check what a crawler receives before JavaScript runs? On a modern site it is the costliest problem and the one a front-end review never shows. What a thorough audit includes, item by item sets out the checklist.
- Did it use your own data? An audit that never mentions Search Console or analytics was written from the outside. Read-only access is enough to produce one; be wary of anyone who asks for write access to your site or accounts to deliver a report.
- Does it say what it did not check? A credible report lists its limits: no server logs, no staging copy, a sample rather than a full crawl.
- Can a developer act on it? Each fix should say what to change, where, and how to verify it. “Improve page speed” is a wish, not a fix.
- Does it forecast results? Anything that projects traffic or revenue from an audit is guessing. A sound one says what the fixes make possible and what is outside anyone's control.
- Who did the work, and will they explain it? Ask for a named person and a conversation, not only a PDF. SEO here is led by Bikram Sah, with Saurav Dhoju handling development, and the people who would do the work are named. The same questions apply when you choose an SEO agency for the work that follows.
- Is the scope tied to the site? A small brochure site and a large catalogue are different amounts of work, so an identical quote for both, or one given before anyone has looked at the site, describes a product rather than an assessment. How audits are priced is covered in its own guide.
What is the quickest test of any audit?
Pick any finding and ask the supplier to show you, on your live site, how to see it for yourself. A real finding can be demonstrated in a minute. A finding that came from a tool's default settings, or that was never checked, cannot.
What makes one SEO audit take more work than another?
Size, platform and access. A few dozen server-rendered pages and a catalogue generating thousands of URLs are different amounts of work, and the number of locations, languages and competitors in scope changes it further. We do not publish prices, and an audit is scoped after someone has looked at the site, not before.
- How many URLs there are, and how many of them come from templates. A problem repeated across a catalogue has to be diagnosed once and verified everywhere.
- How the platform behaves. A standard CMS doing known things is quicker to assess than a bespoke application or a headless build whose behaviour is undocumented.
- Whether content is drawn by JavaScript. A rendering problem moves the fix out of the content and into the application.
- How many locations, languages or product categories are in scope, since each adds checks of its own.
- What data exists. Years of Search Console history and properly configured analytics make an audit faster and more reliable than starting from nothing.
- How contested the searches are. A crowded market means a closer look at the competitor set and the backlink gap.
- Whether the audit is expected to continue into written specifications for each fix, which is more work than naming the problem.
When is an SEO audit the wrong first step?
When the problem is not in the site, when the site is about to change, or when nobody can act on the findings. An audit examines what exists. It is the wrong first purchase for a site mid-rebuild, a business whose enquiries are lost after the click, or a team with no capacity to make the fixes.
| Your situation | Why an audit is premature | A better first step |
|---|---|---|
| The site is about to be rebuilt or moved to a new platform or domain | Findings about the old site are overwritten by the new one, and the real risk is losing what already ranks. | Plan the move so rankings survive it, starting with SEO for a website migration. |
| Traffic is steady but enquiries are not | The problem is after the click, and an audit of search visibility will not find it. | Look at the pages visitors land on and how well they convert. |
| You need enquiries this month | Search builds over time and an audit adds none of its own; its value is in the fixes that follow. | Compare the two routes honestly in SEO or PPC before committing to either. |
| You already hold an audit that was never implemented | A second audit adds to the pile without changing what gets done. | Review the priorities in the existing one, or ask an independent adviser to challenge them and the plan. |
| Traffic fell suddenly after a specific change | A full audit is broader than the question. The cause is usually in what changed. | Start with Search Console messages, indexing and redirects around the date of the change. |
| The site has not launched | There is nothing to crawl and no search history to read. | Build search requirements into the brief, then audit the finished site. |
What SEO Audit Service covers
Not a menu to pick from. Most projects use some of this and ignore the rest, and the scope conversation is mostly about working out which half you actually need.
Crawl and indexation review
Which pages a search engine can reach and which it has indexed, checked against the pages you actually want found: noindex tags, canonicals, sitemaps, redirects and duplicate URLs.
Rendering check
What a crawler receives before any JavaScript runs, compared with what a browser shows, because content that only exists after rendering is invisible to many crawlers.
Internal links and architecture
Click depth, orphaned pages, anchor text, and whether your commercial pages are supported by the pages around them or competing with them.
Core Web Vitals assessment
Loading, responsiveness and layout stability judged on field data where there is enough of it, with lab tests used to find causes rather than to award a pass mark.
Structured data review
What markup exists, whether it validates, whether it matches the visible page, and what is missing for each page type.
Content gap and overlap analysis
The buyer questions that no page of yours answers properly, and the pages that overlap so badly they compete with each other.
Backlink profile review
Which pages attract links, whether those pages still exist, and whether the profile looks like a real business or contains a pattern of paid placements.
Competitor set comparison
Your three closest genuine competitors, chosen from the search results rather than from memory, assessed on the same checks as your own site.
Analytics and tracking validation
Whether Google Analytics 4 and Search Console are set up so that forms, calls and emails count as enquiries and the work can be judged afterwards.
Why this is worth doing properly
Ranked by what it costs you
Each finding is rated by its likely effect on enquiries, not by the severity label a crawler attaches. The handful that matter come first and the rest are marked as safe to ignore.
Evidence you can check yourself
Every finding names the affected URLs and shows how to see the problem on the live site. If you cannot reproduce it, it should not be in the report.
It tells you what to leave alone
A report that flags everything has not made a decision. Ours lists the warnings that do not affect search so nobody spends a sprint on them.
Written for whoever does the fixing
Fixes are described for a developer: what to change, where, and how to confirm it worked. They can be done by our developers, yours, or another agency's.
Honest about its limits
No promise of rankings, traffic or enquiries, and a plain note of anything we could not check, such as server logs or a staging copy.
What you are handed, and who this suits
Deliverables
- A one-page summary of what is holding the site back, in order of importance
- Findings rated high, medium, low or ignore, each with the affected URLs and how to see the problem
- A fix for every finding, described for a developer
- A content section covering gaps, overlapping pages and thin pages
- A backlink section: where authority sits, what is at risk and anything needing attention
- A competitor comparison against your three closest genuine competitors
- A baseline of crawl, vitals and indexation status for measuring the work later
- A statement of what was left alone and what could not be checked
Everything on that list is yours on final payment, in your own company's accounts — the handover is written into the contract, not offered as a favour at the end.
Usually a fit for
- Sites whose traffic has stalled or fallen and where nobody can say why
- Sites that were redesigned or migrated and never checked afterwards
- Owners holding an audit from another supplier who cannot tell how much of it is right
- Businesses whose last agency reported rankings but never showed enquiries
- Sites built on a JavaScript framework where nobody has checked what crawlers receive
- Teams with someone able to make the fixes who need to know which to make first
Built with
- Google Search Console
- Google Analytics 4
- Core Web Vitals (LCP, INP, CLS)
- Schema.org structured data
- JSON-LD
- XML sitemaps
- Robots.txt
- hreflang
- 301 redirect mapping
- Non-rendering HTML fetch
Chosen per project, not per fashion. If your team already runs something else and can maintain it, that usually wins.
Ask what this would involve for youFour gates, and you sign off at each one
Every stage ends with something you can read and approve — access and baseline, crawl, render and compare, rank the findings, walk-through and handover. Nothing moves to the next stage until you say so.
- 01
Access and baseline
Read-only access to Search Console and Google Analytics 4, plus the live site. We record the current crawl, vitals and indexation status so later work can be measured, and we check whether enquiries are being tracked at all.
- 02
Crawl, render and compare
A full crawl, a non-rendering fetch of key templates, a look at your links and content, and a comparison against your three closest genuine competitors, chosen from the search results.
- 03
Rank the findings
Every finding is rated high, medium, low or ignore by its likely effect on enquiries, with the evidence and a fix a developer can act on. What was left alone, and what could not be checked, is written down too.
- 04
Walk-through and handover
We take your team through the report, agree the order of work and answer what a document cannot. What happens next is your call: implement with us, with your own developers, or take the report elsewhere.
Frequently Asked Questions
What is included in an SEO audit?
A good SEO audit covers crawl and indexation, rendering, internal links, Core Web Vitals, structured data, content gaps, backlinks, your closest competitors and whether your analytics can measure enquiries. The output is a written report with findings ranked by impact, the evidence for each, and fixes described clearly enough for a developer to act on.
How do I choose an SEO audit service?
Choose on evidence, not the pitch. Ask to see how findings are ranked, check that rendering and indexation are examined against your own Search Console data, and make sure the supplier says what it cannot check. Be wary of anyone who forecasts a traffic uplift or asks for write access merely to produce a report.
Do I need an SEO audit?
You probably do if traffic has stalled, you have never had the technical basics checked, or earlier work cannot be judged. You probably do not if the site is new and small, enquiries are lost after the click, or nobody can make the fixes. Whether you need an agency at all is a related question.
What is the difference between an SEO audit and a technical SEO audit?
A technical SEO audit goes deep on crawling, rendering, performance and structured data. An SEO audit is broader: it adds content gaps, backlinks, competitors and tracking, then ranks everything together. Small sites often need the broad one. Large, JavaScript-heavy or recently migrated sites usually need the technical depth as well.
Will an SEO audit improve my rankings?
Not by itself. An audit is a diagnosis: rankings change only if the fixes are made and crawled, and the result also depends on competitors and on search engines nobody controls. Be wary of any supplier that attaches a promised outcome to an audit. What it does is stop you spending effort on the wrong things.
What access do you need to audit my website?
Read-only access to Google Search Console and Google Analytics 4 is enough, plus the live site itself. Server logs and a staging copy help where they exist, and the report says so where they are missing. Nothing in producing findings requires write access to your site or your accounts.
Do you fix the problems the audit finds?
Yes, if you want us to. We build websites and software, so implementation by our own developers is normal work, scoped separately once the findings exist. The report stands alone, so you can give it to your own developers or another agency instead. A call to talk through the site is the quickest way to decide.
How often should a website be audited?
When something significant changes rather than on a fixed calendar: a redesign, a platform or domain change, a large content push, or an unexplained drop in traffic. Between those, routine monitoring in Search Console catches most problems. A site that changes constantly deserves attention more often than one that rarely changes.
Can I use an automated SEO audit tool instead?
An automated crawler is a fair first look, and a good audit uses one for its data. What it cannot do is rank findings against your business, read a rendering problem in context, or tell a harmless warning from a costly fault. Treat a tool report as raw material for an audit, not as the audit.
Should the audit come from the people who do the fixing?
Either can work, and each carries a risk. A supplier that also sells the fixes can drift towards findings that need its services, while an independent one may underestimate how hard a fix is. The protection is the same either way: ranked findings, evidence you can check, and a report you could hand to anyone.
Let's Build Something That Matters.
Tell us what you're trying to achieve and we'll help you find the right solution.