Nobody publishes a price for an SEO audit, and the usual explanation is that every site is different. That is true, and it is also the answer a buyer cannot do anything with. This page gives the thing underneath it: what actually moves the number, so you can work out roughly where your own site sits and tell whether a quote you have been handed is reasonable.
The short version is that page count barely matters, and almost everyone prices on it anyway.
The five things that actually drive the cost
1. Template count, not page count
An audit finds faults, and on a site built from templates the faults repeat. Fix the product template and you have fixed forty thousand pages. That means a large ecommerce site running six templates is genuinely quick to audit, because after the first few hundred URLs the crawler stops finding new kinds of problem and starts finding more of the same ones.
The expensive site is the small one with no templates. Four hundred pages, hand-assembled over ten years, three developers and two redesigns, where every page is its own decision and no fault predicts the next. That site takes longer than the fifty thousand page store and it will almost always be quoted lower, because it looks smaller.
If you want one question to ask yourself before reading any quote: how many genuinely different kinds of page does this site have. That number is closer to the cost than the page count is.
2. Whether the history is knowable
The second driver is how much of the past is still visible. Search Console holds sixteen months. A site that has had Search Console running for years, through its migrations, is straightforward, because the cause of a drop can be found by looking at when it happened.
A site where the property was created last month is a different job. The faults are still findable, but the ranking of them by what they cost becomes inference rather than measurement, and inference takes longer and is worth less.
3. Access
Search Console access is the one that changes the price most. Without it, roughly half of the checklist becomes guesswork, and guesswork is slower than measurement, not faster. Analytics access helps. Server log access is a luxury that occasionally settles a question nothing else can.
Sites where access has to be chased through three people at an agency that no longer replies are, genuinely, more expensive to audit than sites where the owner hands over a login.
4. Whether fixes are included
An audit is a diagnosis. Fixing is separate work with a separate cost, and the two get bundled and unbundled differently by everyone selling them, which is most of why quotes are hard to compare.
Both models are legitimate. What is not legitimate is a quote that leaves it ambiguous, because the failure mode is predictable: you pay for a document, discover the fixes are another engagement, and the document sits unactioned. If you only have budget for one of the two, buy the fixes and accept a rougher diagnosis.
5. What the output has to be
A list of faults for someone technical is quick to produce. The same findings written so a developer who has never seen the site can implement them without asking questions takes longer. Written so a board can approve a budget from it takes longer again, and is a different skill.
This is a real cost driver and it is almost never discussed during quoting, which is why so many audits arrive in the wrong register for whoever has to act on them.
The pricing models, and what each does to your incentives
Fixed fee. The most common and usually the fairest, because the incentive is to find the important things quickly rather than to spend time. The risk is scope: a fixed fee set before anyone looked at the site is a guess, and guesses get protected by narrowing what counts as in scope.
Hourly. Honest on unusual sites where nobody can predict the work. The incentive problem is obvious and worth naming out loud rather than pretending it does not exist.
Bundled into a retainer. Common and the one to look at hardest. An audit inside a monthly fee is not free, it is amortised, and the arrangement only makes sense if the fixes are also in there. A retainer that produces a quarterly audit and no fixes is monitoring.
Free. A crawler run with a logo on it. It will find blocking faults, which is genuinely useful, and it will not prioritise anything, because prioritising needs data it does not have. Take the findings, ignore the ranking.
What I publish, and what I do not
Being consistent about this: indexing has published prices because it is the same job every time. One-time work runs from $25 for up to 50 pages, $50 up to 120, $70 up to 170 and $100 up to 250. Monthly monitoring is $50, $100 or $150 depending on how closely it is watched. Those are fixed because the work is fixed, and the outcome is verifiable per URL.
Audits are quoted rather than published, for the reasons above, and the diagnosis conversation before the quote costs nothing. That is not a sales device, it is the only way to price the five drivers. A number given before seeing the site would be a number for an imaginary site.
The distinction is worth applying to anyone you are considering. A narrow, repeatable service should have a published price, and a service whose scope is genuinely unknown before looking should not. Someone publishing a fixed price for auditing any website has decided in advance how much of it to look at.
What makes an audit cost more than it should
Priced on page count. Now covered at length, and it is the single most common way to overpay for a simple site.
Priced on report length. A 300-item export is cheaper to produce than a 12-item list, because sorting is the work. Paying more for the longer document is paying more for less thinking.
Tool licences passed through. The tools that matter for this are Search Console, which is free, and a crawler, which is nearly free at ordinary site sizes.
Recurring audits on a stable site. A full audit earns its cost after something structural: a migration, a redesign, a platform change. Between those, checking Search Console for a new status appearing in volume takes minutes. Quarterly full audits on a site that has not changed is a subscription to a document.
Working out where your own site sits
Answer these four and you will have a better estimate than most quotes:
- How many genuinely different page templates does the site have?
- Has Search Console been collecting data through the changes that matter?
- Can you hand over access today, or does it need chasing?
- Do you need the fixes done, or a document your developer will implement?
A site with few templates, clean access, live history and its own developer is at the cheap end and should be. A site with none of those is not being overquoted when it costs more, it is genuinely more work.
If you are still deciding whether an audit is even the right purchase, technical SEO vs content SEO has the question that settles it: if every technical fault were fixed tonight, what would rank tomorrow. An empty answer means the audit is not what your budget should buy.
The short version
SEO audit cost tracks templates, history, access, scope and output format. It does not track page count, which is what almost everyone prices on, so small complicated sites get underquoted and large simple ones get overcharged. Ask what moves the number before comparing two quotes, be suspicious of anyone who prices before looking, and treat a shorter report as a better one rather than a thinner one.
Quick answers
Why will nobody publish an SEO audit price?
Because the work genuinely is not the same job twice, and a published number would either be too high for a small site or a loss on a complicated one. That is a real reason. It stops being a reason and starts being a dodge when nobody explains what moves the number, which is the gap this page exists to fill. Anyone quoting you a price before looking at the site is quoting a template, and you will get a template.
Is a free SEO audit worth anything?
It is worth exactly what a crawler run is worth, which is not nothing. A free audit will find blocking faults, broken links and missing tags, because those are mechanical. What it cannot do is prioritise, because prioritising requires Search Console data showing what Google actually did with your URLs, and a lead-magnet tool does not have that. Treat it as a smoke alarm rather than a diagnosis.
Should the fixes be included in the audit price?
They are separate work and should be priced separately, but you should know the fix cost before you commit to the audit, or you have bought a list you cannot act on. The question to ask upfront is what the fixes typically cost as a proportion of the audit. If the answer is vague, the audit is the product and the fixes are the upsell.
Does a bigger site always cost more to audit?
No, and this is the most common mispricing in the field. What costs time is the number of distinct templates and the amount of accumulated history, not the number of URLs. A 50,000-page store built from six templates is faster to audit than a 400-page site hand-assembled over a decade by four different developers, because on the first one every fault repeats and on the second every page is its own decision.
How do I compare two audit quotes?
Ask both for the same four things: whether they need Search Console access and what they do with it, how findings will be prioritised, whether fixes are included or quoted separately, and roughly how long the report is. The last one is the tell in reverse. A shorter report is usually the better one, because length in this field is a proxy for how much was left unsorted.