A professional SEO audit delivers four things: evidence, priority, developer-ready specs, and a validation plan. Each finding is backed by evidence from your own data. The findings are ranked by impact and effort. Each one is written as a specification a developer can build, and each carries a plan to confirm the fix held. A document missing any of those is a report.
That is the short answer to what deliverables are included in a professional SEO audit. By the end of this page, you’ll be able to judge any audit offer on deliverables, timeline, and cost. Every MDRN audit is diagnosed and written by a senior consultant with 15+ years in SEO. It isn’t passed to a junior analyst.
A professional audit ends in a fix plan, not a list of errors
An SEO site audit reviews crawlability, indexation, on-page elements, content, and links. If Google can’t crawl or index a page, it can’t rank it, so those checks come first.
Guides use the word “audit” loosely, and some define it in opposite directions, so here is the line we hold. A report lists what a tool found. An audit explains which findings matter, why they matter for this site, and exactly what to change. Our inputs are Google Search Console, SEMrush, a crawler, PageSpeed Insights, and Google Sheets. The diagnosis is the part no input produces on its own.
We start with what Search Console says is already working: which pages earn impressions, and where impressions are being lost. The fastest wins usually sit on pages that are close. Before anyone runs a crawl, decide which of three questions you’re asking.
Technical, content, and competitive audits answer different questions
Each audit type answers one question, and that question decides the scope.
Can Google crawl, render, and index your pages?
That’s the technical audit. It covers canonicals, redirects and 404s, and index bloat (hundreds or thousands of low-value URLs sitting in the index). It also covers internal linking, site architecture, structured data, and Core Web Vitals: LCP, INP, and CLS. Older checklists still list FID, but Google replaced it with INP in March 2024. One industry roundup put the share of sites passing Core Web Vitals at about a third, so speed findings are common. They’re rarely the first blocker.
Does the right page match each query?
That’s the content audit. It maps each target query to one page and checks whether that page answers what the searcher came for. Typical findings include cannibalization, where two pages compete for one query and split the signals between them. The audit also flags thin or outdated pages, and titles and H1s that describe something other than what the query needs.
Where do competitors win, and which gaps are worth closing?
That’s the competitive audit. It uses SEMrush overlap data to find queries where competitors rank and you don’t. It then sizes each gap by volume and fit, so only the worthwhile ones reach the roadmap. Jen Senwoo makes the point that your search competitors are often different companies from your business competitors. The data usually agrees: a publisher or a marketplace can own queries your sales team never thinks about.
Backlink, local (Google Business Profile, multi-location), and AI visibility checks join the blend when the business needs them. The scope follows the problem, so a site with an indexing problem can skip the competitive study until the blocker is fixed.
Every finding should arrive in five parts a developer can build
Here is one finding from a technical audit, written the way we deliver every row.
- Issue. Several high-value pages are excluded from indexing because canonical signals conflict with internal links. A canonical is the tag that tells Google which version of a URL is the preferred one. Here the tag points one way and the site’s own links point another.
- Impact. These pages are unlikely to rank consistently for their target queries, because Google keeps choosing a different URL to index.
- Evidence. The crawl shows internal links pointing at the non-preferred URLs. The GSC Pages report lists these URLs with the status “Alternate page with proper canonical tag.”
- Recommendation. Align the canonicals, update internal links to point at the preferred URLs, and resubmit those URLs through the XML sitemap.
- Acceptance criteria. Each preferred URL returns a self-referencing canonical. Internal links point only to preferred URLs. After recrawl, the URLs report as “Indexed” in the GSC Pages report.
Acceptance criteria are how a developer knows the ticket is done, and how you know the fix held.
Most recommendations described as actionable stop at the fourth part. Almost nobody writes the fifth, so the ticket sits open or gets closed on a guess. The five-part format also lets a mixed audience work from the same row: marketing reads the issue and impact to prioritize it, and engineering reads the recommendation and criteria to implement it.
The full package fits in a short summary, an issue sheet, a roadmap, and a validation plan
The package maps to the four items we list for every Technical SEO Audit & Fix Plan: findings, severity and impact prioritization, fix specifications for developers, and a validation plan. It arrives as a PDF or slide deck plus a working spreadsheet. The spreadsheet is what gets worked.
- A summary stakeholders read in ten minutes. It covers the top blockers, the page groups they affect, and the impact areas to watch. Agency Dashboard advises keeping summaries under two pages, and we agree. A thick deck impresses at kickoff and then sits unread.
- An issue sheet developers work from. The columns are URL or template, issue, evidence, impact, effort, priority, owner, acceptance criteria, and status. Findings are logged per template where possible, so one row for a product template can cover 4,000 URLs. If you want an SEO audit template, those nine columns are the useful part.
- A phased roadmap. Phase 1 fixes the blockers; Phase 2 scales the wins. Each phase names its dependency (dev support, approvals, or content workload), so nobody discovers in week three that a fix needed a sprint no one booked.
- A validation plan. For each fix, it names the signal that proves the fix landed: the GSC Pages report, URL Inspection, performance data by page group, and a recrawl. Server logs join that list where access is available.
Judge the package by whether the issue sheet can be assigned on day one.
Impact times effort is how 400 flagged issues become a short Phase 1 list
Score every issue for impact and effort, then fix in the order the ratio gives.
Impact runs from 1 to 3 across three questions. Reach asks whether the issue affects one template or one page. Page value asks whether it hits revenue or lead pages, or pages GSC shows losing impressions. Severity asks whether it blocks indexing or is cosmetic. Effort runs from 1 to 3 on developer hours and approvals. Divide impact by effort and sort.
Take three rows from a typical sheet. An indexation blocker on the product template scores impact 3 and effort 1, for a score of 3. It goes in Phase 1 because one template change touches thousands of URLs. Title rewrites on the top 10 converting pages score impact 2 and effort 1, for a 2. They sit close behind, because each page already earns traffic and the change goes live through the CMS without a developer. Missing alt text warnings on 300 blog images score impact 1 and effort 2, for 0.5, and that row goes to the backlog. Warnings like these fill an export and rarely move rankings. A sprint spent on them is a sprint the product template stays out of the index.
This is the Pareto principle applied to triage: a small share of issues usually drives most of the impact.
Ask to see the scoring column before you approve the roadmap.
Tool exports count problems but cannot rank them
Say a tool reports 1,200 pages with duplicate meta descriptions. The team spends two weeks rewriting them. Then someone checks the URLs and finds they’re faceted filter pages, color and size combinations that should never have been indexed. The real fix was indexation control on one filter template, and the rewrite budget is gone.
An export treats every row as equal, so working through it in order spends the budget on low-impact rows. The tool did its job. Judgment about which rows matter is what turns that export into an audit, and an AI summary of the export is still an export.
Three signs tell you which one you bought. The first is speed: SEO audits that arrive within an hour of purchase were never read by a person, so nobody checked the evidence. The second is missing evidence, meaning findings with no URLs, no GSC status, and no crawl data behind them. The third is a flat list, where nothing is prioritized and no row carries an effort estimate.
Whether you hire an SEO audit agency or an independent consultant, ask for one sample finding before you buy. If it lacks any of the five parts (issue, impact, evidence, recommendation, and acceptance criteria), the full document will lack them too.
One week covers a scoped check, while migrations and international sites need about three
When Academy Sports moved to a new ecommerce platform, the audit work was redirect mapping, faceted navigation, and indexation checks before and after launch. Migrations fail at the pattern level. One redirect rule governs thousands of URLs, and a quick scan that samples a few pages can miss a broken rule entirely. Academy Sports came through the move with sustained or improved visibility and was recognized as one of the top sites to gain SEO visibility within the year.
Before anyone names a duration, four things need to be in place: Google Search Console and GA4 access, CMS or staging access, a developer contact, and named approval owners. Access quality is the variable buyers don’t expect. A week spent waiting for a GSC invitation is a week the audit doesn’t move.
One week covers a crawl and indexation review, the top templates, priority pages, and a list of quick wins. It can’t cover template sampling across a catalog of tens of thousands of products, migration redirect validation, or JavaScript rendering diagnosis. It also can’t cover log analysis where log access exists, or hreflang across many country sites. On Intel’s site, that work spanned indexing and localization across 20 to 30 TLDs and ccTLDs, plus end-of-life content, redirects, and 404s. That kind of scope takes weeks of pattern checks.
The roughly three-week version fits four triggers: a traffic drop, a platform migration, an international site, or a large faceted catalog. Other practitioners quote similar ranges. Pella Force gives 3 to 5 business days for small sites and 2 to 4 weeks for sites over 500 pages, and Alan Bleiweiss puts deep audits at about five to six weeks. Clean access and a standard CMS shorten the timeline, while JavaScript front ends and slow approvals lengthen it. A small, healthy site doesn’t need the three-week diagnosis, and that’s where you should save your money.
Cost follows site size, platform complexity, and access quality
AgencyAnalytics puts SEO audit services at $300 to $15,000, taking anywhere from 20 to 200 hours. By size, it estimates $500 to $1,000 for sites under 20 pages and $10,000 to $15,000 for sites over 1,000 pages. An Ahrefs pricing survey found $2,500 to $5,000 the most common range for an SEO project, which is a broader unit than an audit. Alan Bleiweiss quotes $5,000, $8,000, and $15,000 per audit.
Three drivers move a quote inside those ranges. Site size comes first, since a 50-page site and a catalog of 50,000 product URLs need different sampling. Platform complexity comes second: a standard CMS audits faster than a JavaScript-rendered front end or a set of ccTLDs. Access quality comes third. Clean GSC and GA4 access with a responsive dev team shortens the work, while missing data and no staging environment add hours of reconstruction.
AI-assisted analysis lets us cover more ground in less time, and a consultant still reviews every finding. That speed goes into coverage and depth, so it shouldn’t be read as a discount.
Compare quotes on deliverables before price. A cheaper audit with no implementable specs costs more once your team has to work out the fixes itself. A $15,000 audit on a 50-page site is as mismatched as a $300 audit on a migration. Ask one thing of every quote: does it include fix specifications and a validation plan?
For our own work, the entry point is a one time $1,500 SEO Diagnostic at a fixed price: full crawl, Search Console analysis, competitor gap analysis, keyword-to-page mapping, and a prioritized fix list, credited toward your first month if you continue into a monthly program. That is a fixed scope diagnostic, and $1,500 sits inside the wide market range above. Ongoing monthly programs are a different unit of work than a one time audit: ecommerce programs start at $4,000 a month and run to $6,000 at catalog scale. Every figure is published on the MDRN SEO pricing page.
The audit pays off only when the Phase 1 fixes are implemented and hold
The expensive failure is a good audit that never gets implemented. Delivery ends with a roadmap walkthrough call, where we agree on Phase 1 and assign an owner to every row in it.
From there, you have two paths. Your team can work from the specs, or we can execute directly or work alongside your developers, with QA checklists, launch validation, and ongoing iterations. That hands-on work is part of our technical SEO consulting.
At SITKA, we worked with their developer to build internal linking rules through custom logic and CSV import. Over the four-year relationship, traffic doubled and the site ranked first for “hunting jacket” and “hunting clothes.” A multi-year result has many variables, including seasonality and product changes. The linking rules were implemented because they were written as specs a developer could build.
After deployment, we recheck in GSC. One industry guide says ranking movement typically takes four to twelve weeks to settle. Alan Bleiweiss notes that many clients need six months to a year to roll out the full task list. The audit usually feeds an SEO strategy and growth plan that sequences that longer list.
Questions buyers ask before commissioning an SEO audit
These five questions come up on almost every scoping call.
What are SEO deliverables?
It depends on whether you’re buying an audit or a retainer. Audit deliverables are findings, prioritization, fix specifications, and a validation plan. Retainer deliverables are ongoing implementation, a content system, and a reporting cadence. Check which one a quote is selling.
What does an SEO audit involve?
Discovery comes first, then data. The audit runs through goal alignment, a data pull and baseline from GSC, GA4, and a crawl, opportunity and constraint analysis, a prioritized roadmap, and a walkthrough. Access to those accounts on day one keeps it on schedule.
What is the 80/20 rule in SEO?
A small share of issues and pages usually drives most of the results. In an audit, that share is indexation blockers, high-reach templates, and revenue pages. That’s why the impact and effort score puts them in Phase 1.
Can ChatGPT do an SEO audit?
Partially. It can’t crawl a site at scale, it can’t see your Search Console data unless you hand it exports, and it can produce findings that sound right and aren’t. It’s useful for summarizing crawl exports and drafting title rewrites. We use ChatGPT Pro alongside other tools, never as a replacement for data, and a consultant reviews every finding before it reaches the sheet.
How often should you get an SEO audit?
At least once a year. Jen Senwoo suggests every 6 to 12 months, and any migration or redesign should get its own audit before launch. Between audits, watch the GSC Pages report for new exclusion statuses.
Scope the audit before anyone runs a crawl
Most disappointing audits were mis-scoped before the first crawl. Agree in writing on which site sections are in scope, which data access you’ll grant, which of the three audit types you need, and what the deliverable set includes. Then everyone judges the finished audit against the same page.
Request an audit scope conversation. No hard sell. We’ll point you to the next step most likely to move your numbers.
