Technical SEO That Ends in Shipped Fixes, Not a Spreadsheet

Most technical SEO engagements end with a 200-row spreadsheet, a call to walk through it, and an invoice. Six months later row 12 still is not done, because the person who found the problem never had commit access. We diagnose and we implement — same person, both jobs. Once the foundation holds, the same work compounds through SEO copywriting and local SEO.

Get a Technical SEO Audit See How We Work

The Problems Worth Paying Someone to Fix

None of these announce themselves. They show up as a flat traffic line while you keep publishing. If two or more sound familiar, start with an SEO audit before you spend anything more on SEO:

LCP Past 2.5 Seconds in the Field

Your lab score looks fine. Chrome field data — the 75th percentile of real visits — puts Largest Contentful Paint past the 2.5s good threshold.

Interactions That Lag on Real Phones

INP over 200ms is almost always long JavaScript tasks blocking the main thread, not a slow server.

Rankings Fell After a Redesign

Dropped redirects, a changed URL structure, or canonicals that quietly started pointing somewhere new.

Index Bloat

Faceted navigation, filters, and tracking parameters generating thousands of near-duplicate URLs that dilute the pages you care about.

Googlebot Crawling the Wrong Pages

Server logs show crawl budget going to parameter URLs and dead archives while new pages sit undiscovered.

Rendering Blocked Without You Knowing

Mixed content, a CSP blocking a script the page needs, or a robots.txt rule hiding the CSS Google renders with.

What We Fix, and How We Know It Is Fixed

Every item below has a number attached to it or a file it lives in. None of it is a matter of taste.

Crawl, Render, and Log Analysis

Three passes that are meant to disagree: what the raw HTML says, what the rendered DOM says, and what Googlebot actually requested.

What's Included:

Raw HTML vs rendered DOM, diffed
Googlebot hit frequency by directory, pulled from server logs
Index Coverage export reconciled against your sitemap
Findings ranked by the sessions they put at risk

Core Web Vitals, Measured in the Field

LCP at or under 2.5s, INP at or under 200ms, CLS at or under 0.1 — at the 75th percentile of real visits, not on your laptop.

What's Included:

Render-blocking JavaScript cut out of the hydration path
TTFB budget set against Google's 800ms threshold and served from the edge
LCP element identified and preloaded rather than guessed at
CLS traced to the element that actually shifts — usually a late font swap or an unsized embed

Mobile-First Indexing Parity

Google indexes the mobile rendering of your page. If that rendering is a lighter version of the desktop one, the lighter version is your site.

What's Included:

Content, headings, and internal links matched across breakpoints
Structured data confirmed present in the mobile render
Interaction latency profiled on mid-range hardware, not a flagship phone
Tap targets and viewport handling checked on real devices

Crawl Budget and Index Control

Most sites do not have an indexing problem. They have an indexing-too-much problem, and it is costing them the pages that matter.

What's Included:

Faceted navigation and filter parameters stopped from spawning near-duplicate URLs
Index bloat quantified by reconciling indexed count against real page count
Self-referencing canonicals on paginated archives, not everything pointed back at page one
Sitemaps restricted to canonical, indexable, 200-status URLs

Structured Data That Validates

Schema that describes what is actually on the page, checked in the Rich Results Test rather than pasted out of a generator.

What's Included:

LocalBusiness, Service, and FAQ markup matched to visible content
Cross-referenced @id graph instead of duplicate disconnected blocks
Enhancement reports in Search Console watched after release
Markup removed wherever it claims content the page does not have

Redirects, Rendering, and Security Headers

The quiet category. None of it looks broken to you, because your browser has a warm cache and Googlebot does not.

What's Included:

Redirect chains collapsed to a single hop
Mixed content and HSTS configuration resolved
CSP audited so it is not blocking scripts the page needs to render
robots.txt checked for blocked CSS and JS that Google renders with

What We Run, and What Each One Is Actually For

Every tool here is available to you too. The tools are not the service — knowing which output to ignore is.

Crawling and Change Monitoring

Where the candidate list comes from, before anything on it gets verified by hand.

Screaming Frog for the render diff

Crawls raw HTML and the rendered DOM separately, so the two can be diffed and content that only exists after JavaScript runs shows up

SEMrush for scheduled re-crawls

Catches the regression your next plugin or theme update introduces, in between full audits

Performance Measurement

Lab numbers diagnose. Field numbers are what you are graded on. You need both, for different reasons.

PageSpeed Insights for lab against field

The one place both sit side by side — when they disagree, the field data is the one that counts

GTmetrix for the waterfall

Which request blocks render, and what it is waiting on before it can even start

Ahrefs for what a migration must preserve

Redirect chains, broken internal links, and which URLs hold external links worth keeping alive

Google's Own Reports

The only data that shows how Google is treating your site rather than how a third party models it.

Google Search Console for coverage and crawl stats

Index Coverage, Crawl Stats, and the 28-day rolling Core Web Vitals field report

Google Analytics for prioritisation

Sessions per URL, so the fix list gets ordered by traffic at risk instead of by a tool's severity label

Server Logs, Which No Dashboard Will Show You

Crawl frequency is the one thing you cannot infer from a crawler. We pull raw server or CDN logs, verify Googlebot by reverse DNS so spoofed user agents do not skew the count, and total hits per directory over a date range. That is how you find out Google is spending its budget on filter URLs and has not touched your service pages in weeks. Then we open the page source and read it.

What Changes After the Fixes Ship

Each one is observable in a report you already have access to.

Pages Leave Discovered, Currently Not Indexed

Removing crawl traps and linking pages properly is what moves URLs out of that status in Search Console.

Your Money Pages Stop Competing With Each Other

Correct canonicals consolidate duplicate and parameter URLs onto the one page you actually want ranking.

The Core Web Vitals Report Turns Green

Not because a lab score moved, but because the 75th percentile of real visits did.

Crawl Budget Goes Where the Revenue Is

Log data shows Googlebot hitting service and product pages instead of filter combinations.

The Same Fixes Help Conversion

The render path Google grades is the one your visitor waits through before they ever see your form.

Your Next Redesign Does Not Cost You Rankings

You end up owning a redirect map, a canonical rule, and a documented template inventory — the three things every replatform loses.

The One Result We Will Quote You

We do not publish percentages we cannot attach to a named person. This one we can.

“Grindstone Design has been doing SEO for my company for the past 7 months and has made our rankings in Google shoot up. We've increased the amount of people coming to our website by over 500% from when we started. Can't recommend enough!”

Anton Montgomery — 7-month SEO engagement, Google review. One of 47 Google reviews averaging 5.0.

Technical work was part of that engagement and it was not the whole of it. We are not going to tell you which share of that number belongs to a redirect map, because we do not know — and neither does anyone selling you an attribution model for it.

Start With an Audit

How the Work Actually Runs

Five stages. What you hand over, what you get back, and who is doing it.

Access and Baseline

Before anything is touched, we record the numbers we intend to be judged against later.

Search Console, analytics, and log access set up

Field Core Web Vitals captured at the 75th percentile

Indexed URL count and template inventory recorded

The one metric this engagement has to move, agreed in writing

Crawl, Render, and Log Analysis

The diagnostic pass. Automated where automation is reliable, manual where it is not.

Full crawl of raw HTML and rendered DOM, then diffed

Googlebot hits totalled by directory from server logs

Coverage export reconciled against the sitemap

Every finding reproduced by hand before it goes on the list

Prioritisation by Traffic at Risk

Findings ranked by pages affected against the sessions those pages earn, not by a tool's severity colour.

Fix list ordered by revenue exposure, not issue count

Each item names the file or setting it lives in

Anything your platform will not let us change, flagged now

You approve the list before implementation starts

Implementation

The step most engagements skip. James Price ships the fixes — the same person who found them.

Changes made in your repo, CMS, or CDN config

Staged first, crawl-diffed against the baseline, then released

Or handed to your developer as a ticket with the exact change and a verification step

No account manager relaying messages between you and the work

Verification and Monitoring

A fix nobody verified is a claim. We re-run the baseline and show you the difference.

Post-release crawl compared against the pre-work snapshot

Field Core Web Vitals tracked across the 28-day rolling window

Crawl Stats and log hits reviewed for the pages you care about

Regressions caught before they compound

Three Ways to Work Together

The difference between them is who implements, and how often.

Audit Only

You have a developer. You need to know exactly what to tell them.

Full crawl plus rendered DOM diff
Log-file crawl analysis where logs are available
Fix list ordered by traffic at risk, each item naming the file or setting
One call to walk your developer through it

Get the Audit

What Actually Drives the Price

Retainers run on the SEO pricing published on our services page. Audit and implementation projects are scoped after we have looked, because three things move the number and raw page count is not really one of them:

Template Count

A 5,000-URL catalog built from six templates is less work than a 200-page site where every page was hand-built

Platform Access

A repo we can branch costs less than a hosted builder that locks the head of the document

Who Implements

Handing your developer a spec is cheaper than us shipping it, and slower

Get a Technical SEO Quote

Why Hire Us for This

Six reasons, all of which you can hold us to.

The Person Who Finds It Ships It

James Price runs the crawl, the log analysis, and the implementation. No account manager relaying messages, no junior handed the file after you sign.

Field Data, Not Lab Scores

A green score on your laptop proves nothing. We work against the 75th percentile of real visits, because that is the number Search Console grades you on.

You Get Changes, Not Documents

The deliverable is a released fix. The write-up exists to explain what changed and how to verify it — it is not a substitute for the work.

We Do Not Break What Is Already Ranking

Every release is crawl-diffed against a baseline captured before we started, so a fix to one template cannot quietly strip the canonicals off another.

Credentials

8+ years in conversion optimization and web design, CXL Institute CRO certified, Google Ads and Meta Ads certified, and 5.0 across 47 Google reviews.

We Build Sites, So We Know What Breaks

Technical SEO advice from someone who never has to ship it ages badly. A recommendation that would wreck your build pipeline or your editors' workflow does not leave the audit.

Questions We Get Before Signing

The honest answers, including the ones that cost us work.

Will you implement the fixes, or do we just get a report?

We implement. The person who finds the problem is the person who ships the fix, working in your repo, your CMS, or your CDN config. If you would rather your own developer do it, we write the ticket at a level a developer can act on — the exact file, the exact change, and how to verify it — and we review the release. What we do not do is hand over a 200-row spreadsheet and call that the deliverable.

How long until we see results?

Speed fixes are measurable in the lab the day they ship. Search Console grades you on field data instead, over a 28-day rolling window, so a Core Web Vitals fix takes roughly a month to fully show there. Indexing changes move as Google recrawls, which you can watch in Crawl Stats and in your own server logs. We will not quote you a ranking date — anyone who does is guessing.

What access do you need?

Google Search Console, your analytics, and ideally raw server or CDN logs — the logs are what tell us where crawl budget is actually going. For implementation we need the repo, CMS admin, or CDN dashboard, plus a staging environment if one exists. If your platform locks the part that needs changing, we tell you before you pay for the audit, not after.

What tools do you use?

Screaming Frog to crawl raw HTML and the rendered DOM separately so the two can be diffed, PageSpeed Insights and Search Console for Chrome field data, GTmetrix for request waterfalls, Ahrefs and SEMrush for link and change monitoring, and raw log parsing for Googlebot hit frequency by directory. Tools produce candidates. Reading the page source is what confirms them.

Is technical SEO a one-time project or ongoing?

Most of it is one-time. A redirect map, a canonical strategy, a fixed render path — those stay fixed until something changes them. What changes them is your own release cycle: a plugin update, a theme change, a new filter added to a category page. If you ship changes often, a retainer earns its keep on release review alone. If your site is stable, take the project and re-audit in a year.

What do you need to give us a quote?

Your URL, roughly how many pages and how many distinct page templates you have, your platform, whether you have a developer, and what you have already noticed going wrong. Template count matters more than page count — a 5,000-URL catalog built from six templates is less work than a 200-page site where every page was hand-built.

Find Out What Is Actually Wrong

You will get a straight answer on whether your problem is technical at all. Sometimes it is not — sometimes the crawl is clean and the offer on the page is what is losing you the job. We will tell you that instead of selling you an audit.

Request a Technical SEO Audit Call (904) 608-8534

A 15-minute call, no deck. You talk to James Price.

Technical SEO That Ends in Shipped Fixes, Not a Spreadsheet

Find Out What Is Actually Wrong

A 15-minute call, no deck. You talk to James Price.