A technical SEO checklist makes sure search engines and AI systems can find, read and trust every page of your website. For a therapy practice the essentials are: pages indexed with no stray 'noindex', an XML sitemap, a sensible robots.txt, canonical URLs, HTTPS, a mobile-friendly and fast site, valid structured data, no broken links, proper redirects, no duplicate pages, unique page titles, image alt text and accurate privacy pages. Most can be checked free in Google Search Console.
Prefer to tick it off as you go? Our free launch checklist covers these checks and more, and saves your progress as a PDF. Open the SEO and GEO website checklist →
What is technical SEO, and why does it matter for therapists?
Technical SEO covers everything that affects whether search engines can find, crawl, understand and trust your pages, as opposed to what the pages say. It's the foundation the rest sits on: excellent writing about EMDR or couples therapy won't rank if Google can't index the page.
The same foundation now serves AI search. Google says pages need to be indexed and eligible for a snippet to appear in AI Overviews, and ChatGPT search draws on Bing's index alongside OpenAI's own crawler. A technically sound site is easier for all of them to read.
The good news is that, unlike content, most technical issues are fixed once and then only need an occasional check.
Before you start: set up the free tools
- —Google Search Console (search.google.com/search-console): shows which pages are indexed, any errors, and what people searched to find you.
- —Bing Webmaster Tools (bing.com/webmasters): the same for Bing, which also feeds ChatGPT search and Microsoft Copilot. You can import your Search Console setup in a few clicks.
- —PageSpeed Insights (pagespeed.web.dev): speed and Core Web Vitals.
- —Google's Rich Results Test and the Schema Markup Validator (validator.schema.org): structured data.
The technical SEO checklist
| Check | Why it matters | How to test |
|---|---|---|
| Pages are indexed | A page that isn't in Google's index can't rank or appear in AI Overviews | Search Console > Pages report; or URL Inspection for a single page |
| No stray 'noindex' | Builders often hide sites during development and the setting is left on at launch | URL Inspection shows 'Excluded by noindex tag'; also check the platform's search visibility setting |
| XML sitemap exists and is submitted | Helps search engines find every page you want found | Visit yoursite.co.uk/sitemap.xml; submit it in Search Console and Bing Webmaster Tools |
| robots.txt is sensible | A wrong line can block your whole site, or crawlers you'd want to allow | Visit yoursite.co.uk/robots.txt; look for 'Disallow: /' and check which bots are named |
| Canonical URLs are set | Tells Google which version of a page is the main one, avoiding duplicates | URL Inspection shows the user-declared and Google-selected canonical |
| HTTPS everywhere | Secure connections are expected, and browsers warn visitors on insecure pages | Padlock in the browser; http:// versions should redirect to https:// |
| One preferred domain | www and non-www versions should not both be live as separate sites | Type both versions; one should redirect to the other |
| Mobile-friendly | Google indexes the mobile version of your site, and most visitors are on phones | Open the site on your phone; Lighthouse in PageSpeed Insights flags tap targets and font sizes |
| Fast and stable | Core Web Vitals are a ranking signal and affect whether visitors stay | PageSpeed Insights on mobile; Search Console Core Web Vitals report |
| Content readable without JavaScript | Some crawlers, including several AI crawlers, run little or no JavaScript | View page source (not Inspect) and check your main text is in the HTML |
| Structured data is valid | Helps machines understand who you are; errors can make it ignored | Rich Results Test and Schema Markup Validator |
| No broken links | Broken links waste crawling and frustrate visitors | Search Console Pages report for 404s; a free link checker for outgoing links |
| Redirects from old URLs | After a redesign, old addresses that people and Google know should lead to the new page | Visit old URLs; they should land on the right page via a permanent (301) redirect |
| No duplicate pages | Near-identical pages split signals and can look like low-quality content | Search for site:yoursite.co.uk and scan for repeats; check tag and archive pages |
| Unique page titles | Titles are a strong clue to what a page is about and are what people click | Search site:yoursite.co.uk and read your titles; look for repeats like 'Home' or 'Untitled' |
| Meta descriptions written | Not a ranking factor, but often used as the snippet people read | Same site: search, or a page's source code |
| One clear H1 per page | Helps people and machines see what the page is about | Lighthouse and most browser SEO extensions show heading structure |
| Image alt text | Describes images for screen reader users and search engines | Lighthouse accessibility audit lists images missing alt text |
| Clean, readable URLs | Short descriptive URLs are easier to understand and share | Look at your service page addresses |
| Privacy and cookie pages | UK GDPR requires a clear privacy notice; accurate legal pages are also a trust signal | Check the footer links exist and the content matches the tools you actually use |
The checks that catch therapists out most often
The leftover 'noindex'
Website builders and WordPress both have a setting that asks search engines not to index the site while it's being built. If nobody turns it off at launch, the site is invisible to Google indefinitely. Search Console will tell you; it's the first thing we check on any audit.
A redesign without redirects
When a site is rebuilt, page addresses often change: /services/anxiety becomes /anxiety-therapy. Without permanent redirects from the old addresses, Google treats the new pages as brand new and any links or rankings the old ones earned are lost. Keep a list of old URLs before any redesign.
Duplicate and thin pages
Blog platforms create tag, category and author archive pages automatically, and some builders create separate pages for every image. These can multiply into dozens of thin, near-identical pages. Either keep them useful or keep them out of the index.
Content hidden in the wrong places
Fees in a PDF, FAQs inside an accordion that only loads when clicked, a 'how I work' section inside an image. If it matters to clients, it belongs as plain text on a page.
Blocking AI crawlers by accident
Some security plugins and hosting settings block AI crawlers by default. Whether you allow them is a choice, but it should be your choice. Bear in mind that OpenAI's GPTBot (used for training) is separate from OAI-SearchBot (used for ChatGPT search), and that Google-Extended controls use of content for Gemini but doesn't affect Google Search ranking or inclusion in AI Overviews.
What each crawler does and what blocking it would change. Read: AI Crawlers Explained →
Do privacy pages really belong on an SEO checklist?
Yes, for two reasons. Legally, a therapy practice handles sensitive personal data, so a clear privacy notice under UK GDPR is required, and cookie information should reflect the tools your site uses. For trust, a mental health website is held to a high standard; Google's quality guidelines describe clear contact details and transparent information about who runs a site as signs of trustworthiness. A privacy page that mentions tools you no longer use, or none at all, undermines that.
How often should I run this checklist?
- —Fully: at launch, after any redesign or platform move, and once a year.
- —Monthly or every other month: glance at the Search Console Pages and Core Web Vitals reports, and Bing Webmaster Tools.
- —Whenever you add a plugin, widget or tracking script: re-test speed on mobile.
- —Whenever you change fees, hours or address: update the website, structured data and listings together.
How we build this into every website we make
Every item in the table is part of our build and launch process, so clients don't have to think about it.
- —Custom-coded Next.js sites, server-rendered or statically generated, so the full content is in the HTML. No plugins to misconfigure.
- —XML sitemap, canonical URLs, clean URLs, and unique meta titles and descriptions on every page, plus Open Graph images.
- —A robots.txt that allows search engines and, by default, AI crawlers including GPTBot, OAI-SearchBot, PerplexityBot, ClaudeBot and Google-Extended. You can choose to block any of them.
- —JSON-LD structured data (Organization, Person, WebSite, BreadcrumbList, BlogPosting, Service, FAQPage and VideoObject where relevant), validated before launch.
- —Optimised images with alt text, WCAG contrast and readable type, and Core Web Vitals checked on mobile.
- —Permanent redirects mapped from your old site when we replace one.
- —Google Search Console and Bing Webmaster Tools verified on launch, sitemap submitted to both.
The technical work is one stage of eight. Here's the full process, from research to after launch. Read: How We Build Websites to Rank →
If one check here failed and you're not sure what to do about it, our free resources page includes a website audit tool. Browse free resources →
Or bring your site to a free call and we'll go through it together. Book a free call →
Frequently asked questions
How do I know if Google has indexed my therapy website?
The reliable way is Google Search Console: the Pages report lists indexed and excluded pages, and URL Inspection checks a single page. A site:yoursite.co.uk search gives a rough idea but isn't a complete count.
What should a therapist's robots.txt file contain?
For most practices, very little: permission for crawlers to access the site and a line pointing to your sitemap. Make sure there's no 'Disallow: /' blocking everything, and decide deliberately which AI crawlers, if any, you want to block.
Do I need Bing Webmaster Tools if I have Google Search Console?
We recommend it. Bing's index is used by Bing itself, Microsoft Copilot and ChatGPT search, and verifying takes a few minutes if you import your Search Console setup.
Is HTTPS a ranking factor?
Google has confirmed HTTPS is a lightweight ranking signal. The bigger reason is trust: browsers warn visitors about insecure pages, which is the last thing you want on a therapy contact form.
Can I do technical SEO on Wix or Squarespace?
Many of these checks, yes: titles, descriptions, redirects, alt text and sitemap submission. You'll have less control over speed, structured data and how pages are rendered, because the platform decides those.
Adam works with UK therapy practices, EMDR trainers and mental health organisations on websites, social media and digital campaigns. He started Psyche Creative to give clinicians a marketing partner who understands the sector — and won't compromise on professional standards to get results.




