Check your website
Type your website address. This site’s server reads that page as it is right now and tells you, in plain words, what Google and AI assistants can read, what the page loads, and how easily people can call you or ask for a quote.
What the check readsChecking .0.0 s
- Read the page Secure connection, errors and how quickly your server answers.
- Read the content Title, description, headings, phone setup, contact links, forms and tags.
- robots.txt Which search and AI crawlers may read your site.
- Sitemap The list of pages search engines read.
- llms.txt A plain text guide for AI assistants, if you have one.
- Weigh the files Every file the page names, how heavy it is and who serves it.
The check reads the page as it is right now, from this site’s server. Results are not stored.
Google’s speed test is running on an emulated phone
The check did not finish
Results
What the files weigh
Two pages, measured two ways and drawn on one scale.
This opens your email app with the results filled in. Nothing is sent until you press send.
What each reading means
The results come in groups. Here is what each reading looks at, and why it matters to a business that runs on orders, calls and quote requests.
How the results are marked
- Problem
- Likely costing you. Worth fixing first.
- Worth a look
- Not broken, but worth checking.
- Note
- A plain fact about the page, for context.
- Looks fine
- Nothing to do.
Basics
Can people reach the page safely, and does it work on a phone?
How I work on websites and stores- Secure connection
- Whether the page ends up on an
https://address, the padlock in the address bar. Without it, browsers label the page as not secure, and nobody wants to type a phone number into a form marked that way. - Status
- What your server answered: the page itself, an error page, or a redirect to another address. Redirects are normal, but each one adds a wait, and a chain of them adds several.
- Server response time
- How long your server took to start sending the page, including any redirects on the way. It is measured from this site’s server, not from a visitor’s phone, so treat it as a guide to how quickly your server answers rather than a full load time.
- Mobile viewport
- Whether the page has the viewport tag that tells phones to fit it to the screen. Without it, a phone shows a shrunken desktop page that people have to pinch and zoom.
- Language
- Whether the page says which language it is written in. Screen readers use this to pronounce the words correctly, and browsers use it to decide when to offer a translation.
- Page size
- How much HTML the page sends, before the images, scripts and styles it names. A very large HTML file takes longer to arrive and longer for a phone to read.
- Title
- The title tag. Google often uses it as the clickable headline in search results, and browsers show it on the tab. The check shows the text and its length, because long titles get cut off.
- Description
- The meta description, a short summary Google may show under the headline. When it is missing or does not fit the page, Google picks a passage from the page instead.
- Main heading (H1)
- The biggest heading on the page. A single clear H1 that says what you do, in the words customers search for, tells people and search engines what the page is about.
- Canonical address
- A tag that names the preferred address of the page. It matters when the same page can be reached at several addresses, for example with and without
www. The check shows whether it points to the page itself or somewhere else. - Noindex
- An instruction, in the page or in the server’s response, that asks search engines to leave the page out of their results. On a home page it is usually a leftover from when the site was being built.
- Image alt text
- Short descriptions written on images. Screen readers read them aloud, and search engines use them to understand pictures. The check counts the images in the HTML that have none.
- Sitemap
- A file that lists the pages you want search engines to find. The check looks for one named in robots.txt or at the usual address on your site.
- Open Graph
- Tags that set the title, text and image shown when someone shares your link in a message or on social media. Without them, the preview is whatever the app guesses.
- Structured data
- Facts about your business written for machines inside the page, with schema.org types such as Organization, LocalBusiness or Product. Google uses some of them for richer results, and they help any system match the page to your business. The check reads JSON-LD, the format Google recommends, and lists the types it finds. Structured data written as microdata inside the HTML tags is not counted, so if the check finds none, it is worth confirming.
- Robots.txt rules
- A plain text file at the root of your site, robots.txt, tells crawlers what they may fetch, and well-known crawlers say they follow it. The check reads yours and reports, for each crawler in the table below, whether it may fetch the page you checked. With no robots.txt at all, crawlers treat the whole site as open. Some companies say their on-request visits may not follow it, because a person asked for the page.
- Search crawlers and training crawlers
Search crawlers fetch pages so a search engine or assistant can find them, quote them and link to them when someone asks a question. On-request fetchers visit a page when a person asks an assistant to look at it. Training crawlers collect pages that may be used to train future AI models.
Whether to let training crawlers in is a business choice, and owners decide it both ways. What matters is that each crawler reads its own rules: blocking a training crawler does not block the same company’s search crawler, and the reverse is true too.
| Name in robots.txt | What it does |
|---|---|
| GooglebotGoogle, search | Crawls for Google Search, including AI Overviews. |
| Google-ExtendedGoogle, Gemini training and answers | Not a separate crawler. A name in robots.txt that tells Google whether pages it already crawls may be used to train Gemini models and to ground Gemini answers. Google says it does not affect Google Search. |
| BingbotMicrosoft, search | Crawls for Bing search, whose results also feed Microsoft Copilot. |
| OAI-SearchBotOpenAI, search | Finds pages to show and link to in ChatGPT search. |
| ChatGPT-UserOpenAI, on request | Visits a page when someone using ChatGPT asks it to. |
| GPTBotOpenAI, training | Collects pages that may be used to train OpenAI’s models. Blocking it does not block OAI-SearchBot. |
| Claude-SearchBotAnthropic, search | Crawls to improve the search results Claude gives. |
| Claude-UserAnthropic, on request | Visits a page when someone asks Claude a question that needs it. |
| ClaudeBotAnthropic, training | Collects pages that may be used to train Anthropic’s models. Blocking it does not block Claude-SearchBot. |
| PerplexityBotPerplexity, search | Finds pages to show and link to in Perplexity answers. Perplexity says it is not used to train AI models. |
| Perplexity-UserPerplexity, on request | Visits a page when someone asks Perplexity a question that needs it. |
- llms.txt
- An optional, emerging convention: a plain text file at the root of a site that gives AI tools a short, clean summary of what the site is and links to its key pages. It is not a standard, and support for it varies, so having one is a small extra rather than a requirement. The check looks for one and shows what it finds.
- Text in the raw HTML
- How much readable text is in the HTML as your server sends it, before any scripts run. Google runs scripts before it reads a page, but many AI crawlers read only the HTML as it arrives. A page that builds its text with scripts can look almost empty to them.
- Files the page names
- Every script, stylesheet, image and font named in the page’s HTML, fetched by this site’s server and weighed one by one. These are what a phone may have to download to show the page. Files that scripts load later are not in the HTML, so they are not counted.
- Script files
- How many script files the HTML loads, and how many of them come from other companies. Each script has to download and run before parts of the page work, which adds up on a phone.
- Other companies’ files
- Files served from someone else’s domain: app scripts, chat and review widgets, tag managers, font services. Every other company is one more server for the phone to connect to and wait for, and together their files can outweigh your own.
- Sizes
- The size of each file as delivered after unpacking, which is roughly how much the phone has to process. Large images and large scripts are the usual places to look first. If the page names more files than the check weighs, the results say so.
- Google’s speed test on a phone
- When it is switched on, the check also asks Google’s PageSpeed Insights to load your page on an emulated phone with a slowed connection. That gives timings (when the first content appeared, when the main content appeared, how long scripts kept the phone too busy to respond, how much the layout jumped) plus Google’s own scores and suggestions.
- Real Chrome visitor data
- When enough people visit your site in Chrome, Google also reports how it went for them over recent weeks: how quickly the main content appeared, how quickly the page reacted to taps, and how much it moved around. If the page itself has too few visits, Google may report for your whole site instead, or not at all.
Calls and quote requests
Can someone who is ready to call or ask for a quote find the way?
How I run quote and lead flows- Tap-to-call links
- Phone numbers written as
tel:links, so a visitor on a phone can tap to call. A number in plain text has to be copied or remembered, and some people will not bother. - Email links
mailto:links that open the visitor’s email app with your address filled in.- Forms
- Contact, quote and booking forms in the HTML. The check sees whether a form is there, not whether its messages reach you, which is worth testing yourself.
- A visible contact or quote link
- A link with words like contact, quote or book, so someone ready to act can find the next step without hunting for it.
- Analytics and ad tags
- The tags that count visits, calls and sales: Google Analytics, Google Tag Manager, Google Ads, Meta Pixel and other common ones. They are how you tell which channels bring in orders, calls and quote requests.
- Only what is in the HTML
- The check lists only the tags written into the page’s HTML. Tags loaded through Google Tag Manager, or added by an app after the page loads, do not show up on their own. So if none are found, look in your analytics account before assuming nothing is measured. (This site has none, on purpose.)
Where the readings come from
When you press Check my site, this site’s server fetches the public page you type in (normally your home page) once, the way many crawlers do: it reads the HTML your server sends and does not run the page’s scripts. Every request it makes carries a clearly named user agent, HamzaGrowthSiteCheck, with a link back to this page, so the visit is easy to spot in your server logs.
It follows any redirects, reads your robots.txt, looks for a sitemap and an llms.txt file, then weighs the first files the HTML names, up to a set limit. Each step you see while the check runs is real work, timed as it happens, and How the check ran under the results lists every step.
Every reading comes from what your server actually sent. Nothing is estimated or filled in. The check does not log in, fill in forms or change anything on your site.
When the speed test is switched on, Google’s PageSpeed Insights runs it separately, from Google’s own servers. Those timings and scores are Google’s, shown as Google returns them.
What the check cannot see
- Pages behind a login, such as customer accounts or your store admin.
- Your Google Business Profile and how you show up on Google Maps.
- Merchant Center and your product feed.
- Your inbox, so it cannot tell whether form messages and quote requests reach you, or how they are followed up.
- Anything that loads only after scripts run, such as many chat widgets, booking tools and tags added by a tag manager.
- Any page other than the one you typed: product, service and collection pages, and the rest of the site.
- Firewall or bot protection settings that treat some crawlers differently from this check.
Those are the parts I look at when you email me. I take a close look at your whole online side, not only one page, and tell you what I would fix first. Here is how it starts.
What happens to the address you enter
It goes to this site’s server, only to run the check. The server uses it to fetch that public page and sends the readings back to your browser. Neither the address nor the results are stored.
When the speed test is switched on, the address is also sent to Google’s PageSpeed Insights, which tests the page under Google’s own terms.
Nothing else happens with it. There is no account and no sign-up. If you email the results to me, that is your choice, and nothing is sent until you press send in your own email app. The privacy page covers your connection details and the rest of the site.
Questions owners ask
Is it safe to run the check on my website?
Yes. It reads the page you type in (normally your home page) once, plus robots.txt, a sitemap, llms.txt and a limited number of the files the page names: the same things a browser or a search crawler asks for. It does not log in, fill in forms or change anything.
Why does a reading differ from what I see in my browser?
The check reads the HTML your server sends, the way many crawlers do. Your browser also runs the page’s scripts, which can add text, links, forms and tags after the page loads. If something only appears after scripts run, you see it, but the check and some crawlers do not.
Should I block AI training crawlers like GPTBot and ClaudeBot?
That is your call, and businesses decide it both ways. Blocking a training crawler does not block the same company’s search crawler, so you can opt out of training and still let ChatGPT search or Claude search read your pages. The check shows each crawler on its own line, so you can see what your robots.txt actually says.
Do you keep the address I check?
No. The address goes to this site’s server only to run the check, and neither the address nor the results are stored. If the speed test is switched on, the address is also sent to Google’s PageSpeed Insights. The privacy page has the details.
Can you fix what the check finds?
Most of what it finds is the kind of work I do every week: websites and Shopify stores, getting found on Google and in AI answers, and the measurement behind a live dashboard. Email me the results with the button under them, and I take a close look at your whole online side and tell you what I would fix first.
Want a closer look?
The check reads one page, usually your home page. When you email me, I take a close look at your whole online side and tell you what I would fix first.
Email HamzaEmail is the only way in for now.