AM Most SEO problems on a store or content site are simple once you see them: a product page with no meta description, a template that prints two H1s, a staging noindex tag that went live, a canonical pointing at the wrong URL. This checker fetches one page the way a crawler does, reads its HTML, HTTP headers, robots.txt and sitemap, and reports those problems in plain language, with a score and a pass, warning or fail for each check.
You can run 20 checks per day with no sign-up. Results for the same URL are reused for 30 minutes, so re-running a page right away does not use another check. After you fix something, wait a moment and run it again to see the new result.
How to check a page
- Copy the full address of the page you want to test. Start with the page that matters most: the home page, a top collection or a best-selling product.
- Paste it into the box above and press the check button. The tool fetches the page, follows up to 6 redirects, then fetches robots.txt and looks for a sitemap. It usually takes a few seconds.
- Read the verdict and the score. Then click the Fails filter to see only the serious problems.
- Open each failed or warning check. Every one shows what was found, why it matters and how to fix it.
- Scroll to the heading outline to see your H1 to H3 structure as Google reads it, with skipped levels highlighted.
- Click Copy report to paste the full checklist into a ticket, an email to your developer or a client report.
- Fix the issues in your platform, then run the check again. The page address updates with
?url=, so you can bookmark or share the report link.
Test one page of each type. On Shopify or WordPress, a problem in a product template appears on every product, so one fix often covers hundreds of pages.
The 23 checks
| Group | Checks | What a pass looks like |
|---|---|---|
| Technical | HTTPS, status code, redirects, response time, page size | https, HTTP 200, one redirect at most, HTML in under 800 ms, HTML under 800 KB |
| Indexing | noindex, robots meta and X-Robots-Tag, canonical | No noindex or nofollow, canonical pointing at this page |
| Site files | robots.txt, XML sitemap | A robots.txt, and a sitemap listed in it or at /sitemap.xml |
| Page setup | Mobile viewport, lang attribute, favicon | width=device-width, a lang value, a link rel="icon" |
| Content | Title, meta description, H1, heading outline, word count, image alt text | Title of 30 to 60 characters, description of 70 to 160, one H1, no skipped levels, 300+ words, alt on every image |
| Social | Open Graph tags, Twitter card | og:title, og:description, og:image, og:url, og:type and twitter:card |
| Structured data | JSON-LD schema | At least one schema type, and Product markup on product pages |
| Links | Internal and external link counts | At least one internal link |
How the score works
Each check has a weight based on how much it can hurt a page. A noindex rule carries the most weight (14), followed by the status code and the title (10 each), HTTPS and the H1 (8 each) and the meta description (7). Small items such as the favicon and lang attribute carry 2. A pass earns the full weight, a warning half, a fail nothing. If a page returns an error status or has noindex, the score is capped at 40 whatever else passes, because nothing else matters until the page can be indexed.
The score summarizes this checklist. It is not a ranking prediction. Google does not publish a score like this, and a page can score 100 and still rank poorly if its content does not match what searchers want.
Worked examples
A Shopify product page scoring 85
A product page loads over https with one redirect (http to https), returns 200 in 650 ms and has a canonical, sitemap and robots.txt. The checker finds six warnings: a 72-character title that Google will cut, no meta description, two H1s (the theme prints the product name and the store logo as H1), 3 of 12 images without alt text, no twitter:card, and og:type "product" with no Product schema. Those warnings cost 18 of 117 possible points, so the score is 85 and the verdict is "Good on-page basics". It is still worth fixing the title and description, because they decide how the listing looks in Google.
A blog that launched with noindex
A new WordPress site gets no Google traffic after a month. The checker shows a fail on "Indexing (noindex)" with the source: the robots meta tag. The score is capped at 40 and the verdict is "Critical problem found". The cause is the "Discourage search engines" box under Settings > Reading, left on from development. Unticking it and requesting indexing in Search Console fixes it.
Who uses it
- Shopify and WooCommerce store owners checking a product, collection and home page after installing a theme or an SEO app.
- Bloggers and niche site owners confirming that a new template has one H1, a canonical and schema before publishing 50 posts on it.
- Freelancers and agencies running a quick audit on a prospect's site to open a sales conversation, with the copied report as evidence.
- Developers doing a pre-launch check for leftover noindex tags, http links and missing Open Graph images.
- Students and new SEOs learning what each on-page element does, with the reason and fix shown for every check.
Fixing the most common problems
Missing or long titles and descriptions
Google has no fixed length limit for either, but it cuts both to fit the screen by pixel width. About 30 to 60 characters for titles and 70 to 160 for descriptions shows in full on most devices. The meta tag analyzer shows a Google preview of both. On Shopify, the SEO title and meta description guide shows where these fields live.
Canonical pointing elsewhere
A canonical that names a different page tells Google to index that page instead. That is correct for filtered or variant URLs and wrong for a main product page. Canonicals should point to a live URL that returns 200.
Redirects before the page loads
One redirect, such as http to https, is normal. Two or more means links or ads point at old addresses. Trace the chain with the redirect checker.
No robots.txt or sitemap
WordPress uses /wp-sitemap.xml and some plugins use /sitemap_index.xml, so a sitemap warning is not a problem if robots.txt has a Sitemap: line or you submitted the file in Search Console. If you have no robots.txt, the robots.txt generator builds one.
No structured data
Product pages should carry Product markup, and most sites benefit from Organization and BreadcrumbList. The schema markup generator produces valid JSON-LD. Note a 2026 change: Google stopped showing FAQ rich results on May 7, 2026, so FAQPage markup no longer earns the expandable questions under a result. It can stay if the questions are on the page, but it will not change how the listing looks.
Warning: A noindex fail on a page you want in Google is the most serious result this tool can show. Check theme settings, SEO plugins and platform password or staging options first, since those are the usual sources.
Troubleshooting
- "That does not look like a web address." Include the domain with its extension, such as example.com/page. Local and private network addresses are not checked.
- The check fails with a 403 or times out. Some firewalls and bot protection services block automated requests. The page may be fine for visitors. Try again later, or check with Google Search Console's URL Inspection tool.
- A tag I can see in my browser is reported missing. It is probably added by JavaScript after the page loads. This checker reads the HTML the server sends. Google renders JavaScript, so it may see the tag, but server-side tags are safer.
- The result did not change after my fix. Results are reused for 30 minutes, and your platform or CDN may cache the page too. Clear the cache and try again later.
- Password-protected store. A Shopify store with the password page on returns the password page, not your products. Remove the password before testing.
What this checker does not do
- It checks one page, not your whole site. It does not crawl for broken links, duplicate titles or orphan pages. For a full review, follow the Shopify SEO audit guide.
- It does not show rankings, traffic or backlinks. Use Google Search Console for those. To watch positions over time, AM Jarvis also has a Rank Tracker that checks your keywords daily.
- It does not measure Core Web Vitals. Response time here is the server time for the HTML only. PageSpeed Insights measures real loading, including Interaction to Next Paint.
- It does not judge content quality. Helpfulness, accuracy and search intent still need a human read.
Privacy: the checker only fetches public pages, and it identifies itself as JarvisTools in its user agent. We keep a copy of the result for 30 minutes so repeat checks are fast, and we do not store who checked which site.
If you are building new niche sites and want these basics right from the start, the AM Jarvis AI Site Builder generates pages with titles, meta descriptions, schema markup and internal links already in place.