Docs / Side panel

SEO tab

How the page looks in Google, every SEO problem on it with how to fix it, broken links, the heading outline, structured data, language versions and share previews, in one tab.

Open it#

Open the side panel (0) and choose SEO. It checks the page as soon as it opens; Refresh checks it again after you change something. The number on the tab counts the errors and warnings found.

Besides reading the page in your tab, the SEO tab asks the site for a few things a search engine would: the page again (to see the status code, the X-Robots-Tag header and the HTML before JavaScript runs), /robots.txt, the sitemap, the share image, and one made-up address to see whether missing pages answer 404. These requests go only to the site you’re checking. See what TruePx sends.

Search Preview#

The page as a Google result: address, title and description. Under it, a meter for each shows its length in characters and in pixels, because Google cuts titles and descriptions by width, not by character count. A title over about 600 px or a description over about 920 px is marked Too long, with where Google would cut it.

No meta description? The preview says so: Google will pick text from the page instead.

SEO Issues#

Every problem found, most serious first. Each row has a severity icon:

Error stops the page from being found, shown or shared properly. Fix these.
Warning hurts search results or accessibility. Fix when you can.
Notice is worth knowing; often fine, sometimes a quick win.

Click a row to open it. You get the elements it’s about (click Show to scroll to one and spotlight it on the page, or open a link in a new tab), a How to fix explanation, and Copy for AI, which copies the issue, its elements and the fix as a prompt for Claude, Cursor or another coding assistant.

Links without text, links that go nowhere and empty headings inside content that’s hidden right now (a closed menu, an inactive tab) are reported separately as notices, so they don’t inflate the count.

Check links#

Links aren’t checked until you ask, because it means visiting every address on the page. Click Check links: TruePx requests each unique link (up to 500), first with a quick HEAD request, then with a normal request if that fails, and waits up to 12 seconds for each. The result says how many addresses were checked, broken, redirecting or had other problems, and adds them to the issues list. Check again repeats it.

Checking links on other sites needs Chrome’s permission to visit them. The first time, Chrome asks; if you decline, the check explains how to allow it.

Overview#

The page’s key facts: URL, HTTP Status, Canonical address, Robots rules (or “index, follow (default)”), the X-Robots-Tag header if the server sends one, Language, word count, and the author, publisher and keywords tags when present.

Headings#

How many headings of each level (H1 to H6) the page has, and its outline as a search engine reads it: each heading, indented by level. Empty headings show as “(empty)”. A good page has one H1 and goes down one level at a time.

Links#

Every link on the page with its text and address. Filter by All, Internal, External, Nofollow (links marked nofollow, sponsored or ugc) and, after a link check, Problems.

Highlight nofollow outlines nofollow links on the page itself.
Export CSV downloads the list (address, text, rel, internal or external, and the status after a link check) for a spreadsheet.

Images#

How many images the page has and which have no alt text. Click one to find it on the page. When every image has alt text but some are marked decorative (alt="", which screen readers skip), those are listed for you to review: make sure none of them carries information, such as a product photo, a chart, a linked logo or text in an image.

Structured Data#

Every JSON-LD block on the page, one card per item (for example Organization, SoftwareApplication, Product, Article). Open a card to see its properties as a tree, and copy the JSON. For types that can show as a Google rich result, TruePx checks the required properties and says either “Has every property Google’s … rich result requires” or which ones are missing. A block that isn’t valid JSON is marked Invalid JSON.

To confirm a result in Google’s own tool, use Rich Results Test in Quick Links.

Hreflang (language versions)#

Shown when the page lists other language or region versions of itself with hreflang. Each version shows its language code and address; This page marks the current one.

Google only uses a pair of versions when each links back to the other. Click Check each version links back: TruePx loads every listed version and reads its HTML and HTTP Link header. Each version then says Links back, Doesn’t link back: Google ignores this pair, Redirects to … (point hreflang at the final address) or Couldn’t load it.

Hreflang with three versions: en (This page), de and x-default (This page); after Check each version links back, de shows Links back; below, Share preview in Facebook style and the start of Meta Tags — The German version links back, so Google can use the pair.

Return links declared only in a sitemap can’t be seen this way; the result says so.

How a link to the page looks when shared on Facebook, X or LinkedIn (pick one from the menu): the card with its image, title, description and site, built from the page’s Open Graph and Twitter tags. The image only shows if it loads and is big enough to be used.

Meta Tags#

All the page’s meta tags as code, grouped into SEO, Open Graph and Twitter, with a copy button.

Quick Links#

One click to the tools you’d open next, for this page or site: robots.txt, sitemap.xml, PageSpeed Insights, Google’s Rich Results Test, the Schema Validator, a site: search on Google, and the Wayback Machine. They open in a new tab.

Every SEO check#

What each issue means. Titles are as TruePx shows them, with # standing for the number found on your page.

Title, description and headings#

Issue	Severity	What it means and how to fix it
No <title> tag	Error	The page has no title. Give it one of around 50–60 characters; it’s the headline of the search result.
Title may be cut off in Google results	Notice	Desktop results show about 600 px of a title and cut the rest with “…”. Put the key words first, or trim it to roughly 50–60 characters.
Title under 30 characters	Notice	A short title says little in search results; roughly 50–60 characters is ideal.
No meta description	Error	Write a summary of about 70–155 characters; search engines often show it under the title.
Meta description under 70 characters	Notice	Expand it to about 120–155 characters so the snippet says more.
Meta description may be cut off in Google results	Notice	Desktop results show about 920 px (roughly 150–160 characters). Put the important part first.
No <h1> heading	Warning	Add a single `<h1>` that names the page’s topic.
# <h1> headings found	Warning	Keep one `<h1>` and turn the rest into `<h2>` or lower.
Heading level skipped / # skipped heading levels	Notice	Go down one level at a time (h2 → h3, not h2 → h4) so the outline reads as a clear structure.
# empty headings	Warning	A heading with no text adds a blank entry to the outline. Give it text, or use a `<div>` or `<p>` if it’s only for styling.

Indexing and the server’s answer#

Issue	Severity	What it means and how to fix it
Page blocks indexing (noindex)	Error	A robots meta tag tells search engines to skip the page. Remove `noindex` if it should be findable.
HTTP header blocks indexing (noindex)	Error	The server sends `X-Robots-Tag: noindex`, so search engines skip the page even though the HTML allows it. Remove it from the server or CDN settings.
Page returns status …	Error	The address answers with an error status (for example 404 or 500), and search engines drop such pages. Make it return 200.
No canonical URL	Notice	Add `<link rel="canonical" href="…">` pointing at this page’s main address.
Canonical points to another page	Notice	Search engines will index that address instead. Fine for duplicates; on the main page, point it at this URL.
# tags are given conflicting values	Warning or Notice	A tag that should appear once (title, description, canonical, og:title, og:url, twitter:…) appears with different values, often one from the site builder and one from an SEO plugin. Remove the extras.
# important parts only appear after JavaScript runs	Error or Warning	Compares the HTML the server sends with the page after scripts run: the title, meta description, canonical link, H1 or most of the text is added by JavaScript, or a `noindex` in the HTML is removed by it. Google may render scripts late or not at all, and other search and AI crawlers often never do. Render these on the server (SSR or prerendering). A missing title, a removed noindex or mostly script-added text makes it an error.
The HTML before JavaScript and the X-Robots-Tag header couldn’t be checked	Notice	The site answered TruePx’s second request differently (a bot check, a login or an error). Check the page in Google Search Console’s URL Inspection to see what Google gets.
This is a bot check, not the site’s page	Error	A service like Cloudflare showed a challenge instead of the page, so nothing on it was checked. Complete it (or open the page in a normal window), then check again. It’s the only issue shown in that case.
No viewport meta tag	Error	Phones will show the desktop layout, shrunk. Add `<meta name="viewport" content="width=device-width, initial-scale=1">`.
<html> has no lang	Warning	Set the page’s language, for example `<html lang="en">`, for screen readers and search engines.

Links and images#

Issue	Severity	What it means and how to fix it
# broken links	Error	After Check links: these addresses answer “not found”, a server error or nothing. Point each at a working page, or remove it.
# links returned errors	Warning	The server answered with an error code or took over 12 seconds. Open the link to confirm, then fix or replace it.
# links redirect	Notice	Link straight to the final address; every redirect is an extra hop for visitors and search engines.
# links could not be verified	Notice	These sites answered 401, 403 or 429 to the check. Open them to confirm they work.
# links without text	Warning	Give every link visible text, an `aria-label`, or an image with alt text, so search engines and screen readers know where it goes.
# links go nowhere	Notice	`href="#"` and `javascript:` links can’t be followed. Give each a real address, or use a `<button>` for actions.
# jump links point to missing sections	Warning	Nothing on the page has the id the link jumps to. Add the id to the right section, or correct the link.
# links have an address in their rel attribute	Warning	`rel` takes keywords (nofollow, sponsored, ugc, noopener), not a URL, so a nofollow meant here isn’t applied. Put the address in `href`.
# images without alt text	Warning	Describe each image in an `alt` attribute; use `alt=""` for purely decorative ones.

Structured data and language versions#

Issue	Severity	What it means and how to fix it
# structured data blocks have JSON errors	Warning	Search engines ignore JSON-LD that doesn’t parse. Fix the syntax, then confirm in the Rich Results Test.
Structured data has an HTML code in its text	Notice	Inside JSON-LD nothing is decoded, so `Buyer's` is read literally. Write the character itself.
Structured data names this page differently from its title	Notice	The name shares no words with the title or H1, which usually means it was copied from another page’s template. Make it describe this page.
hreflang does not include this page	Warning	Each language version should list itself as well as the others.
No x-default hreflang	Notice	Add `hreflang="x-default"` for visitors whose language has no version.

Issue	Severity	What it means and how to fix it
No share image (og:image)	Warning	Shared links show a plain box or a random image. Add `og:image` at 1200 × 630 px.
The share image address isn’t a full URL	Warning	Most platforms ignore a relative `og:image`. Use the full address starting with `https://`.
The share image doesn’t load	Error	The address returns an error or isn’t an image, so shared links show no picture.
The share image is only … px	Warning	Small images show as a thumbnail or not at all. Use 1200 × 630 px (at least 600 × 315).
The share image is … px, so platforms will crop it	Notice	Large previews are 1.91:1. Use 1200 × 630 px and keep text away from the edges.
No share title (og:title)	Notice	Platforms fall back to the page title. Add `og:title` with the headline you want people to see.
No share description (og:description)	Notice	Some platforms show no text under the link without it. Add one or two sentences.
No robots.txt	Notice	Not required, but crawlers look there for your sitemap. Add one with a `Sitemap:` line.
robots.txt blocks the whole site	Error	`User-agent: *` with `Disallow: /` tells every search engine to stay out, often left over from a staging site. Remove it before launch.
No sitemap found	Notice	A sitemap helps search engines find new pages. Generate `/sitemap.xml` and list it in robots.txt and Search Console.
Missing pages answer “200 OK” instead of 404 (or redirect to the home page)	Warning	A made-up address on the site didn’t return 404. Search engines treat such pages as real (“soft 404”) and broken links never show up. Return status 404 with a helpful not-found page.