Free tool Chrome extension · v2.0.0Get the Chrome extension →

SEO Readout, a free SEO Chrome extension, in one click.

SEO Readout reads the page you are on and tells you, in plain words, what is right and what is wrong with it for search. One score, quick wins first, nothing sent anywhere.

Add to Chrome, free →

We built it because we wanted one. The free SEO extensions either nag you with pop-up tabs, ask to read every site you visit, or stop at listing what is on the page without saying whether it is right. This one judges. It runs on the tab you click it on and on nothing else, and it has no server to phone home to.

Click any finding and the page scrolls to it and outlines it: the image with no alt text, the heading that skips a level, the link with no text.

Its sister, Rank Readout, works on the Google results page instead: every result outlined, where your site ranks, and what the AI Overview cites.

Every tab in under two and a half minutes, on a real Newcastle wedding page. Plays from YouTube when you press play, or watch it on YouTube.

What it checks

  • Title and meta description.

    Length in characters and in pixels, the way Google actually cuts them. Duplicates, placeholders, a description that repeats the title.
  • Headings.

    One H1, no skipped levels, nothing hidden, and the whole outline drawn as a tree.
  • Images.

    Missing alt text, placeholder alt text, missing width and height (that is layout shift), a lazy-loaded hero.
  • Links.

    Every link on the page with its anchor text, internal or external, nofollow, sponsored and ugc, and whether it sits in the menu, the body or the footer. It flags links with no text, "click here" links, nofollow on links to your own pages, and http links on an https site. Filter them, click one to see it on the page, or copy the lot as CSV. Links are read, never visited.
  • Canonical.

    Missing, relative, pointing somewhere else, or http on an https page.
  • Indexability.

    The robots meta tag, the X-Robots-Tag header most tools never read, and robots.txt for the exact URL you are on. A noindex page gets a banner across the top of the report so it cannot be missed.
  • Structured data.

    Every JSON-LD block parsed and every node checked for the properties Google requires and recommends for its rich result: Article, Product, LocalBusiness, Breadcrumb, Event, Recipe, Video, JobPosting, Review, Organization, WebSite and more. FAQPage is checked too: Google stopped showing FAQ rich results in May 2026, but the markup still helps AI answer tools read the questions. It catches references to an @id that is not on the page, dates that are not ISO 8601, and relative URLs.
  • Open Graph and X cards.

    Every tag with what it falls back to, and a mockup of the card on Facebook, LinkedIn, X, Slack and Discord with the real image loaded and measured against 1200 by 630.
  • The Google result.

    A preview on desktop and mobile with the title and description editable, so you can try a rewrite and watch the width meter before you touch the site.
  • The page before JavaScript.

    The page fetched again as plain HTML, the way ChatGPT, Claude and Perplexity's crawlers read it, and set beside what the browser shows. Words, H1, title, canonical and schema that only appear after a script runs are flagged, and so is a noindex in the HTML that a script takes away.
  • AI crawlers.

    Which AI answer and training crawlers the site's robots.txt lets onto the page, one row each, with the rule that decided it. Blocking an answer crawler costs the site its place in AI answers and is flagged; blocking a training crawler is a choice and is noted as one. llms.txt is checked too, and never costs points.
  • The whole site.

    New in 2.0: press Site check and it reads every page it can reach from the site's own links and sitemap, up to 500, a few a second, slowing down if the site asks it to, obeying robots.txt. It finds broken internal links with the page and the link text, links through redirects, missing and duplicate titles, descriptions and H1s, canonicals pointing at dead pages, noindexed pages in the sitemap, orphan pages nothing links to, and thin pages. Only that site's pages are ever requested.
  • Local details.

    The business name, phone number, postcode and suburb in the LocalBusiness schema, looked for on the page and in its click-to-call links. It catches a schema phone number the page does not show, a call link that shows one number and dials another, and schema that never points at the Google Business Profile.

What the score means

Four modules, each out of 100: Basics (title, description, canonical, indexability), Structure (headings, images, text, links), Schema and Social. A fail costs its full weight and a warning half of it, and the four combine at 40, 25, 20 and 15 per cent into the one number in the ring. Ninety and above is an A.

Passing checks are folded away, so the first thing on the screen is what to fix. Quick wins, the one-line changes a developer does in minutes, come first. The whole report copies out as Markdown for a ticket or a client email, as JSON, every link as CSV, or as a score card image for Slack.

What it never does

  • Send page content, URLs or anything else to a server. There is no server.
  • Ask for an account, a login or a licence key. There is no paid tier.
  • Open a tab on install, on update, or ever. No changelog pop-ups, no donation asks. Safe to open in front of a client.
  • Ask for permission to read every site you visit. It runs on the tab you click it on, and only then.

Privacy

SEO Readout does not collect anything. There is no account, no analytics, no telemetry, no error reporting and no server. We cannot see which sites you use it on, because nothing is ever sent to us. Every request it makes goes to the site you are checking, and nowhere else.

What it reads. When you click the icon, it reads the page in the tab you clicked it on: the title, the meta tags, the headings, the images, the links, the text and any JSON-LD. That reading stays in the popup's memory and is gone the moment the popup closes. Nothing runs on any other tab, and nothing runs before you click.

The site check. Only when you press Site check, and then Check the whole site, it reads the rest of that site: it starts from the page you were on, follows the site's own links, and reads the site's sitemap. It reads each page's HTML without cookies, a few pages a second, two at a time, slowing down by itself if the site asks it to or answers slowly, up to 500 pages, and it obeys the site's robots.txt, including any Crawl-delay. It only ever requests pages on that one site: links to other websites are listed, never visited. The results stay in the site check tab's memory and are gone when you close it, unless you export them. It never runs in the background or on a schedule; every check is one click.

What it changes. Nothing, unless you ask. When you click a finding or a link in the report, it scrolls the page to that element and draws an outline over it for a few seconds. The outline sits on top of the page and removes itself; nothing on the page is edited and nothing is saved.

What it requests. On a single page, four things, and all of them go from the page to its own site, not to us:

  • a HEAD request for the page's own URL, to read the X-Robots-Tag header, which can carry a noindex that never shows in the HTML;
  • a GET request for the same URL, without cookies, to read the HTML before JavaScript runs, the way a crawler does. It is read in the page and only a summary (word counts, the title, the H1, the canonical) reaches the popup; the HTML itself is discarded;
  • a GET request for /robots.txt on the same site, to check whether that URL is blocked, for Google and for each AI crawler;
  • a GET request for /llms.txt on the same site, to see whether there is one.

When you run a site check, it also requests the site's pages, /robots.txt and its sitemap, from that same site only, as described above.

The social preview loads the page's own og:image so it can show you the card and measure the image. That request goes to wherever the page hosts the image, the same as sharing the link would.

What it stores. A few preferences in the popup's own local storage: which tab you had open, the desktop or mobile preview, the social card you last looked at, and the Links tab's filter. No page data is stored.

Permissions. activeTab (read the tab you clicked on, and for a site check the rest of that same site) and scripting (run the reader in that tab, and the outline when you click a finding). It does not ask to read every site you visit, and it never will.

Third parties. None. No fonts, scripts or images are loaded from anyone else's servers. The links out are the three in the footer: the "Built by Newcastle Web Co" credit, "Rank Readout" (our other free extension, from version 1.2.1), and, from version 2.0.0, "Help us keep this free · Review it on the Chrome Web Store", which opens the extension's review page. None of them goes anywhere unless you click it. Exported reports end with the same review link, as text.

Changes. If any of this ever changes, the extension's version number will change with it and this page will say what changed and when. Questions: admin@newcastlewebco.com.au. Last updated 18 October 2026: version 2.0.0 adds the site check, which requests up to 500 pages of the site you are checking, from that site only, when you press it. Version 1.2.0 added the outline that shows a finding on the page, which makes no requests. Version 1.1.0 added two requests, the HTML before JavaScript and /llms.txt.

Why a web studio gives this away

Every site we build goes through these checks before it ships, and every site we take over gets them on day one. Putting the checks in a popup means a business owner can run them on their own site in ten seconds and see the same list we would send them. Some of those people will fix it themselves. Some will ask us to. Both are fine. If the report on your site is longer than you would like, our SEO work in Newcastle starts with exactly this list, and the structured-data half of it is what our GEO work builds on.

We asked the same questions of 742 business websites across Newcastle, Lake Macquarie and Maitland in our study, The State of Small Business Websites in the Hunter. The median mobile speed score was 57 out of 100 and only 34% carried LocalBusiness structured data. That is the region. This is the tool that tells you where your own site sits in it.

Questions

Is SEO Readout really free?

Yes. There is no paid tier, no trial, no account and no licence key. It is a tool we wanted for our own work, and giving it away costs us nothing but the time it took to build.

Does it send my pages anywhere?

No. Every check runs inside the popup on your machine. We have no server for it to talk to, so we cannot see which sites you use it on. The only requests it makes go from the page to its own site: the X-Robots-Tag header, the page's own HTML before JavaScript runs, robots.txt and llms.txt.

Why does it ask for the scripting permission?

That is the permission Chrome uses to run the reader in the tab you clicked on, and to draw the outline when you click a finding to see it on the page. Paired with activeTab, it means nothing runs until you click the icon, and it only ever runs on that one tab. It does not ask to read every site you visit, and it never will.

How is the score worked out?

Four modules, each out of 100: Basics (title, description, canonical, indexability, the page before JavaScript, AI crawlers), Structure (headings, images, text, links), Schema (including local details) and Social. A fail costs its full weight, a warning half. They combine as 40, 25, 20 and 15 per cent. A noindex page does not lose points; it gets a banner instead, because a page can be perfectly built and still deliberately hidden.

Does it work on WordPress, Shopify, Wix and Squarespace sites?

It works on any page open in Chrome, whatever built it. It reads what the browser rendered, the way a visitor sees it, and then fetches the same page as plain HTML to show what a crawler that does not run JavaScript gets. A page that only fills itself in with JavaScript is flagged.

Get it. It is on the Chrome Web Store, free, no account. Click Add to Chrome, open any page, click the icon.

← All work