Small tool, one job, done properly

Why this tool exists

Anyone who has published on the web has had the same small humiliation. You write something you are proud of, you post the link, and the preview that appears is a grey rectangle with the words "Home | My Site" next to it. The post is already live. People are already scrolling past it. By the time you notice, the moment has gone.

The tools that exist to prevent this are mostly owned by the platforms themselves. Facebook has a Sharing Debugger, X has a Card Validator, LinkedIn has a Post Inspector. They all work, and they are all worth using — but each one only shows you its own platform, several of them ask you to sign in to a developer account first, and none of them tells you what Google will do with the same page. Checking one URL properly means opening four tabs and logging into two of them.

Weblytic collapses that into one box. Paste a URL, see all four previews at once, and get a plain-language list of what is missing. That is the entire product. There is no dashboard, no upsell, no free tier with a locked feature behind it.

Who is behind it

Weblytic is built and maintained by a small independent team that spends most of its time on the practical end of the web: building WordPress sites, running blogs, publishing on YouTube, and helping other people do the same. This tool started as an internal script used to spot-check client sites before launch. It was useful enough often enough that it seemed worth cleaning up and putting online.

That background shapes the tool. It is written for the person who has to ship something this afternoon, not for the person writing a specification. The warnings tell you what to do, not just what is wrong. The image size guide gives you one number that works everywhere instead of a table of eleven exceptions.

How the tool actually works

It is worth being transparent about this, because "paste a URL and we analyse it" can mean a lot of different things.

When you submit a URL, your browser sends it to a small PHP script on our server. That script validates the address, refuses anything that is not an ordinary public http:// or https:// page, and then downloads the raw HTML using a standard request with an honest user-agent string. The HTML comes back to your browser as JSON, and from that point everything happens on your own machine: your browser parses the document, pulls out the meta tags, and draws the preview cards.

The server step exists for one reason only. Browsers are not allowed to read another website's HTML directly — the same-origin policy blocks it, and rightly so. A tiny server-side proxy is the standard way around that limitation. It is not doing anything clever, and it is not sending your URL anywhere else.

There is no database. The fetched page is not stored, the URL is not logged against you, and no account is created. The only thing our server writes to disk is a short-lived counter that stops a single IP address from making hundreds of requests a minute.

What we believe about tools like this

Free should mean free

A utility that takes half a second of server time should not need a subscription. Weblytic is supported by advertising, which is a fair trade for something you will use for thirty seconds and close. There is no paid tier holding back the useful part.

No account for a stateless task

Checking a meta tag does not require knowing who you are. Asking for an email address in exchange for it is a marketing decision dressed up as a product one, and we would rather not.

Explain, do not just flag

"Missing og:image" is a fact. "This is the single biggest one — a link with no image gets a fraction of the clicks, add a 1200×630 PNG" is advice. A tool that only does the first leaves you to go and search for the second, which rather defeats the point.

Work everywhere

The whole site is plain HTML, CSS and vanilla JavaScript with one small PHP file. No framework, no build step, no npm install, nothing that will break in three years when a dependency is abandoned. It loads fast on a bad phone connection because there is barely anything to load.

What is on the roadmap

We add things when they are genuinely useful rather than to have something to announce. The list currently under consideration includes checking several URLs at once, previews for Pinterest and Discord, a WhatsApp card mock, and a shareable report link so you can send a client the result instead of a screenshot.

If there is something you keep wishing this tool did, tell us. Requests from people who actually use the thing are worth more than any roadmap we write on our own.

Get in touch

Bug reports, feature ideas, corrections to anything written here, or a site that fails to load when it should not — all of it is welcome. The contact page has the form and the email address, and messages get read by a person.