What is microdata? Structured data inside your HTML explained

Microdata is one of the three formats you can use to add structured data to a web page, and it is the format that weaves that data straight into your existing HTML through a set of attributes: itemscope, itemtype and itemprop. It is the older alternative to JSON-LD, the format Google recommends today. In this article you will learn what microdata is exactly, how it works and why, as a B2B company, you rarely choose it deliberately anymore.

Microdata belongs to the wider story of structured data and schema markup, the way you tell SEO and AI models what is on your page.

What exactly is microdata?

Microdata is a specification that forms part of the HTML standard. It lets you add extra meaning to the text that is already on your page, by “labelling” that text with attributes. Instead of a separate block of code, as with JSON-LD, you stick the labels right in the middle of your visible HTML.

Like the other formats, microdata leans on the vocabulary of Schema.org, the shared list that the major search engines maintain together. In other words, microdata is the syntax (the way of writing) and Schema.org supplies the words (Organization, Product, Article, and so on). You pick a type from that list and attach properties to it.

The core idea: microdata tells a machine “this sentence is not just text, this is the name of a company” or “this number is a price”. As a result, Google has to guess a lot less about what your page is about.

How does microdata work? The three attributes

Microdata revolves around three HTML attributes that you place on ordinary elements (such as a div or a span):

• itemscope marks the start of an item: it says “everything inside here belongs together”.
• itemtype indicates which type of item it is, via a Schema.org URL, for example https://schema.org/Organization.
• itemprop names a property of that item, such as the name, the address or the price.

A simple example for a company looks like this:

<div itemscope itemtype="https://schema.org/Organization">
  <span itemprop="name">Customer Impact</span>
  <a itemprop="url" href="https://www.customerimpact.be">customerimpact.be</a>
</div>

A person just sees a name and a link. Thanks to the attributes, a search engine reads that this is an organisation, with a name and a URL. You use that same principle for articles, products, events or reviews.

Microdata versus JSON-LD and RDFa

There are three formats for structured data, and Google supports all three: JSON-LD, microdata and RDFa. Still, in practice the choice is made quickly. In its documentation on structured data, Google explicitly recommends JSON-LD, because it is the easiest to add and maintain, and because you do not break it by accident when you change your page layout.

RDFa is the third format and works in a comparable way to microdata (it weaves data into HTML, but with other attributes such as vocab and property). That one, too, is rarely used deliberately anymore.

The big drawback of microdata is precisely that it sits inside your layout. If a developer changes something in your template, the markup easily gets wiped out without anyone noticing. JSON-LD stands apart and therefore stays more stable. If you want to see how to tackle that concretely, read how to add JSON-LD and get found by ChatGPT, Claude and Google.

Where do you still come across microdata?

Microdata is not dead, it is mainly outdated in use. You still regularly run into it on:

• Older websites that were set up years ago, when microdata and RDFa were more common.
• Outdated themes and templates in CMS platforms such as WordPress, which bake the markup automatically into the HTML.
• Sites that were never migrated to JSON-LD because “it works anyway”.

And that last point is true as well: existing, correct microdata does not need to be replaced in a panic. Google still reads it. But if you are building something new, or you are finally going to set up your structured data properly, there is no reason to pick microdata.

Is microdata relevant for your B2B site?

Honestly: for most B2B companies the choice between microdata and JSON-LD is not a real discussion. Pick JSON-LD and you are done. Knowing microdata is mainly useful to recognise what is on your existing site, and to understand why a developer or an audit is talking about it.

More important than the format is that you have the right structured data at all, and only what is actually on your page. False or exaggerated markup works against you instead of for you. For a B2B company without a webshop, a handful of types is usually enough (Organization, Article, FAQPage), not every conceivable schema. How to approach that pragmatically, you can read in our guide on implementing schema markup for B2B.

And as with everything in SEO: structured data is a means, not an end. It helps Google and AI understand your page, but a pretty rich result that brings in no enquiries is a vanity metric. Steer on the leads and revenue that extra visibility delivers, not on the technology itself.

Frequently asked questions about microdata

What is the difference between microdata and JSON-LD? Microdata weaves the structured data straight into your visible HTML with attributes such as itemscope and itemprop. JSON-LD puts that same data in a separate script block, separate from your layout. Google supports both, but recommends JSON-LD because it is easier to manage.

Is microdata bad for SEO? No. Correct microdata is simply read by Google and produces the same rich results as JSON-LD. The drawback is practical: it is harder to maintain and breaks faster during a redesign, not that it harms your rankings.

Should I replace my existing microdata with JSON-LD? Not urgently, if it works correctly. If you are building something new or you are reworking your structured data anyway, JSON-LD is the logical choice. Do not replace something that works perfectly well just for the sake of the format.

Does microdata use Schema.org too? Yes. Microdata is only the way of writing; the types and properties (such as Organization or Product) come from the Schema.org vocabulary, just as with JSON-LD and RDFa.

Unsure about your structured data?

We look at what is on your site right now, whether that is microdata or JSON-LD, and tell you honestly what difference it makes for your findability in Google and in AI answers. We are a small team that moves fast and steers on leads, not on isolated technical scores. Schedule your free intake.