Most sites we audit already have schema markup on them. That’s not the problem.
The problem is that a large share of it is either broken, outdated, or quietly doing nothing at all, while the business assumes it’s working because someone set it up once and never looked again.
Schema markup is the structured data that tells search engines and AI systems what a page actually is: who published it, what it’s about, what question it answers. When it’s right, it helps a page qualify for rich results and makes it easier for AI search to extract and cite. When it’s wrong, it fails silently. Nothing breaks visibly, nothing throws an obvious error on the page, and the business has no idea anything’s off.
These are the schema markup mistakes we run into again and again, roughly in order of how often we see them.
Still Using FAQPage and HowTo Schema That No Longer Does Anything
This is the newest common mistake, and it catches a lot of people who did everything “right” a couple of years ago.
Google retired most FAQPage and HowTo rich results for regular sites. Those enhanced FAQ accordions and step-by-step results in search now only show for a narrow set of site types, mostly government and health.
Plenty of sites still ship FAQPage schema on every blog post and service page, expecting rich results that simply won’t appear anymore. The markup isn’t harmful, but it’s dead weight, and worse, it gives a false sense that “the SEO is handled” when that particular piece stopped doing anything.
The FAQ content itself is still worth keeping on the page, since it helps human readers and AI extraction. It’s the schema wrapper around it that no longer earns the rich result it used to.
Schema That Doesn’t Match What’s Actually on the Page
This is the one Google actively cares about, and the one most likely to cause real trouble rather than just wasted effort.
We regularly find pages marked up as one thing while clearly being another. A blog post tagged as a Product because someone wanted star ratings. A homepage carrying Product schema. Review or rating markup for reviews that aren’t visibly on the page.
Google has gotten good at detecting this mismatch, and it doesn’t read as a harmless mistake, it reads as an attempt to game rich results. Unlike a plain syntax error, this kind of mismatch can trigger a manual action that strips rich results from the whole site until it’s fixed.
The rule we work by is simple: the schema has to describe what a visitor actually sees on that page, not what the business wishes were there.
Wrong Date Formats on datePublished and dateModified
This one is boring and extremely common, which is exactly why it’s worth naming.
Dates in schema have to follow a specific format (ISO 8601, the 2026-04-06 style). We find sites feeding in dates in a human-readable format instead, which quietly invalidates the markup.
It matters more than it used to. Freshness is a signal AI search systems weigh, and dateModified is part of how a page communicates that it’s current. A broken date field means a page can’t cleanly signal how recently it was updated, right when that signal has started to matter more.
Missing Required Properties
Every schema type has properties Google treats as required, and leaving one out disqualifies the whole block from rich results, often without an obvious warning until someone goes looking.
The classic version is a product marked up without a price currency, or an Article missing a valid publisher logo. Everything looks fine in the code at a glance. The rich result just never shows up, and weeks pass before anyone connects the two.
This is why “we added schema” and “the schema works” are genuinely different claims. The only way to know which one is true is to run the page through validation, not to assume the plugin handled it.
Relying Only on a Plugin and Never Validating
Plugins like Rank Math or Yoast handle a lot of schema automatically, which is genuinely useful. The mistake is treating that as the finish line.
Automatic schema still breaks in specific ways: a CMS update changes an output, a plugin ships FAQPage markup that no longer earns anything, an eCommerce plugin adds Product schema to category pages where it doesn’t belong. The plugin did its job as configured. The output is still wrong.
We treat validation as a required step, not an optional one. Google’s Rich Results Test shows what a page is eligible for, and the Schema Markup Validator catches structural issues the first tool sometimes misses. Running a page through both takes a couple of minutes and is the only thing that turns “we set up schema” into something verified.
Thin or Missing Organization Schema
This one used to be minor and has quietly become important because of how AI search works.
Organization schema, with proper links out to a business’s real profiles (LinkedIn, its other platforms, anywhere it has a verifiable presence), is part of how AI systems confirm a business is a real, identifiable entity. That verifiable identity feeds directly into the credibility signals AI models weigh when deciding whether to cite a source.
A site with thin or missing Organization schema isn’t breaking any rules. It’s just handing AI systems less to work with when they’re deciding whether this business is a trustworthy source to pull from, at exactly the point where that decision has started to matter for visibility.
Why These Schema Markup Mistakes Slip Through in the First Place
Almost none of these are dramatic failures. That’s the pattern underneath all of them.
Schema breaks silently. A page keeps loading, the content still reads fine, and nothing on screen announces that the structured data underneath is invalid or outdated. Search Console will flag some of it, eventually, but only if someone’s actually watching the enhancement reports, and most businesses aren’t.
Which is why we treat schema as something to re-check on a schedule, not set once and trust forever. Most schema markup mistakes aren’t introduced by a person editing the page, a CMS update, a plugin change, or a shift in what Google supports can turn working markup into dead markup on its own. The sites that stay clean are the ones where someone actually looks, on purpose, every few months.
Frequently Asked Questions
Can broken schema markup hurt rankings, or does it just fail to help?
For plain errors like a wrong date format or a missing property, the markup simply won’t produce a rich result, there’s no direct ranking penalty. The exception is schema that deliberately misrepresents a page’s content, which can trigger a manual action that removes rich results across the whole site until it’s resolved.
How often should a site’s schema actually be checked?
A review every few months is a reasonable baseline for most sites, plus a check any time the CMS or a major plugin gets updated, since those are the moments working markup tends to break quietly. Sites that change content or products frequently benefit from checking more often.
Is it better to have no schema than incorrect schema?
For the misrepresentation case, yes, no schema is safer than markup that misdescribes a page, because the second can be penalized while the first just does nothing. For honest-but-incomplete schema, it’s less either/or: incomplete markup that accurately reflects the page is usually worth fixing rather than removing.
