Schema markup is not a ranking factor.

What it actually buys you

Structured data is a translation layer. It restates what is already on the page in a format a machine can parse without inference: this string is a price, this one is an author, these are the steps of a recipe. Google uses it for two things, and ranking is not either of them.

The first is eligibility for rich results. Correct Product markup makes a listing eligible to show a price and availability, Recipe markup makes it eligible for the cooking time, Breadcrumb markup changes the URL line into a path. Eligible is the operative word, because Google decides per query whether to show the feature at all. The second is entity clarity: connecting an organisation, a person, a service and an article so a machine can tell which Acme you are and who wrote what.

Valid, and doing nothing

The most common outcome we see is markup that passes every validator and earns absolutely nothing, for one of four reasons.

It is a type with no rich result attached. Organization and WebSite markup are worth having for entity reasons, but there is no visible feature waiting for them, so a site that adds them and watches for a change will see none. It is a type whose result was withdrawn, which is what happened to FAQ markup: since 2023 Google shows FAQ rich results only for well-known government and health sites, so on most sites it earns nothing, and we removed ours rather than maintain code for a benefit that had gone.

Or it is disconnected: a pile of separate blocks with no relationships between them, so nothing resolves into a graph and the entity work does not happen. Or it contradicts the page, which is the actively harmful case. Markup describing a price, a rating or an author that a visitor cannot find in the visible content is against the structured data policies, and it is the one version of this that can genuinely cost you, up to a manual action.

The review stars problem

Ratings are where the temptation concentrates, because stars in a result are visibly valuable. They are also where the enforcement is real. Self-serving review markup on your own organisation, ratings with no genuine reviews behind them, aggregate scores assembled from nothing: all of it is against the policies, and it is exactly the pattern that gets checked. Our generator will not fabricate any, which is deliberate.

So is it worth doing

Yes, for the right reasons. Rich result eligibility is worth having wherever a relevant type exists, because the click-through difference is real even when the position is unchanged. Entity clarity is worth more than it used to be, because it is a large part of how a model decides who you are, which is the substance of LLM SEO rather than a ranking play.

And it is cheap. A connected graph implemented once in the templates keeps working, which is how we do it on client sites and on this one, where every page carries JSON-LD that validates on each build. Just hold the correct expectation: you are buying eligibility and legibility. If someone is selling schema as the reason you will outrank a competitor, they are describing a mechanism that does not exist.