Schema won’t lift your rank. It lifts everything around it.

Not directly, and anyone claiming otherwise is overselling. What it does is more valuable in practice: it earns rich results that raise click-through from positions you already hold, it removes ambiguity about what your entities are (which increasingly decides whether AI engines cite you), and it makes your content machine-legible in a search landscape where machines mediate more of the traffic every quarter. Better CTR at the same rank plus AI citability is a stronger business case than a phantom ranking boost.

The ones with real payoffs for your model: Organization or LocalBusiness for entity clarity, Service and Product for what you sell, Article with genuine author entities for content, FAQPage where the questions are real, BreadcrumbList for navigation clarity, and Event, Job, Recipe or HowTo where they genuinely apply. Everything else is usually markup theatre. Depth matters more than breadth: a complete, accurate Organization graph beats twelve shallow types.

Firm, and learned expensively: self-serving review and AggregateRating markup on your own Organization or LocalBusiness is against Google's review-snippet policy, and in the 2026 spam-update environment it is an active liability. We have removed it from sites site-wide as recovery work, including our own. Ratings belong on third-party platforms that actually host the reviews. Any provider offering to "add star ratings to your search results" for your own business is selling you a future manual action.

Substantially: generative engines cite sources whose facts are unambiguous and corroborated, and structured data is the cleanest way to state who you are, what you offer, where you operate, what things cost and who wrote what. Entity consistency across your site and external profiles is the underlying goal; schema is how you declare it machine-side. This is now one of the strongest practical arguments for depth in structured data.

In code where possible (templates, CMS fields, or a centralised generator so one source of truth feeds every page), then validated in rendered output rather than source intentions, using Google's Rich Results Test and Schema.org validation plus manual entity-graph review. The most common problem we fix is not absence but duplication and contradiction: a theme, a plugin and an embed each emitting their own Organization block with different details.

Practically always: hand-coded sites get generators, WordPress gets consolidated plugin plus code, Shopify and Webflow get theme or custom-code implementation with dynamic fields, Squarespace and Wix get header and page injection. The platform decides the mechanism, not the possibility. What varies is validation discipline, and that is the part most implementations skip.

Audit and remediation projects from £600 (existing markup consolidated, errors fixed, gaps filled, entity graph made coherent). Full implementation for complex sites from £1,200. Ongoing schema governance inside retainers, because templates change and markup silently breaks on release, which is exactly how sites lose rich results without noticing.