Schema Markup in Singapore (2026): The Practical SEO Guide

Last updated: 23 June 2026 · By Adrian Tan, Singapore Digital Marketing

Schema markup is one of the few parts of SEO that is genuinely a checklist item — get it right once, maintain it, and it quietly works in the background. But 2026 changed the why. Schema used to be mostly about winning pretty “rich results” in Google. This year Google removed one of the most popular rich results entirely, while structured data simultaneously became more important than ever — because it is now the language that AI search engines read to understand and cite your business.

This guide explains what schema markup actually is, what changed in 2026, which types still earn results, and how Singapore businesses should use it — with concrete, copy-ready examples. It is a technical topic, so we keep the jargon defined in passing and the steps practical.

What schema markup is (in plain English)

Schema markup (also called structured data) is a standardised block of code you add to a web page that tells search engines what the content means, not just what it says. A human reads “Dr Tan, 4.8 stars, open until 9pm, Orchard Road” and understands it instantly. A search engine sees a wall of text. Structured data hands it a labelled summary: this is a LocalBusiness; its name is X; its rating is 4.8; its opening hours are Y; its address is Z.

The vocabulary comes from Schema.org, a shared standard backed by Google, Microsoft, Yahoo and Yandex. You don’t invent the labels — you use the agreed ones (Organization, Product, LocalBusiness, Article, and so on). The recommended format is JSON-LD: a small script block in the page’s code, separate from the visible content, so it never affects how your page looks.

You do not strictly need schema to rank — but without it you are forcing search engines and AI assistants to guess at meaning they could simply be told. In a market as competitive and mobile-first as Singapore — where 98.4% of the population is online (DataReportal, Digital 2026: Singapore) — handing the machines clean data is cheap leverage.

What changed in 2026 (read this before you do anything)

Two things shifted, and they pull in opposite directions.

1. Google removed FAQ rich results. As of 7 May 2026, FAQ rich results no longer appear in Google Search. Google is also retiring the FAQ search-appearance report, dropping FAQ support from the Rich Results Test, and removing it from the Search Console API later in 2026 (Search Engine Journal). This follows the earlier 2023 change that had already limited FAQ rich results to a narrow set of sites.

What this means in practice: if your reason for adding FAQ schema was the expandable Q&A boxes in Google results, that benefit is gone. However, the FAQPage type itself is still valid Schema.org markup — it validates without errors, it is harmless to keep, and crawlers for Bing, Perplexity and the various AI/retrieval systems still read it. So FAQ schema moved from “SERP feature” to “AI-readability signal”. Keep it for the latter; don’t expect SERP boxes from it.

2. Structured data became central to AI search. Google’s AI Overviews now appear on a large share of searches, and AI assistants (Gemini, ChatGPT, Perplexity, Claude) increasingly answer queries by reading and citing the open web. These systems lean on structured data to identify entities — who you are, what you sell, where you operate — and to decide whom to cite. JSON-LD is, in effect, the common machine-readable format they all consume. The upshot: schema is no longer just about a rich snippet on page one; it is about being legible to every system that might recommend you.

So the honest 2026 framing is: do schema for entity clarity and AI legibility first, and for the rich results that still exist second.

Which schema types still earn rich results in 2026

Plenty of structured data still produces visible enhancements in Google Search. The types worth prioritising:

A Singapore-relevant note on Review/AggregateRating: Google’s policies only allow review markup for reviews collected and shown on your own site, and self-serving markup (a business marking up its own glowing rating) can be ignored or penalised. Don’t fake it. For local reputation, your real lever is genuine Google reviews on your Google Business Profile, not schema tricks.

A worked example: LocalBusiness schema for a Singapore clinic

Here is the kind of JSON-LD a local Singapore business should run sitewide (illustrative — replace the values with your own):

{
  "@context": "https://schema.org",
  "@type": "MedicalClinic",
  "name": "Example Aesthetic Clinic",
  "image": "https://example.sg/clinic.jpg",
  "@id": "https://example.sg",
  "url": "https://example.sg",
  "telephone": "+65 6123 4567",
  "priceRange": "$$",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "10 Orchard Road #05-01",
    "addressLocality": "Singapore",
    "postalCode": "238888",
    "addressCountry": "SG"
  },
  "geo": { "@type": "GeoCoordinates", "latitude": 1.3039, "longitude": 103.8318 },
  "openingHoursSpecification": [{
    "@type": "OpeningHoursSpecification",
    "dayOfWeek": ["Monday","Tuesday","Wednesday","Thursday","Friday"],
    "opens": "10:00", "closes": "20:00"
  }]
}

Notice the SG-specific fields: addressCountry: "SG", a +65 telephone, a Singapore postal code, and accurate opening hours. Keep this consistent with your Google Business Profile and website footer — schema is one more place your NAP must match, or you send mixed signals.

How to add schema to your site (three routes)

1. A plugin (easiest for most SMEs). On WordPress, SEO plugins like Yoast, Rank Math or AIOSEO generate Organization, Article, breadcrumb and local schema automatically once configured. On Shopify, most modern themes and apps output Product schema. This covers 80% of needs with near-zero code.

2. Google Tag Manager. You can inject JSON-LD via a custom HTML tag — useful when you can’t easily edit templates. It works, but it’s harder to keep accurate and to debug, so use it as a fallback rather than a default.

3. Hand-coded JSON-LD in the template. The most robust route: place the JSON-LD block directly in the page <head> or body. Best for custom builds and for types your plugin doesn’t cover well. This is what we do on built-for-SEO sites (see building a website for SEO in Singapore when live, and our technical SEO basics).

Whichever route you choose, always validate with Google’s Rich Results Test and the Schema.org validator, then watch the Enhancements reports in Google Search Console after setup (see Google Search Console setup for Singapore).

A second worked example: Product schema for Singapore e-commerce

If you sell online, Product plus Offer schema is the one that pays — it can surface price, currency and availability directly in results. Get the Singapore-specifics right (illustrative):

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Handwoven Rattan Pendant Light",
  "image": "https://example.sg/products/rattan-light.jpg",
  "description": "Handmade rattan pendant light, designed in Singapore.",
  "sku": "RPL-001",
  "brand": { "@type": "Brand", "name": "Example Home" },
  "offers": {
    "@type": "Offer",
    "url": "https://example.sg/products/rattan-light",
    "priceCurrency": "SGD",
    "price": "189.00",
    "availability": "https://schema.org/InStock",
    "itemCondition": "https://schema.org/NewCondition"
  }
}

The make-or-break fields are priceCurrency: "SGD", a price that matches the live page exactly, and an accurate availability. A price in the schema that differs from the page price is a guideline breach and will get the markup ignored. Most Shopify and WooCommerce setups generate this automatically — your job is to verify it’s accurate and uses SGD.

Schema for AI search (the 2026 reason it matters)

Beyond rich results, structured data is now how AI systems build a confident picture of your business — what’s called entity clarity. When Gemini, ChatGPT, Perplexity or Claude decide whom to cite for “best [service] in Singapore”, they favour sources whose identity, offerings and location are unambiguous. Schema is the cleanest way to remove that ambiguity. Practical moves that help:

• One consistent Organization entity sitewide, with sameAs links to your official social and directory profiles — this stitches your identity together across the web.
• @id references so your pages point to the same entity rather than describing a new one each time.
• Author markup on articles (Article/BlogPosting with a named author) — this supports E-E-A-T signals that both Google and AI systems weigh.
• Accurate LocalBusiness data matching your Google Business Profile — consistency across schema, GBP and your site is what makes an AI system trust the data.

None of this is exotic. It’s the same clean, accurate structured data you’d build for Google — it just now does double duty.

One caveat worth adding. Schema is an entity-clarity and rich-results investment, not an AI visibility lever: an Ahrefs study of 1,885 pages that added JSON-LD found no citation lift in AI Overviews, AI Mode or ChatGPT, and Google states there is no special structured data needed to appear in its AI features. Do it for the reasons above. We set out what actually drives citation in GEO vs AEO vs LLM SEO.

How to test and validate (the workflow)

Never ship schema you haven’t validated. The workflow:

1. Rich Results Test (Google) — confirms whether a page is eligible for the rich results that still exist, and flags errors. Note: it no longer tests FAQ.
2. Schema.org Validator — checks your markup against the full vocabulary, catching type and property errors the Google tool ignores.
3. Google Search Console → Enhancements — after deployment, GSC reports valid/invalid items per type over time. This is where you catch issues at scale, not one page at a time (see Google Search Console setup for Singapore).
4. Re-validate after any template or plugin change — CMS updates silently break schema more often than you’d think.

The rules that keep schema safe (Google’s guidelines)

Structured data has policies, and breaking them can get markup ignored or trigger a manual action (Google, structured data guidelines):

• Mark up content that is actually visible on the page. Don’t schema things users can’t see.
• Be accurate and current. No fake ratings, no outdated prices, no events that already happened.
• No self-serving review markup. Reviews must be genuinely collected, not invented.
• Match the page’s purpose. Don’t put Recipe schema on a service page.
• Keep it complete. Fill required and recommended properties; partial markup may not qualify.

These aren’t suggestions — they’re the line between schema helping you and schema getting your enhancements pulled.

Common schema mistakes Singapore SMEs make

The errors we find most often when auditing local sites:

1. Conflicting markup from two sources. A theme outputs Organization schema and so does the SEO plugin — now there are two different entities describing the same business. Pick one source per type and disable the other.
2. Schema that doesn’t match the page. Prices, hours or ratings in the JSON-LD that drifted out of sync with the visible content. Google requires them to match; mismatches get the markup ignored.
3. LocalBusiness data that contradicts the Google Business Profile. A different phone format or address than your GBP. Consistency across schema, GBP and your footer is the whole point — contradiction undoes it.
4. Self-serving review markup. A business inventing its own AggregateRating. This is against policy and can trigger a manual action. Earn real Google reviews instead.
5. “Set and forget” after a CMS update. A plugin or theme update quietly breaks the markup and nobody re-validates for months. Re-test after every significant change.
6. Over-marking up. Stuffing every conceivable type onto a page in the hope of more rich results. Mark up what’s genuinely on the page and relevant to its purpose — nothing more.

Most of these come down to one principle: schema must be a true, consistent description of the page. Treat it as documentation of reality, not a growth hack, and it stays safe.

A sensible 2026 priority order for Singapore businesses

If you’re starting from zero, do it in this order:

1. Organization schema sitewide — establish your brand entity.
2. LocalBusiness (the right subtype) on your contact/location pages — feed local + AI.
3. BreadcrumbList across the site — easy win, near-universal eligibility.
4. Product/Offer if you sell online — prices in SGD, availability.
5. Article/BlogPosting on your blog — author, dates, entity clarity.
6. FAQPage where genuinely useful — for AI legibility, not SERP boxes.
7. Validate everything, then monitor Search Console monthly.

Done properly, schema is a one-time build plus light maintenance that makes your whole site more legible to Google and to the AI assistants now shaping how Singaporeans discover businesses. It won’t, on its own, lift you up the rankings — but it removes a layer of guesswork that your competitors are still leaving on the table.

[CTA] Want your structured data audited and implemented correctly — so you’re eligible for every result you should be, on Google and in AI answers? See our SEO service in Singapore or talk to us. For the bigger picture, start with our SEO guide.

FAQ

Does schema markup improve my Google rankings directly?

Not directly. Schema doesn’t move you up the rankings by itself, but it makes you eligible for rich results, clarifies your content for Google, and increasingly helps AI search engines understand and cite you — all of which can lift visibility and click-through.

Is FAQ schema still worth adding in 2026?

The SERP benefit is gone — Google removed FAQ rich results on 7 May 2026. But FAQPage JSON-LD is still valid and is read by Bing, Perplexity and AI crawlers, so it retains value for AI legibility. Keep existing FAQ schema; just don’t add it expecting Google rich-result boxes.

What format should I use — JSON-LD or microdata?

JSON-LD. Google recommends it, it’s the cleanest to maintain, and it’s the format AI search engines consume. Microdata still works but is harder to manage.

Do I need a developer to add schema?

Usually not. WordPress (Yoast, Rank Math, AIOSEO) and most Shopify themes generate the common types automatically. Hand-coded JSON-LD or Tag Manager is only needed for custom types or custom builds.

Can schema markup get me penalised?

Yes, if you break Google’s guidelines — fake reviews, marking up hidden or inaccurate content, or self-serving rating markup can lead to ignored markup or a manual action. Mark up only real, visible, accurate content.

Which schema matters most for a local Singapore business?

Organization and LocalBusiness (with accurate SG address, +65 phone, postal code and hours), kept consistent with your Google Business Profile. That combination feeds both local search and AI recommendations.

Related guide

• AI SEO in Singapore — including the controlled study finding that adding schema markup did not increase AI citations, and why structured data is still worth implementing.

Related guide

If you are implementing schema partly in the hope of being cited by AI assistants, read how to rank on ChatGPT, Gemini and Perplexity first — it carries the controlled study that tested exactly that assumption, and the result argues for implementing structured data on its own merits rather than as a citation tactic.