Should you ditch Data Highlighter for JSON-LD?

Quick SEO Quiz

Test your SEO knowledge in 3 questions

Less than 30 seconds. Find out how much you really know about Google search.

🕒 ~30s 🎯 3 questions 📚 SEO Google

Official statement

Google advises implementing JSON-LD or schema.org markup directly on the site for the long term to enable better understanding and usage by various search engines, rather than relying on Data Highlighter.

16:40

🎥 Source video

Extracted from a Google Search Central video

⏱ 55:57 💬 EN 📅 03/04/2020 ✂ 23 statements

Watch on YouTube (16:40) →

✂ Other statements from this video 22 ▾

📅

Official statement from April 3, 2020 (6 years ago)

⚠ A more recent statement exists on this topic Why does Google recommend JSON-LD over Microdata or RDFa for structured data imp... Ryan Levering · August 23, 2022 View statement →

TL;DR

Google recommends replacing Data Highlighter with native structured data markup (JSON-LD or schema.org) directly in the code. While Data Highlighter remains a handy testing tool, it does not guarantee longevity or multi-engine compatibility. In practical terms? Migrate to a clean technical implementation that will remain under your control and work across all search engines.

What you need to understand

What exactly is Data Highlighter?

Data Highlighter is a free Google Search Console tool that allows for visual structuring of data without altering the site’s code. You point your mouse, identify elements (prices, reviews, event dates), and Google registers this mapping on the server-side.

The concept is appealing for beginners or for sites where editing HTML code is challenging. The problem is that this approach creates total dependency on Google. If the tool disappears or if Google decides not to honor these mappings anymore, everything is lost. Worse, no other search engine (Bing, Yandex, DuckDuckGo) can utilize these annotations — they exist solely within the Google ecosystem.

Why is Google pushing for JSON-LD?

JSON-LD is a structured data markup format that fits directly into the <head> or <body> of a page. Unlike HTML microdata or RDFa, it remains isolated from the visible DOM, which limits the risks of errors and facilitates maintenance.

Google has favored this format for several years because it is easier to parse, less prone to malformations, and perfectly compatible with Schema.org — the shared standard vocabulary among all engines. Implementing JSON-LD means your markup works everywhere, not just on Google. It’s a portable and sustainable approach.

Is Data Highlighter still usable today?

Yes, the tool has not been officially removed. But Mueller's recommendation is clear: it's a temporary crutch, not a production solution. If your site regularly generates content (blog, e-commerce, directory), maintaining mappings manually becomes unmanageable.

Moreover, Google does not guarantee that Data Highlighter will continue to exist indefinitely. Search Console tools evolve — some disappear (remember the old structured data testing tool). Building an SEO strategy on a Google proprietary tool is a technical risk that can be avoided.

Data Highlighter: visual Search Console tool, Google dependency, no multi-engine portability
JSON-LD: native markup in the code, compatible with all engines, automatable maintenance
Official recommendation: migrate to a sustainable technical implementation for the long term
Data Highlighter remains a testing or troubleshooting tool, not a sustainable strategy
JSON-LD is easier to debug, version, and integrate into a modern CMS

SEO Expert opinion

Is this recommendation consistent with field practices?

Absolutely. Sites that use Data Highlighter heavily encounter maintenance issues as soon as they exceed a few dozen pages. Modifying an existing mapping, adding new content types, or correcting an error identified by the validator quickly becomes time-consuming.

Serious SEO agencies have massively migrated to JSON-LD since 2018-2019. Modern CMSs (WordPress via plugins like Yoast/Rank Math, Shopify, PrestaShop) generate JSON-LD automatically. Data Highlighter was never designed to scale — it was a tool for evangelization to popularize structured data among small sites.

What concrete limitations does Data Highlighter impose?

First limitation: partial coverage. You must manually map each type of page, and if your HTML structure evolves, everything breaks. Sites with dynamic templates (facets, filters, complex pagination) simply cannot use Data Highlighter effectively.

Second limitation: zero control over the Schema.org vocabulary used. Google chooses the implementation; you cannot refine properties, add complex relationships (nested @types), or test experimental types. With homemade JSON-LD, you control every property and can finely adapt your markup to the specifics of your content.

Are there cases where Data Highlighter remains relevant?

Yes, in two specific situations: legacy sites where modifying the code costs a fortune (old proprietary CMSs, overwhelmed technical teams), and quick tests before industrialization. If you want to verify how Google interprets a new type of content before coding, Data Highlighter remains a good sandbox.

But be careful: do not confuse testing with production. If your test via Data Highlighter works, the next step should be an implementation in JSON-LD. Leaving Data Highlighter running in production long-term creates a technical debt that will eventually cost you dearly — in time, fragility, and missed opportunities (Bing compatibility, future Schema.org developments).

Practical impact and recommendations

How to properly migrate from Data Highlighter to JSON-LD?

Start by auditing what Data Highlighter is currently marking up. In Search Console, go to Appearance in Search results → Data Highlighter, and list all the sets of marked-up pages. Note the types used (Article, Product, Event, LocalBusiness…) and the mapped properties.

Next, generate the equivalent JSON-LD. If you are using a modern CMS, activate or configure an existing plugin. If you are coding manually, use the Schema.org validator (validator.schema.org) and the Google rich results testing tool (search.google.com/test/rich-results) to validate each template before deployment. Once the JSON-LD is in place and validated, gradually disable Data Highlighter: start with one category of pages, check in Search Console that the structured data is correctly reported via the dedicated report, and then extend to the entire site.

What mistakes to avoid during migration?

Classic mistake: allowing Data Highlighter and JSON-LD to coexist on the same pages. Google could receive contradictory or duplicate signals, generating warnings in Search Console. Choose one system, only one, per type of page.

Another trap: not validating the JSON-LD before going live. Google's validator generates errors that you would never see with Data Highlighter (missing required properties, improperly nested types, non-compliant values). Poorly formatted JSON-LD can completely prevent Google from utilizing your structured data — while with Data Highlighter, Google “guessed” and corrected some errors in the backend. With native code, you are responsible for the quality.

Should you implement all available Schema.org types?

No. Focus on types that trigger eligible rich results in your niche: Product (rich snippets price/reviews), Recipe, Article, Event, FAQ, HowTo, VideoObject… Google publishes official documentation listing supported types for rich results — that’s your starting point.

For other types (Organization, WebSite, BreadcrumbList), implement them if they provide a clear semantic benefit (knowledge graph, sitelinks search box, breadcrumbs in SERPs). But don’t make it a religion: it’s better to have 3 types perfectly implemented and maintained than 15 types partially functional. JSON-LD should be actionable and measurable, not a blind checklist.

Audit current Data Highlighter sets and note types + properties used
Generate equivalent JSON-LD via CMS plugin or custom script, then validate with Google and Schema.org tools
Deploy gradually (by category of pages) and monitor the Structured Data report in Search Console
Disable Data Highlighter once migration is validated to avoid contradictory signals
Prioritize testing types eligible for rich results (Product, Recipe, Article, FAQ…)
Automate JSON-LD generation in templates to ease long-term maintenance

Migrating from Data Highlighter to JSON-LD represents a significant technical undertaking, especially for sites with dozens of templates or complex architectures. If your team lacks expertise in structured data or if development resources are limited, hiring a specialized SEO agency can significantly speed up the process while avoiding costly errors. Targeted support allows for the proper industrialization of the markup, integration into publication workflows, and ensuring maximum compatibility with future Schema.org developments.

❓ Frequently Asked Questions

Data Highlighter va-t-il disparaître prochainement ?

Google n'a pas annoncé de date de fin de vie, mais la recommandation officielle de Mueller indique clairement que l'outil n'est pas pensé pour du long terme. Mieux vaut anticiper et migrer vers du JSON-LD plutôt que d'attendre un éventuel retrait brutal.

Le JSON-LD est-il plus difficile à mettre en place que Data Highlighter ?

Techniquement oui, car il nécessite de modifier le code. Mais les CMS modernes proposent des plugins qui automatisent tout. Une fois en place, le JSON-LD est beaucoup plus facile à maintenir et à scaler que des mappings manuels Data Highlighter.

Peut-on utiliser Data Highlighter et JSON-LD en parallèle ?

Non recommandé. Cela crée des risques de signaux contradictoires et complique le debugging. Choisis une méthode unique par type de page pour éviter les erreurs dans Search Console.

Les autres moteurs de recherche comprennent-ils Data Highlighter ?

Non, Data Highlighter est un outil propriétaire Google. Seul le balisage natif (JSON-LD, microdonnées, RDFa) est exploitable par Bing, Yandex et les autres moteurs.

Faut-il supprimer immédiatement tous mes mappings Data Highlighter ?

Non, procède par étapes. Implémente d'abord le JSON-LD sur une catégorie de pages, vérifie dans Search Console que tout fonctionne, puis désactive progressivement Data Highlighter au fur et à mesure de la migration.

🏷 Related Topics

JSON-LD Data Highlighter données structurées Schema.org rich snippets balisage sémantique Search Console résultats enrichis

Domain Age & History Structured Data JavaScript & Technical SEO

🎥 From the same video 22

Other SEO insights extracted from this same Google Search Central video · duration 55 min · published on 03/04/2020

🎥 Watch the full video on YouTube →

Related statements

« Previous

Mobile-first indexing: SEO Focus...

Site Indexing by Google...

« Back to results