Structured data is certainly not optional for a serious scientific website.
It’s the language “tag” that search engines use to understand your content and index it correctly. For websites specializing in life sciences, where accuracy, provenance, and discoverability are essential, implementing schema.org markup is important.
Why is this important, and how can it help life science websites?
Schema.org structured data helps machines understand the meaning of web pages rather than guessing from plain text. This is important because:
- Search engines index pages more easily when they contain clear structured data and metadata.
- Users find content faster: structured data improves click-through rates and helps users access the exact resource they need (protocol, dataset, training).
Some tags to use
- Organization: for societies, labs, companies, and directories. Include name, url, logo, sameAs (social profiles), and contactPoint.
- WebSite and WebPage: site level and page level metadata; use mainEntity to link a page to its primary entity.
- ScholarlyArticle: for research summaries, white papers, and article pages; include author, datePublished, isPartOf (journal), and citation.
- Dataset: for data releases, supplementary files, and public datasets; include name, description, creator, distribution (file formats, access URLs), and license. This is essential for research reproducibility and discovery.
- Event: for conferences, webinars, and training; include startDate, endDate, location, and organizer.
- FAQPage and HowTo: for practical guidance, protocols, and site help; these can generate rich snippets that improve visibility.
Practical implementation steps
1. Audit your content and map types
Inventory pages and assign the most specific schema type.
Example mapping for LifeScience.directory:
- Directory home → WebSite + Organization (directory as an organization)
- Listing page for a society → Organization + WebPage
- Article about SEO or training → ScholarlyArticle or Article
- Resource pages with datasets → Dataset
- Events and webinars → Event
2. Use JSON LD (recommended)
Major search engines prefer JSON LD embedded in the <head> or just before </body>. It’s relatively easy to author and maintain.
Common traps and how to avoid them
- Over marking: Don’t add irrelevant schema types. Use the most specific type that fits the content.
- Outdated metadata: Keep dateModified and version up to date for datasets and articles.
- Broken URLs in contentUrl: Ensure all distribution links are accessible and use HTTPS.
- Missing provenance: Always include creator or author with affiliation for scientific credibility.
- Ignoring validation: Run structured data tests after every deployment.
Ressources
- Schema.org documentation (Organization, Dataset, ScholarlyArticle, Event, HowTo, FAQPage).
- Google Search Central: Structured data overview and rich result types.
- FAIR principles and dataset metadata best practices for research data.
- Google Rich Results Test and Schema Markup Validator for validation.
- How to make your data FAIR?