Blog SEO with JSON-LD Structured Data

Meta tags and structured data will not save a thin post, but they do help search engines and social apps understand what a page is. For blogs, the usual win is clean titles/descriptions plus Article (or BlogPosting) JSON-LD that matches visible content. This is a practical setup guide, not a promise of rich-result magic.

Meta tags that earn the click

Tag	Practice
`title`	~50–60 characters; primary topic near the front
`meta description`	~120–160 characters; concrete outcome, not keyword soup
`canonical`	Absolute URL of the preferred copy
`og:title` / `og:description` / `og:image`	Match or thoughtfully adapt the page; real 1200×630-ish image
`twitter:card`	`summary_large_image` when you have art

Write descriptions like a human subtitle. Duplicate titles across posts dilute branding and confuse SERPs.

JSON-LD Article skeleton

Google documents preference for JSON-LD in a script tag. Minimal shape:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "BlogPosting",
  "headline": "Blog SEO with JSON-LD Structured Data",
  "description": "Practical Article JSON-LD for blogs...",
  "image": ["https://example.com/og/seo-structured-data.png"],
  "datePublished": "2025-03-01T00:00:00+09:00",
  "dateModified": "2026-07-16T00:00:00+09:00",
  "author": {
    "@type": "Person",
    "name": "Your Name",
    "url": "https://example.com/about"
  },
  "publisher": {
    "@type": "Organization",
    "name": "Your Site",
    "logo": {
      "@type": "ImageObject",
      "url": "https://example.com/logo.png"
    }
  },
  "mainEntityOfPage": {
    "@type": "WebPage",
    "@id": "https://example.com/en/blog/seo-structured-data/"
  }
}
</script>

Keep headline aligned with the visible <h1>. Mismatches look manipulative and help no one.

Dates, authors, and updates

Use ISO 8601 with timezone offset or Z. Update dateModified when you meaningfully change content — not on every typo deploy if that would mislead readers. Multi-author blogs should model the real byline; fake authors for “E-E-A-T theater” backfire.

If the page truly has an FAQ section, FAQPage schema can be appropriate. Do not dump invisible FAQs solely for stars. BreadcrumbList helps understanding site hierarchy on articles nested under /blog/....

Only mark up content users can see. Hidden schema is a policy risk.

Implementation tips for Astro/Next

Generate JSON-LD from the same frontmatter that drives the page title
Absolute URLs only — relative paths in image are common breakage
One primary Article object per post page; avoid contradictory duplicates from theme + plugin
Escape </script> inside JSON if you ever inline user text (prefer JSON.stringify)

const jsonLd = {
  "@context": "https://schema.org",
  "@type": "BlogPosting",
  headline: post.data.title,
  datePublished: post.data.pubDate.toISOString(),
  // ...
};

Validation workflow

Deploy a preview.
Run Google Rich Results Test on the URL.
Fix errors (missing fields, invalid images).
Spot-check Facebook/Slack debuggers for OG tags (cache may lag).
Monitor Search Console enhancements for ongoing issues.

What structured data will not do

Rank a hollow 200-word post above strong competitors
Replace inbound links or clear information architecture
Guarantee rich results — eligibility ≠ entitlement

Prioritize helpful content and clean HTML first; schema is documentation for machines.

Launch checklist

Unique title + description
Canonical set
OG image resolves (200 OK, adequate size)
JSON-LD validates
Visible author/date consistent with schema
Noindex only when intentional (previews, thin duplicates)

Bottom line

Treat SEO metadata and JSON-LD as a contract with crawlers and social scrapers: accurate titles, real dates, real authors, absolute image URLs. Set the template once in your site framework, validate on a sample of posts, and invest remaining energy in articles people actually want to read.

Tools