Queen of San Diego — Tech Blog

Under the hood at queenofsandiego.com

```html

Injecting Structured Data Into Concert Event Pages: A Multi-Site JSON-LD Deployment Strategy

Over the past development session, we identified a critical SEO gap across our event subdomain portfolio: 12 active concert pages lacked any structured data markup. This meant search engines couldn't reliably extract event details, venue information, or review aggregates. We built and deployed a structured data injection pipeline that added Event and LocalBusiness JSON-LD to all concert pages, then synced them across S3 and invalidated CloudFront caches to ensure immediate indexing.

The Problem: Missing Structured Data at Scale

Our event subdomain architecture spans multiple independent concert sites:

  • paulsimonradyshell.com
  • sailjada.queenofsandiego.com
  • 6 additional event-specific subdomains

A structured audit revealed that none of these pages contained Event schema or LocalBusiness markup. While the HTML was semantically clean and the UX was strong, search engines had no machine-readable way to understand event dates, ticket availability, venue coordinates, or aggregated ratings. This directly impacted click-through rates in rich snippets and knowledge panel eligibility.

Solution Architecture: Automated Injection with Idempotent Validation

We created a Python-based injection script at /Users/cb/Documents/repos/tools/inject_structured_data.py that:

  • Scans page head tags to detect existing structured data (preventing duplicate injection)
  • Generates Event JSON-LD by parsing page metadata and event-specific attributes
  • Appends LocalBusiness schema with venue coordinates and contact information
  • Preserves existing markup — only adds schema if none exists

The script uses this core logic:

def has_event_schema(html_content):
    """Check if page already contains Event or LocalBusiness JSON-LD"""
    return '"@type":"Event"' in html_content or '"@type":"LocalBusiness"' in html_content

def inject_event_schema(html_content, event_data):
    """Insert Event JSON-LD before closing head tag"""
    schema = generate_event_jsonld(event_data)
    return html_content.replace('', f'{schema}')

This idempotent approach meant we could safely re-run the script without fear of duplicate schemas stacking up.

Infrastructure: S3 Buckets, CloudFront, and Cache Invalidation

Each event subdomain uses a dedicated S3 bucket and CloudFront distribution:

  • Bucket naming convention: paulsimonradyshell-production, sailjada-events, etc.
  • CloudFront distribution IDs: Stored in configuration, matched per subdomain
  • Deployment workflow:
    1. Inject structured data into local HTML files
    2. Sync updated pages to S3 using aws s3 sync
    3. Invalidate CloudFront cache for affected paths

Example sync command (no credentials shown):

aws s3 sync /Users/cb/Documents/repos/sites/sailjada.queenofsandiego.com/ \
  s3://sailjada-events/ \
  --delete \
  --exclude ".git/*" \
  --exclude "*.pyc"

Post-sync, we invalidated the root and all concert pages:

aws cloudfront create-invalidation \
  --distribution-id E2ABC1234XYZ \
  --paths "/*" "/concert-pages/*"

CloudFront invalidations typically propagate within 60 seconds, ensuring that subsequent crawls by Google and Bing fetch the new structured data immediately rather than serving stale versions.

Data Modeling: Event and LocalBusiness Schemas

The injected JSON-LD follows Google's Event schema specification:

{
  "@context": "https://schema.org",
  "@type": "Event",
  "name": "Concert Name",
  "description": "Full event description from page metadata",
  "startDate": "2024-06-15T19:30:00-07:00",
  "endDate": "2024-06-15T22:00:00-07:00",
  "location": {
    "@type": "Place",
    "name": "Venue Name",
    "address": {
      "@type": "PostalAddress",
      "streetAddress": "123 Music St",
      "addressLocality": "San Diego",
      "addressRegion": "CA",
      "postalCode": "92101",
      "addressCountry": "US"
    },
    "geo": {
      "@type": "GeoCoordinates",
      "latitude": "32.7157",
      "longitude": "-117.1611"
    }
  },
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.9",
    "ratingCount": "63"
  }
}

Embedding both Event and LocalBusiness schemas in the same page allows search engines to understand not just when and where the concert happens, but also to surface venue reviews and ratings in a unified rich snippet.

Deployment Results and Validation

We successfully injected structured data into 12 concert pages across the event subdomain portfolio. Post-deployment checks confirmed:

  • All 12 pages now contain valid Event JSON-LD
  • No duplicate schemas (idempotent injection worked as intended)
  • Existing page markup and styling remained untouched
  • CloudFront cache invalidations completed within standard TTL windows

We validated the markup using the Schema.org Validator to ensure compliance before pushing to production.

Why This Approach?

Automation over manual edits: With 12 pages across multiple repos and S3 buckets, a one-time manual approach would have been error-prone and difficult to audit. The script-based approach is repeatable and auditable.

Idempotent logic: By checking for existing schemas before injecting, we can safely re-run the script if new events are added, without breaking deployed pages.

Immediate cache invalidation: Waiting for CloudFront's standard 24-48 hour TTL would delay indexing by days. Explicit invalidation ensures Googlebot picks up the new markup within the hour.

What's Next

With structured data now live, the next phase focuses on distribution channels. The CMO assessment identified 3,676 emailable contacts, 157 aggregated reviews, and strong organic brand presence that aren't being leveraged. Upcoming work includes:

  • Automated email campaigns targeting inactive subscribers with event highlights
  • Google Business Profile optimization to surface events in local search results
  • OTA platform integration (Ticketmaster, Eventbrite) to expand reach beyond organic

The structured data we deployed today