Sail JADA — Engineering Log

Technical work on sailjada.com and the JADA booking systems

```html

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

During this session, we identified a critical SEO gap across the JADA concert event subdomain network: 12 active event pages were completely missing structured data markup. This meant search engines couldn't understand event details, dates, locations, or pricing—leaving significant organic visibility on the table. Here's how we diagnosed the problem, built an automated injection solution, and deployed it across multiple S3-backed CloudFront distributions.

The Problem: Zero Structured Data on Concert Pages

The event subdomain architecture includes sites like paulsimonradyshell.com and sailjada.queenofsandiego.com with individual event pages for each concert. A systematic audit revealed:

  • 12 active event pages across multiple subdomains
  • Zero Event schema markup (no JSON-LD)
  • Zero LocalBusiness markup for venue/organizer context
  • No date/time/pricing information parseable by search engines

Without this structured data, Google Search can't create Rich Results for event listings, Knowledge Panels don't populate with event details, and schema-based carousel snippets never appear. For a ticketed event business, this is invisible revenue loss.

Solution Architecture: Automated JSON-LD Injection Pipeline

Rather than manually editing 12 HTML files, we built a reusable injection script that:

  1. Parses event pages for existing metadata (title, date, price elements)
  2. Generates valid Event + LocalBusiness JSON-LD blocks
  3. Injects structured data into the <head> tag
  4. Deploys updated files to S3 buckets
  5. Invalidates CloudFront cache to serve fresh versions

Technical Implementation Details

Structured Data Injection Script

Created: /Users/cb/Documents/repos/tools/inject_structured_data.py

The script uses BeautifulSoup to parse HTML and inject Event schema following schema.org specifications:


# Key function structure:
def inject_event_schema(html_content, event_data):
    """
    - Parse HTML with BeautifulSoup
    - Extract or derive: name, startDate, endDate, price, priceCurrency
    - Validate against schema.org/Event requirements
    - Create JSON-LD block with proper escaping
    - Inject before closing </head> tag
    - Return updated HTML
    """

The JSON-LD payload includes:

  • @context: "https://schema.org" — Schema.org namespace
  • @type: "Event" — Event schema type
  • name, description, startDate, endDate — Core event data
  • location — Venue details as LocalBusiness or Place
  • offers — Pricing and availability
  • organizer — JADA/QOS brand reference

All datetime values use ISO 8601 format (YYYY-MM-DDTHH:MM:SS) as required by schema.org validators.

Multi-Site Deployment Pattern

Event pages live in subdomain-specific directories with their own S3 buckets and CloudFront distributions:


Directory Structure:
/Users/cb/Documents/repos/sites/
  ├─ sailjada.queenofsandiego.com/
  │  └─ ranch-and-coast.html (and other event pages)
  └─ [other-event-subdomains]/
     └─ event-pages/

Each subdomain has a dedicated AWS infrastructure stack:

  • S3 Bucket: Stores HTML, CSS, JS for that event subdomain
  • CloudFront Distribution: CDN caching layer with automatic cache invalidation
  • Route53 DNS: CNAME records pointing domain to CloudFront distribution

Infrastructure: S3 & CloudFront Invalidation

After injecting structured data, we needed to push updated files to S3 and clear the CDN cache:


# Example commands (no credentials shown):

# Sync updated files to S3 bucket for event subdomain
aws s3 sync ./updated-event-pages s3://[bucket-name]/

# Invalidate CloudFront distribution to force cache refresh
aws cloudfront create-invalidation \
  --distribution-id [DISTRIBUTION-ID] \
  --paths "/*"

Key decisions:

  • Wildcard invalidation (/*) — Since event metadata can change, we invalidate entire distributions rather than individual paths. Cost is minimal; correctness is critical.
  • Immediate deployment — Structured data affects SEO ranking immediately. No staging delay.
  • Bucket per subdomain — Isolates event sites, enables independent scaling and permission boundaries.

Why This Approach?

JSON-LD vs. Microdata vs. RDFa

We chose JSON-LD because:

  • Search engine preference — Google explicitly recommends JSON-LD for Event schema
  • Non-invasive — No need to modify existing HTML structure; inject into <head>
  • Maintainability — Easy to validate, test, and version-control as separate data blocks
  • Content separation — Keeps presentation markup (HTML) separate from semantic data (JSON-LD)

Automated Injection vs. Manual Markup

Building a script instead of hand-editing 12 files:

  • Consistency — All pages get identical schema structure and validation
  • Scalability — Adding 5 new event pages costs zero engineering time; script runs automatically
  • Correctness — No human copy-paste errors; standardized datetime formatting
  • Auditability — Script output can be version-controlled and reviewed before deployment

Validation & Testing

Before deploying, each injected page should be validated:


# Option 1: Google's Rich Results Tester
https://search.google.com/test/rich-results

# Option 2: Schema.org validator
https://validator.schema.org/

# Option 3: JSON-LD syntax check (no external calls)
python -m json.tool < extracted-jsonld-block.json

Post-deployment, monitor Google Search Console for:

  • Rich Results — Check "Enhancements" section for newly-detected Event schema
  • Indexing status — Confirm pages are re-crawled after JSON-LD injection
  • Search Appearance — Verify event date/time/price appear in SERP previews within 1-2 weeks