MOOSERANK SERVICE-BUSINESS SCHEMA OBSERVATION
Review date: October 22, 2026 (America/Winnipeg)
Article: https://mooserank.com/blog/local-business-schema-service-businesses

QUESTION
What does a small set of service-business homepages actually emit in server HTML,
and which selected statements can be compared with the public business text?

COHORT
20 unique provider domains: five Long Island handyman providers, five Long Island
HVAC providers, five Austin plumbing providers, and five Denver-area electricians.
The public JSON/CSV lists each domain, discovery query, source URL and timestamp.
This is a convenience sample selected from public searches, excluding directories
and marketing publishers. It is not random, representative, a Google top-20 list,
or a study of ranking performance. Provider affiliations were not assessed.
One plumbing-domain connection failed and was replaced. Exact attempt information
is retained in the JSON dataset. Initial local sandbox network failures do not
establish that a provider was unavailable.

COLLECTION
Read-only HTTP retrieval on October 22, 2026. Twenty homepage responses and fifteen
supplementary contact pages linked from those homepages were captured. No browser
JavaScript rendering, Google URL Inspection, Schema.org validator, Rich Results
Test, backlink data, paid keyword data, or ranking measurement was used for this
cohort. Public pages can change; the dataset preserves fetch timestamps and SHA-256
hashes. Original responses remain in our internal research archive; they are not
republished. Repeat collection may yield different observations.

COUNTING RULES
- Headline counts use one homepage per provider, never all 35 pages as a denominator.
- JSON-LD blocks are script elements with application/ld+json type. A strict JSON
  parser checked those blocks. Passing JSON syntax is not vocabulary validation.
- Typed business/organization declarations are counted even when incomplete.
  Repeated descriptions of one @id and multiple legitimate entities are not
  automatically duplicate businesses or conflicts.
- Presence metrics inspect direct properties on typed business/organization
  JSON-LD nodes. They do not resolve linked nodes or include nested contactPoint.
- Microdata/RDFa are markup indicators, not complete parsed graphs. RDFa requires
  Schema.org evidence; generic OpenGraph property attributes do not count.
- No JSON-LD observed means no JSON-LD in that response. JavaScript-injected data,
  other formats, and what Google ultimately renders remain outside this method.
- Missing optional fields are not errors; no completeness score was calculated.
- Own-business aggregateRating presence is an observation, not a finding of false
  reviews, a policy violation, or eligibility for Google review stars.

SELECTED FACT COMPARISONS
Business name, telephone, hours, location or coverage were compared where there
was a substantive claim and readable source text. Supported means agreement in
the specific retrieved sources, not independent real-world verification.
Unresolved includes no claim to compare, insufficient server text, mixed office/
emergency/answering scopes, or ambiguity in the source. It is not an error rate.
An internal schedule discrepancy means differing emitted schedules were observed;
it does not establish which schedule is correct.

FINDINGS
16 of 20 homepages had JSON-LD; 25 blocks parsed; 0 parse failures.
14 of 20 had typed business/organization declarations.
Direct business-node properties: phone 9, hours 9, areaServed 9, streetAddress 7.
Own-business aggregateRating appeared on 2 homepages.
Microdata indicators appeared on 1 homepage; Schema.org RDFa on 0.
These are narrow descriptive counts, not SEO success or failure measures.

REVIEW AND PRIVACY
Collection and selected source-text comparisons were AI-assisted. Aggregation
and all 35 source hashes were checked programmatically. No business-owner
verification or human expert review is claimed.
Published provider data omits exact street addresses, telephone numbers, emails, raw HTML,
full provider graphs, and verbatim evidence passages. The public domain/URL list
allows readers to identify the source scope without a league table of bad sites.

MOOSERANK IMPLEMENTATION RECORD
MooseRank was reviewed separately and is not one of the 20 sampled providers.
Public homepage/contact snapshots preceded the change. The two availability
assertions removed from the implementation are documented in downloadable before
and after records. No ranking lift, Google approval, GBP edit, or rich-result
appearance was measured or claimed.