Soniteq — The Reference Implementation
We built this on our own site first. Soniteq is the founder's other company, and it's where EntityMesh's structured-data templates were proven before being sold to anyone.
Disclosure. Soniteq is the founder’s own sister company and EntityMesh’s reference implementation — not a client engagement. Every figure on this page describes soniteq.co, not blueninja.systems’ own footprint.
Measured 13 August 2026. All schema figures below come from a single scan on that date. The method, the full 28-type inventory, and the raw results are published in the archived scan report so anyone can re-run it.
Context
What This Case Study Is
Soniteq.co is the team’s own reference implementation of EntityMesh — built by the same people behind Blue Ninja Systems. This case study documents how that team applied its own philosophy to their own website, creating a reference-level implementation that demonstrates what AI-era readiness looks like in practice.
This distinction matters: EntityMesh delivers the Support Hub, Answer Hub, schema templates, and monitoring playbook. The site design, product pages, technical infrastructure, and AI crawler configuration were built separately by the site owner. The results below reflect the combination of both — the product doing its job, and the owner applying the underlying philosophy across the full site.
This is the standard we hold ourselves to, and the standard we help our customers reach.
Product Scope
What EntityMesh Delivered
Delivered by EntityMesh
- Support Hub category architecture and navigation model
- Answer Hub content map, page-type standards, and question-intent coverage
- Schema templates and implementation guidance by page pattern
- Monitoring playbook for narrative tracking and iteration planning
Built by the Site Owner
- Full site design, visual identity, and front-end engineering
- Product pages, pricing, and marketing copy
- Technical infrastructure: SSR, pre-rendering, and deployment pipeline
- llms.txt, llms-full.txt, humans.txt, and AI crawler configuration
Note on schema: EntityMesh provides schema templates and implementation guidance as part of every engagement. Customers implement schema on their own pages; those on a Managed Retainer have implementation handled for them. The schema coverage results below reflect the site owner applying those templates across the full site.
Implementation
Modules Applied
Diagnostic
Baseline crawl and schema diagnostics identified structural gaps and prioritized fix order across the Answer Hub.
Auth Graph
The entity, proof, comparison, source, and next-action map shaped the publishable architecture.
EchoScan
Controlled prompt-set monitoring posture was prepared to track SOMV, citation presence, and definition drift over time.
Positioning
AI-Era Readiness
2,987
Schema instances
across 181 pages · scanned 13 August 2026
181
Pages scanned
every URL in sitemap.xml · 0 failed to fetch
0
Parse errors
across 1,023 JSON-LD blocks
On every page — 100% coverage, 13 August 2026
Eight schema types appear on all 181 pages. Not most pages — every one:
Across the whole site that is 2,987 schema instances spanning 28 distinct @type values — a count that includes nested types such as ListItem inside BreadcrumbList, and Question and Answer inside FAQPage. They are counted because they are present, not presented as 28 separate implementations.
For outside context: the HTTP Archive Web Almanac 2024 found JSON-LD structured data on 41% of the pages it measured, and BreadcrumbList on 5.66%. Every page of soniteq.co carries both.
That is context, not a ranking. The Almanac samples roughly one or two pages per site — largely home pages — while the figures above come from a full-site crawl of a content-heavy site, where markup like Article and FAQPage naturally concentrates. The two populations are not like-for-like, so no multiple, percentile, or position against other sites is implied or should be inferred.
Technical Foundation
The Critical Role of Pre-Rendering
One of the most consequential technical decisions for AI-era discoverability is whether a site serves pre-rendered HTML or depends on client-side JavaScript to assemble its content.
Most AI crawlers — including GPTBot, PerplexityBot, and Google-Extended — do not execute JavaScript. Content that only exists after hydration is content those crawlers never see, however good the markup underneath it is. soniteq.co is served pre-rendered, which is why a fetch-and-parse scan like the one above can read every page without running a browser.
EntityMesh flags rendering dependency as a critical finding in every Diagnostic, and the schema templates are designed to work within a pre-rendered architecture. We have not published a crawlability percentage for soniteq.co here, because the 13 August 2026 scan measured schema coverage, not rendering — and an unmeasured number does not belong on this page.
Measured 13 August 2026
Schema Coverage
Beyond the eight types on every page, these carry partial coverage across the 181 pages scanned.
How these are counted. Coverage = distinct pages containing a type ÷ pages scanned. Instances and pages-with-type are different quantities and both are reported below: a single page can carry several instances of a type, and coverage still counts that page once. Because the denominator is pages scanned, no figure here can exceed 100%.
Schema Type
Instances / Pages
Coverage
118 / 118
65.2%
116 / 116
64.1%
272 / 116
64.1%
272 / 116
64.1%
70 / 70
38.7%
31 / 20
11%
16 / 16
8.8%
14 / 14
7.7%
12 / 12
6.6%
On Article and FAQPage. Both sit around 65% rather than covering the whole site. The instance counts have barely moved while the site has grown — coverage is lower because pages were added faster than markup, not because markup was removed. That gap is the next build cycle’s work.
Source: schema.org coverage scan of soniteq.co, 2026-08-13 — 181 pages scanned from sitemap.xml, 1,023 JSON-LD blocks parsed with zero errors, 2,987 total schema instances across 28 distinct types. Coverage = distinct pages containing a type ÷ pages scanned. Independently spot-verified in Google Rich Results Test (2 pages, 0 errors). Full results archived at evidence/SONITEQ_SCHEMA_SCAN_2026-08-13.md. Schema implemented by site owner using EntityMesh templates.
Verification
Checked Against Google
A scan is only as good as its method, so two pages were spot-checked in Google’s Rich Results Test on the same day. Both returned zero errors, and Google detected the same types this scan recorded.
soniteq.co/answers/what-is-soniteq
4 valid items detected
Articles · Breadcrumbs · Organization · Paywalled Content
Zero errors. Some items carry non-critical issues — these are optional recommended properties, not errors.
soniteq.co/kora
3 valid items detected
Breadcrumbs · Organization · Software Apps
Zero errors. Some items carry non-critical issues — these are optional recommended properties, not errors.
Screenshots of both results are archived alongside the scan report.
What this does not claim
Schema coverage measures how well a site is structured to be understood by machines. It is not a ranking, not a prediction of AI citations, and not a guarantee of traffic. Whether any engine cites this site is decided by that engine.
Outcomes
Ongoing Monitoring
With the foundational infrastructure in place, the project has now entered the monitoring phase. Using EchoScan, we are actively tracking crawlability, schema coverage, AI citation presence, SOMV, and definition drift using controlled prompt sets. As a new site, Google Search Console data is still accumulating — this is itself a live demonstration of the monitoring phase in action.
Measurement
Metrics Framework
EntityMesh success is measured through a combination of quantitative performance indicators and qualitative narrative analysis. Here’s a guide to the metrics we track.
Quantitative Metrics (The "What")
Hard numbers that track performance over time.
- Crawlability Score (from Diagnostic)
- Schema Coverage (% of pages with valid schema)
- MeshScore readiness baseline
- Organic traffic to Answer Hub pages
- Keyword rankings for question-intent queries
- Number of featured snippets & AI Overviews won
Qualitative Metrics (The "Why")
Narrative-focused indicators of brand strength in AI models.
- Narrative presence (How accurately do AI models describe you?)
- SOMV (How strongly do models recommend you compared with competitors?)
- Definition Drift (Is the AI's definition of you consistent?)
- Citation quality in AI-generated answers
Build your own reference implementation
EntityMesh gives you the system, the templates, and the monitoring playbook. You bring the vision.