PWA + offline Service Worker
All calculation logic in SW cache; field engineers use it offline.
Engineering Utility / SaaS · Case Study
A fast, accurate, schema-rich PWA for engineering unit conversions. Edge cache + Web Worker computation + Google Discover-ready structured data.

Results
The Brief
Engineers won't spend more than 5-10 seconds on a converter mid-workflow. Existing online converters are ad-laden, slow to load, and precision-broken.
The Approach
I applied the Technical pillar (PWA, edge cache, INP) and AEO pillar (H1 + direct answer for engineer queries like "X kPa to MPa"). Entity signals via Schema.org Conversion + UnitDefinition (linked to Wikidata Q-IDs).
See the full process → 47-Point AI-Ready SEO Audit methodology.
Tech Stack
The Build
PWA + offline-first: the converter runs without internet. Service Worker caches all calculation logic — critical for engineer scenarios (field work, in-flight).
Web Worker computation: 12-decimal precision math doesn't block main thread. First render < 800ms p75.
Every unit linked to a Wikidata Q-ID (e.g., kilopascal → Q44395). UnitDefinition schema ties entity signals into Knowledge Graph — we surface in the definition card on "kPa" queries.
Steps Applied
All calculation logic in SW cache; field engineers use it offline.
12-decimal math in worker; UI keeps < 800ms p75 first paint.
Every unit linked to a Wikidata Q-ID; Knowledge Graph entity signals.
Honest Reflection
Version one had a separate page for each of 24 unit families — 180+ programmatic pages total. But UX fragmented: an engineer searching "kPa to MPa" had to click again to switch from kPa to MPa. Version two ships a single SPA + separate SEO pages (programmatic stays). On the next utility, I'll question the "SEO page count = UX page count" assumption.
Screenshots


Same 47-point methodology, $499 fixed price, 5-7 business days. Book the audit.