TrackMyGeo

Algeria American Samoa Angola Argentina Australia Bangladesh Barbados Benin Bolivia Botswana Brazil British Indian Ocean Territory Burkina Faso Burundi Cameroon Canada Cape Verde Central African Republic Chad Chile Colombia Comoros Cook Islands Cuba Democratic Republic of the Congo Djibouti Dominica Dominican Republic Ecuador Egypt Equatorial Guinea Eritrea Eswatini Ethiopia Falkland Islands Fiji French Guiana French Polynesia Gabon Gambia Ghana Grenada Guam Guinea Guinea-Bissau Guyana Haiti Ireland Ivory Coast Jamaica Kenya Kiribati Lesotho Liberia Libya Lithuania Madagascar Malawi Mali Marshall Islands Mauritania Mauritius Mayotte Mexico Micronesia Morocco Mozambique Namibia Nauru New Caledonia New Zealand Niger Nigeria Niue Northern Mariana Islands Palau Papua New Guinea Paraguay Peru Republic of the Congo Réunion Rwanda Saint Helena, Ascension and Tristan da Cunha Saint Lucia Saint Vincent and the Grenadines Samoa Sao Tome and Principe Senegal Seychelles Sierra Leone Solomon Islands Somalia South Africa South Sudan Sudan Suriname Tanzania Togo Tokelau Tonga Trinidad and Tobago Tunisia Tuvalu Uganda United Kingdom United States Uruguay Vanuatu Venezuela Wallis and Futuna Western Sahara Zambia Zimbabwe

My excuse was, I needed to wait a month to see if the subscription renewed correctly. Then I needed to wait another month to see if it correctly expired, stopped billing and paused the service correctly. Thats my excuse, Im sticking to it. It worked.

Architecture The app is built as a Next.js web platform with a Node.js runtime and a custom WebSocket server for realtime updates. It uses PostgreSQL and TimescaleDB for relational and time-series data, Azure IoT Hub and Azure Functions for telemetry ingestion, NextAuth with Microsoft Entra External ID / Azure AD B2C for authentication, and Stripe for subscription billing. Tech stack Web app: Next.js, React, TypeScript, Tailwind CSS Server/runtime: Node.js, custom WebSocket server Authentication: NextAuth, Microsoft Entra External ID / Azure AD B2C Data: PostgreSQL, TimescaleDB Device / telemetry: Azure IoT Hub, MQTT, Azure Functions Maps / geospatial: MapLibre GL, TileServer GL Payments: Stripe Infra / cloud: Azure App Service, Azure Bicep State / data fetching: TanStack Query Testing: Vitest, React Testing Library Observability / communications: Azure Application Insights, Log Analytics, Azure Communication Services

Switch logged in user, is a pain in the auth. Just log auth.

Full stop, end of story.

Geo zone speed events being delivered hours after the event is like being punched in the face, in hospital.

6. No Single Pane of Glass Fragmented views: multiple accounts, multiple maps, no unified fleet picture. 7. Labor Gap & Update Nightmare Fleet-wide policy isn’t a spreadsheet row — it’s physical labor: firmware, APN, or box settings on dozens of hardwired units means paying someone to crawl under every dash, repeatedly. Same class of pain as only being able to change fleet-wide behavior after a yard day. 8. Calibration Drift & Cheap Hardware Truth Budget boxes: heading / motion sensors drift after months of vibration; you think you have truth, you have slowly wrong truth. Phones aren’t perfect either, but the failure mode here is cheap silicon with no field recalibration without a visit. 9. Cloud vs Landfill Legacy fleets rarely get seamless fleet-wide OTA for brain-in-the-box; units age into unsupported SKUs while phones get app updates from the store and policy updates from the cloud. Their pattern is often capex + crawl + landfill; phone-based stacks are enroll + update + redeploy without a bay lift per VIN. 10. Battery Drain — Whole Fleet “If I put tracking on every phone, do I get fifty ‘my battery died’ tickets?” — ops, HR, and IT worry about support load at scale, not one driver complaining. In practice data packets are usually small and battery impact is often modest (device, settings, and coverage vary). 11. Vehicles Still Need to Be Turned or Moved Regularly Hardwired or poorly designed trackers can flatten batteries if vehicles sit; policy becomes “start every vehicle on a schedule” — another hidden ops tax. Parked fleet = flat lines on the map; tracking doesn’t replace dispatch, maintenance, or rolling stock — you’re not promising magic on cars that never move.