A single top-to-bottom queue for a fleet that was outrunning a spreadsheet

The problem

The fleet ran on a spreadsheet, and the spreadsheet was wrong. Stale sum ranges were under-reporting spend across a dozen tabs by more than $30,000 combined, and nobody had a reason to suspect it until someone went looking. Registration stickers, insurance cards and emissions certificates lapsed on whatever cadence someone remembered to check. Engine fault codes from the telematics platform the company was already paying for went unread because nothing surfaced them, and oil changes ran on a rough sense of when a truck was probably due rather than its real odometer. With three dozen active vehicles and trailers and hundreds of historical service records behind them, mileage errors compounded: a vehicle's mileage feeds its retirement date and the capital plan, so a reading that was half of true produced confidently wrong answers everywhere downstream.

What we built

We built a single triage queue: one list, worked top to bottom, and when it's empty the fleet work is done for the day. It pulls together expired and soon-to-expire registration, open engine fault codes, oil changes due against the real odometer, driver-reported issues, and missing paperwork, with the tab's badge count reading the exact same logic as the page so the two can never disagree. It syncs read-only from the client's existing ELD/telematics platform (Motive), matching vehicles by VIN first and unit number second, and never writes back into a system that also holds federally regulated hours-of-service records. Around the queue sits a full vehicle registry with service history and document storage, manual service logging, an insights view with a spend trend and a mileage-driven retire/replace watchlist, a capital plan by year, and a seven-step retirement checklist that archives a vehicle to permanent history rather than deleting it. Every mutation is written against the identity the company's own Microsoft sign-on verified, including who viewed a document, and the browser never holds a database credential: it sends intent, a server resolves roles and does the write, so a compromised browser session can't post an arbitrary number.

The outcome

The spreadsheet's quiet shortfall came to light and got fixed at the source rather than patched over. The mileage-conflict check alone caught a vehicle reading roughly half its true mileage in one system versus its own service history, the kind of gap that would have skewed its replacement date and every mileage-driven decision behind it. Compliance stopped depending on someone remembering: the queue surfaces what's due before it lapses, not after. And because the underlying logic, the queue, the odometer-conflict detection, the mileage-first watchlist, the provenance rules, is decoupled from the specific telematics platform and identity provider it launched with, it fits any business with a similar shape: a light-duty mixed fleet of roughly 20 to 250 owned vehicles, already on telematics for compliance but getting no maintenance value out of the odometer feed it already pays for, with nobody who owns fleet as a full-time job. That covers most field-service trades (pest control, lawn care and landscaping, snow removal, HVAC, plumbing, electrical), environmental and vegetation contractors, small construction and site-services firms, delivery and last-mile courier fleets, security patrol, multi-location franchise operators, non-profit and small public-sector fleets, and non-emergency medical transport.