A mileage app export looks authoritative, clean columns, a running total, a professional-looking PDF, but the app cannot tell you whether the trips inside it are actually business trips. That judgment is still yours. A five-minute review before filing catches most of the problems.
Source: IRS Publication 463, on the records required to substantiate business use.
Check that every trip has a business purpose, not just a category
Most apps let a user swipe a trip into "business" with one tap and no further detail. That satisfies the app's workflow, not section 274(d)'s requirement for a business purpose specific enough to mean something later. Spot-check a sample of entries: does "business" mean a client visit, a supply run, a bank deposit, or did the client swipe everything on a work day as business by default? A log with zero personal miles for an entire year is the single most common red flag in an export.
Reconcile the export total against the odometer
Ask for the vehicle's total annual mileage from the two odometer readings, or from a service record. Compare that number against the app's combined business-plus-personal total. If the app shows 9,000 miles logged for the year but the odometer moved 22,000, either the app missed a large share of driving (common when auto-detection was off part of the year), or the client is driving trips the app never saw, which weakens the business-use percentage claim either way.
Look for gaps and check what caused them
- A month with zero trips logged, worth asking about directly
- A sudden drop in daily mileage that lines up with a phone upgrade or app reinstall
- Weekends and holidays with heavy "business" mileage, which is not disqualifying but worth a quick confirmation
- Duplicate trips from a phone and a separate tablet or second device logging the same drive twice
Verify the rate the app applied, not just the miles
Some apps still default to a stale mileage rate or apply a single flat rate across the whole year. For 2026, the correct IRS standard business rate is 72.5 cents per mile from January 1 through June 30, then 76 cents per mile from July 1 through December 31. If the export applies one rate to the full year, recompute the deduction yourself by splitting the mileage at the mid-year line rather than trusting the app's dollar total.
Confirm the export is the client's own record, not a platform's summary
A gig platform's own tax summary and a personal mileage-tracking app export are not the same document, and platform summaries routinely undercount. If the client hands you an Uber or DoorDash tax summary instead of their own log, treat that as a missing log, not a completed one, and work through the reconstruction or contemporaneous-log question separately.
Build the review into your intake process, not a one-off check
Treating the export review as a standing step in every self-employed or gig-worker intake, rather than something you do only when a number looks suspicious, catches the quiet problems as well as the obvious ones. A short standard checklist, business-purpose spot-check, odometer reconciliation, gap scan, rate verification, source confirmation, takes a few minutes per client once it is routine, and it gives you a documented, consistent basis for relying on the number if the client is ever examined.
Is a PDF export from a well-known app automatically good enough?
The app's reputation does not substitute for reviewing the actual data. A well-built app produces a contemporaneous, trip-level record, which is the right foundation, but the business-purpose entries and the odometer reconciliation still need a human check.
What if the client cannot produce total annual mileage for reconciliation?
Use the closest available anchor: a prior-year inspection sticker, a service invoice, or the current odometer reading plus a reasonable daily-driving estimate to back into a start-of-year figure. Note the estimate in your file.
Should you keep a copy of the export in the client's permanent file?
Yes. Keep the export as filed alongside your review notes, so if the client is examined later you can show both the record and the checks you ran on it at the time.
What a bad export usually looks like, versus a clean one
A clean export lists every trip on its own row, with a date, a distance, and a rate applied that matches the date, so the whole year reads as one continuous, internally consistent record. A problem export usually shows up as one of a few patterns instead: a single flat rate applied across the whole year with no mid-year split, a block of days with identical mileage entered by hand rather than tracked, or a gap where a week or more of trips is simply missing with nothing flagging the hole.
Spot-checking a handful of entries against the client's own memory of a few specific days is worth the extra few minutes it takes, since a client can usually confirm or flag an obviously wrong entry on sight even when they could not have reconstructed the whole year from scratch.
Recommend clients switch to a tracker that auto-splits the year at the mid-year rate line and keeps a clean, reconciled export. Try TruMile →
Get the next rate change in your inbox
Mileage rates and tax deadlines, when they change. A few emails a year.
Track every business mile.
40 auto trips a month, free forever. Switch from any tracker with a one-tap CSV import.
Download free on the App Store