How Much Do Uber & DoorDash Drivers Really Make After Expenses?
Direct answer: gig drivers keep roughly half of what the app screen shows. A representative full-time week of $1,200 gross becomes about $400 of operating costs and a $200 tax set-aside, leaving $600 take-home — $15.00 an hour across the 40 hours you actually worked, against a screen rate of $34.29 on the road. Operating costs typically land somewhere near 30–45% of gross. Below is the full net-versus-gross breakdown, the ranges worth planning around, and the four numbers to track so you are trusting your own logs instead of a forum.
Net versus gross: a full week, line by line
Formula: net hourly = (gross earnings − operating costs − tax set-aside) ÷ total hours worked
| Line | Amount | Note |
|---|---|---|
| Gross earnings (app screen) | +$1,200 | The number the app calls "earnings" |
| Platform add-ons, tolls, parking | −$40 | Charged back to you on some orders and markets |
| Fuel | −$150 | The most visible cost, and not the biggest one |
| Insurance share (delivery/rideshare) | −$45 | Endorsement plus the personal premium allocated to work use |
| Maintenance and tires | −$55 | Lumpy — budget it monthly or it ambushes you quarterly |
| Depreciation and wear | −$90 | The car quietly losing value while it works |
| Phone, data, supplies, cleaning | −$20 | Small enough to forget, large enough to matter |
| Operating costs | −$400 | 33% of gross — inside the usual 30–45% range |
| Net profit before tax | $800 | What the tax system actually sees |
| Tax set-aside at 25% | −$200 | Self-employment tax plus income tax on net profit |
| Take-home for the week | $600 | ÷ 40 hours worked = $15.00/hour |
Three different hourly numbers describe the same week: $34.29 on the app screen ($1,200 over 35 on-road hours), $30.00 if you count every hour you were available or dealing with the app, and $15.00 after costs and taxes. Only the last one tells you whether this beats your alternatives — which is exactly what our gig driver earnings calculator computes from your own figures.
Cost ranges worth planning around
Every market and vehicle differs, so treat these as rough weekly planning figures for a driver working 30–40 hours, not quotes: fuel $100–$220, insurance allocation $30–$70, maintenance and tires $30–$80, depreciation and wear $50–$160, and tolls, parking, phone and supplies $20–$70. Stack them and total operating cost usually lands between 30% and 45% of gross. A paid-off, economical car sits near the bottom of that band; a financed, thirsty SUV sits above it. If your own logs come in consistently above 45%, the vehicle — not the platform — is the problem you need to solve.
The benchmark data
Gridwise Analytics' 2026 driver data puts Uber Eats at roughly $24.68/hour gross and DoorDash at $18.93/hour, with DoorDash (about 67% of US delivery share per the same dataset) winning on daily volume: ~$63.66/day vs ~$52.94/day. Other earnings trackers put DoorDash net earnings — after gas, mileage, and depreciation — at $13–20/hour, with peak-window specialists reporting $22–28/hour.
The consistent finding across every study: take-home runs 30–50% below the app-screen number once vehicle costs are counted. Drivers who ignore this are quietly subsidizing their own cars.
The costs the app never shows
- Mileage: the IRS standard mileage rate for 2026 is 76 cents/mile — a bundled estimate covering fuel, maintenance, tires, insurance, and depreciation. A driver doing 600 miles/week is consuming ~$460/week of vehicle value at that rate.
- Unpaid time: waiting between rides is working time that earns nothing. Studies of driver logs show dead time can be 30–40% of hours on the road.
- Self-employment tax: gig income carries the full 15.3% SE tax plus income tax (see our tax set-aside guide).
Worked example
Driving 35 hours/week for $900 gross, with $90 fuel, $25 maintenance, 600 miles at $0.25/mile depreciation, $20 phone/insurance, and a 20% combined tax rate:
- Vehicle + other costs: $90 + $25 + $150 + $20 = $285/week
- Net before tax: $900 − $285 = $615
- After 20% tax: $492/week → $14.06/hour
The app said $25.71/hour. The truth is $14.06. Both numbers matter — but only one tells you whether the job is worth it versus your alternatives. Run your own figures in our gig driver earnings calculator.
Practitioner note: The mistake I see most often is benchmarking against the wrong column: drivers compare their $25.71 screen rate to a friend's $22 and feel ahead, when both net near $14 after $285 of weekly costs and 20% tax. Compare net-to-net or not at all — and average over four weeks, since any single week can swing 30% on weather, holidays, or one repair.
Why the weekly number swings
A week is a noisy unit. Weather, a holiday weekend, a slow restaurant quarter, a tire that finally gave out — any single week can land 30% above or below your average, and plenty of drivers quit over one bad month. Average over four weeks instead, separate one-off costs (tires, brakes, a repair) from recurring ones (fuel, insurance, phone), and judge the trend rather than the week. The quarter is the honest unit: it absorbs the lumpy repairs, the dead weeks, and the bonus-heavy ones in the same breath.
Track four numbers, not one
You cannot manage an hourly figure you never calculate. Keep a weekly log with four columns and everything else follows from them:
- On-road hours and available hours: the gap between them is dead time, and the first thing to shrink. If dead time sits above 30% of your hours, the fix is when and where you drive, not how hard you drive.
- Gross earnings per app: recorded before any deduction, so it stays comparable with the benchmarks above and with your own history.
- Every work cost: fuel, insurance, repairs, tires, phone, tolls, cleaning. A receipts folder plus one monthly total beats a spreadsheet you abandon in a fortnight.
- Tax set-aside: move a fixed percentage into a separate account the day you are paid — the IRS estimated-tax rules explain who has to pay quarterly and when. Money you never see is money you never spend.
Four weeks of that log tells you more about your real rate than an afternoon of forum reading, and it feeds directly into the net-per-hour method in our side-hustle economics guide.
How better drivers close the gap
The highest earners, per both Gridwise and driver-tracker data: multi-app drivers who decline orders below a personal value threshold, work only peak windows (lunch, dinner, Fri/Sat nights, weather events), and log every mile and receipt for tax time. Strategy moves the number more than the platform does — on a $15/hour baseline, refusing the worst quarter of offers, concentrating on two peak windows a day, and running both apps typically recovers a few dollars an hour. That is why the drivers who look lucky are mostly just selective.
Deciding with the number instead of the vibe
Once your four-week net hourly exists, the decision is a comparison rather than a feeling. If your net hourly beats what the same hours would earn in a regular job once you price the flexibility you actually value, keep driving and protect the profitable hours. If it sits below your alternative, you have three levers in order of effectiveness: raise your effective rate (selectivity, peak windows, both apps), cut vehicle cost (the vehicle you choose matters more than the one you drive today), and cut dead time. Only when all three have been worked does the honest answer become a different job — and a four-week log makes that conclusion safe rather than emotional.
Where the $15-an-hour math goes wrong:
- You count only on-road hours. Dead time of 30–40% is real work that earns nothing — the $34.29 screen rate becomes $30.00 the moment every available hour counts.
- You annualize a bonus week. Streaks and surges are lumpy by design — plan on steady-state net and treat bonuses as upside.
- You ignore depreciation on a financed car. The $90-a-week wear line is the quietest cost and the one that decides whether the car outlasts the loan.
- You compare gross across markets. A dense urban $24.68 and an exurban $18.93 buy different hours, different miles, and different dead time.
Online hours, active hours, and net hours: three rates for one week
The same $1,200 week reads three ways: $34.29 per on-road hour across 35 hours, $30.00 once every available hour counts, and $15.00 net across 40 hours worked after $400 of operating costs and a $200 tax set-aside. The smaller worked example tells the same story: a $25.71 screen rate becomes $14.06 after $285 of weekly costs and 20% tax. Quote the screen rate to compare apps; use the net rate to decide whether to keep driving.
Per-platform pace table: what the benchmarks net out to
The same costing logic behind the worked weeks above, applied to the benchmark gross rates on this page. Take-home runs 30–50% below the app-screen number once vehicle costs are counted:
| Pace check | Gross basis (this page) | After-expenses pace | Source week |
|---|---|---|---|
| Main-week pace | $1,200 gross | $15.00/hr net ($600 take-home ÷ 40 hours after $400 costs and $200 tax) | Net-versus-gross table above |
| Small-week pace | $900 gross over 35 hours ($25.71/hr screen) | $14.06/hr net ($492 ÷ 35 after $285 costs and 20% tax) | Worked example on this page |
| Tool-week pace | $1,050 gross over 38 hours ($27.63/hr screen) | $15.60/hr net ($592.80 ÷ 38 after $290 costs and 22% tax) | Gig calculator worked example |
| Uber Eats benchmark pace | $24.68/hr gross | Inside the $13–20/hr net band once 30–45% operating costs and tax apply | Benchmark data above |
| DoorDash benchmark pace | $18.93/hr gross | Inside the $13–20/hr net band; peak-window specialists reach $22–28/hr | Benchmark data above |
Strategy moves the number more than the platform does: on a $15-an-hour baseline, refusing the worst quarter of offers, concentrating on peak windows, and running both apps typically recovers a few dollars an hour.
One honest caveat
Benchmarks describe markets, not you. Dense urban markets pay more than exurban ones; a 2015 Civic costs half as much per mile as a 2024 pickup. Your logs beat every study — which is exactly why we built the calculator.