Stand in any grocery store parking lot at 6 p.m. and you can watch the modern gig economy in motion: the same drivers, often running two or three apps at once, trying to answer one question the apps never answer honestly. What does this actually pay per hour once the car gets its cut? Every delivery platform advertises its best-case number. None of them advertise the gas, the brake pads, the tires, the unpaid drive back from a far-flung drop-off, or the 45 minutes you sat in a parking lot waiting for an order that never came. This guide does the math the apps skip, app by app, so you can decide whether delivery driving deserves your evenings and weekends in 2026.
One thing up front: delivery can absolutely be worth it. For people who need money this week, it remains one of the fastest legitimate ways to turn free time into cash, usually with a first payout inside seven days. But the difference between a driver clearing $19 an hour after expenses and a driver unknowingly working for $8 an hour is not luck. It is app choice, market, schedule, and a few habits we will walk through below.
Every comparison in this article runs on three numbers, and only the third one matters to your bank account:
A quick example of why this matters. A driver grosses $22 in an hour but drives 25 miles to do it. At a realistic 30 cents per mile, the car quietly consumed $7.50 of that hour. Real wage before taxes: $14.50. That is the honest arithmetic behind every row in the comparison table below.
DoorDash is the largest restaurant delivery platform in the United States, which cuts both ways. There are orders almost everywhere, including suburbs and smaller towns the other apps barely serve. There are also more drivers competing for the good ones.
Pay is a base amount per offer, typically a few dollars, plus customer tips and occasional promotions. Tips carry the economics, which means your earnings depend heavily on neighborhood and time of day. Experienced Dashers in decent markets commonly gross $15 to $25 per active hour during lunch and dinner rushes, and meaningfully less outside them. DoorDash also offers an earn-by-time mode in many markets that pays a fixed hourly rate for active delivery time plus tips. It smooths out the bad offers but usually caps your great nights, so many drivers use it only when the market is slow.
The skill that separates profitable Dashers from frustrated ones is offer selection. A $4 offer that takes 25 minutes and 9 miles is a pay cut wearing a costume. A common driver rule of thumb is to want roughly a dollar per mile or better on the round trip, and to decline the rest. Your acceptance rate can affect access to certain priority programs, but chasing a high acceptance rate by taking money-losing orders is backwards.
Uber Eats works a lot like DoorDash: small base pay per trip plus tips, with upfront information about the offer before you accept. It tends to shine in dense urban cores and college towns, where short trips stack up quickly, and to lag in spread-out suburbs where the miles eat you alive. Gross earnings in active hours typically land in the same $15 to $25 range as DoorDash in comparable markets.
Two structural advantages are worth knowing. First, Uber Eats shares a driver pool and app infrastructure with Uber rides, so drivers who also do rideshare can flip between food and passengers as demand shifts through the night. Second, because most multi-apping drivers run DoorDash and Uber Eats together, it is the natural second app: you keep both online and take whichever offer clears your per-mile bar first.
Grubhub's national footprint is smaller, but in some metros, particularly in the Northeast and Chicago, it remains a serious source of orders. It offers scheduled blocks that give priority access to deliveries, plus open dashing when there is room. Pay per offer is often a touch higher than competitors in its strong markets and noticeably worse in its weak ones, so this app is entirely a question of geography. Sign up, run it alongside the big two for a week, and let your own numbers decide whether it keeps its slot on your phone.
Instacart is a different job wearing the same gig clothing. You are not just driving; you are shopping a grocery list, communicating about substitutions, hauling cases of water up apartment stairs, and then delivering. Batches pay a base amount plus tips that you can see, mostly, before accepting. Good full-service batches gross $20 to $35, and a strong hour can beat anything the restaurant apps offer.
The catches are real, though. Batches can be sparse outside busy windows, so your gross per hour of waiting-plus-working often lands closer to $15 to $22. The work is physical. And Instacart has a known problem the community calls tip baiting, where a customer attaches a large tip to attract a fast shopper and reduces it after delivery. It is not the norm, but it happens often enough that veteran shoppers mentally discount oversized tips on small orders.
Instacart fits people who would rather be moving through a store than sitting in a restaurant lobby, and it pairs poorly with other apps because a 90-minute batch locks you in.
Amazon Flex flips the model. Instead of cherry-picking offers, you reserve a delivery block in advance, commonly two to four hours, with the pay shown upfront. Advertised rates generally run about $18 to $25 per hour, and unlike the restaurant apps, that number is mostly real because it is contractual for the block. Packages for standard Amazon.com blocks do not earn tips; Whole Foods and grocery blocks can.
The honest downsides: blocks in good time slots get claimed fast, sometimes within seconds, so getting consistent work takes persistence or very flexible availability. Routes can be dense apartment complexes or 70-mile rural loops, and the pay is the same either way, so your per-mile cost is the wild card. A $72 three-hour block that involves 90 miles of driving nets very differently than the same block covering 30 miles. Flex suits people who want a predictable, schedulable shift rather than a game of offer roulette.
Spark handles Walmart's curbside and delivery orders, plus shopping orders similar to Instacart. Offers arrive round-robin style, and gross earnings typically run $12 to $20 per hour depending on market saturation and how long the store's curbside pickup keeps you waiting, which is the platform's chronic complaint. Its advantage is footprint: in suburban and rural areas where restaurant delivery is thin, the local Supercenter often has steady order flow all day, not just at mealtimes. In a town with one Walmart and four DoorDash drivers, Spark can quietly out-earn everything else on this list.
Here is where most delivery income quietly leaks away. Restaurant delivery typically involves 15 to 30 miles of driving per active hour. Multiply honest per-mile cost by honest miles and the picture changes fast:
That is how you get to 25 to 35 cents per mile of real marginal cost for a paid-off economy car, and meaningfully more for a truck, an SUV, or anything financed and newer. Drivers in older fuel-efficient cars are structurally more profitable than drivers in new crossovers doing the exact same deliveries. The chart below shows where a typical $100 of restaurant-delivery gross actually ends up for a driver covering about 110 miles to earn it, with a tax reserve included.
Two practical rules fall out of this math. First, judge every offer in dollars per mile, not dollars per order; a dollar per mile is a workable floor in most markets, and the best drivers hold out for more. Second, track every mile from the moment you go online, because business mileage is deductible and, at the IRS rate, a 12,000-mile delivery year generates a deduction worth several thousand dollars in reduced taxable income. A mileage log, even a simple app-based one, is the highest-paid paperwork in gig work.
Almost no experienced driver runs one app. The waiting time between offers is the silent killer of hourly earnings, and the fix is running two or three platforms simultaneously and letting them compete for your next 30 minutes. Here is the system in practice:
A few etiquette-and-survival notes. Never accept overlapping orders you cannot deliver promptly; late food earns bad ratings and deactivation risk on every platform involved. Pause the other apps the moment you accept an offer. And keep food delivery and grocery batches separate in time, because a melted ice cream complaint is not worth an extra $6.
Scheduling matters as much as stacking. Lunch (about 11 a.m. to 1:30 p.m.) and dinner (about 5 p.m. to 9 p.m.) carry restaurant apps; weekend mornings carry grocery; bad weather is quietly the most profitable shift in delivery, because orders surge exactly when drivers stay home. Friday and Saturday dinner in the rain is the closest thing this industry has to a guaranteed good night.
Every app on this list pays you as an independent contractor, which changes three things immediately.
You owe self-employment tax. On top of regular income tax, net self-employment profit is subject to a 15.3 percent self-employment tax covering Social Security and Medicare. On $10,000 of net delivery profit, that is roughly $1,413 after the standard adjustment, before any income tax. This is the line item that shocks first-year drivers every April.
The form does not define the obligation. Platforms issue Form 1099-K or 1099-NEC once you cross federal reporting thresholds, which Congress has changed more than once in recent years. Do not let the presence or absence of a form decide anything: profit is taxable from the first dollar, reported on Schedule C, whether or not paperwork arrives.
Deductions are your raise. The mileage deduction usually dwarfs everything else, but hot bags, phone mounts, a portion of your phone plan, and tolls count too. Most drivers with meaningful earnings should also make quarterly estimated tax payments to avoid an underpayment penalty; the IRS estimated taxes page explains the safe-harbor rules. A simple habit that prevents nearly all gig tax pain: move 20 to 25 percent of every payout into a separate savings account the day it lands, and pay quarterlies from there.
Delivery income has a documented tendency to evaporate, partly because it arrives in small, frequent, spendable amounts. Drivers who treat it as a targeted tool, this money exists to kill this credit card or fund this emergency fund, consistently report better outcomes than drivers who let it blend into the checking account. Park payouts somewhere slightly out of reach, like a high-yield savings account, and watch how quickly part-time miles become a real number:
Expectations calibrate everything, so here is an honest first week. Signup and background checks usually clear in one to four days per app, faster on DoorDash and Uber Eats, sometimes slower on Flex and Spark. Your first few shifts will be clumsy: you will misjudge restaurant wait times, get lost in an apartment complex, and accept at least one order you immediately regret. That is normal, and it is also why first-week earnings usually land 20 to 30 percent below what the same hours pay a month later.
Plan your first week around three two-to-three hour shifts during dinner peaks rather than one long Saturday grind. Short peak shifts teach you the profitable patterns, which restaurants are fast, which neighborhoods tip, which zones strand you, without burning you out on the slow hours in between. Keep a simple note after each shift: hours online, miles driven, gross earned. By day ten you will have something almost no new driver has, an evidence-based answer to whether this is worth your time, and you will have earned a few hundred dollars collecting it.
One more first-week tip that veterans swear by: drive your zone once with no orders, at the time you plan to work. Learn where the restaurant clusters, the hospital and hotel drop-offs, and the parking dead zones. Twenty unpaid minutes of scouting routinely saves hours of paid confusion later.
Honesty cuts both ways, so here is the other side. Delivery is probably the wrong move if any of these describe you:
Delivery can carry you through a tight season, but the ceiling on every app in this comparison is the same. If you are driving because the day job stopped fitting, spend one evening finding out what does fit: the career assessment at RealWorldCareers maps your cognitive strengths to work with a far higher ceiling than any delivery zone.
Whichever you choose, run it like a business for two weeks: log every mile, record every payout, compute your true net hourly, and then decide with real numbers whether the parking lot at 6 p.m. deserves you. The apps will always tell you their best story. Your spreadsheet will tell you the true one.
Most income advice stops at gigs and stacking hours. The bigger move is matching your work to how your brain actually performs. RealWorldCareers measures your cognitive strengths and shows the careers your brain was built for.
Find the career your brain was built forThere is no single winner because pay is dominated by your market. In dense cities, DoorDash and Uber Eats during lunch and dinner rushes usually lead. In suburbs and small towns, Walmart Spark and Amazon Flex often out-earn them. The reliable approach is to run two or three apps for two weeks, track miles and payouts, and keep whichever produces the best net hourly for you.
It can be, if you select offers carefully. Drivers who hold out for roughly a dollar per mile or better and work peak meal windows commonly net $13 to $19 per hour in a fuel-efficient car. Drivers who accept everything in a truck or financed SUV can fall under $10. The car you drive and the offers you decline matter more than the app itself.
Yes. Tax forms are a reporting mechanism, not a definition of taxable income. All net profit from gig driving is taxable and reported on Schedule C regardless of whether a 1099-K or 1099-NEC arrives. Most drivers should set aside 20 to 25 percent of payouts and review the IRS rules on quarterly estimated payments.
Yes, and most experienced drivers do. The standard system is to stay online on two or three apps, accept the first offer that meets your per-mile floor, and immediately pause the others. Never carry overlapping orders you cannot deliver promptly, since late deliveries put your account standing at risk on every platform involved.
Restaurant delivery typically involves 15 to 30 miles per active hour, so a 20-hour-per-week driver can add 15,000 to 25,000 miles per year. That mileage is a deductible business expense at the IRS standard rate, but it is also real depreciation, which is why delivery profits are structurally better in an older, paid-off, fuel-efficient car.
Personal auto policies often exclude commercial delivery activity, and the apps' own coverage is limited and varies by platform and by whether an order is in progress. Many insurers sell an inexpensive rideshare or delivery endorsement that closes the gap. Telling your insurer is the safer and usually cheap option; discovering an exclusion after a crash is the expensive one.
One smart money idea each week, charts included. Join free and get the printable 2026 Money Calendar in your welcome email.
