How to Set Up Direct Delivery Without Third-Party Apps: Keep 100% of Every Order (2026)

8. august 2026, kell 20:15Tabres Team

direct delivery commission free online ordering restaurant delivery app fees first party ordering restaurant delivery radius

A $50 delivery order on a big app can land in your bank account as about $33. The food already cost you $15. That is third-party delivery in one line.

Short answer: you can set up direct delivery in about a week. Give delivery its own hours, draw a radius you can cover in 20 minutes, set a minimum order of $20 to $30, and charge a real delivery fee of $3 to $6. Then pick who drives — your own staff inside a tight radius, or an on-demand courier like Uber Direct or DoorDash Drive at a flat $6 to $9 per drop for anything wider. Put one ordering link everywhere: Google, Instagram bio, receipts, packaging. You keep the full order value minus 2% to 3% card processing. No commission, ever.

The tech is the easy part. The boring operational decisions are what most owners skip — and that is exactly where direct delivery goes wrong.

What direct delivery actually means

Direct delivery means the order comes to you, not to a marketplace. Your menu, your prices, your checkout, your customer. You get the name, the phone number, the address, and the order history. On an app, none of that is yours.

There are three ways to run it, and they are not the same thing:

Your own drivers. Staff or part-timers in their own cars. Cheapest per order in a small radius. You carry the insurance and the scheduling.
On-demand couriers (white label). You take the order, a courier network delivers it in your name. Flat fee per drop, no percentage of the food. Uber Direct and DoorDash Drive both work this way — roughly $6 to $9 a delivery in 2026, depending on distance, market, and time of day.
Hybrid. Own driver inside 2 miles, courier for everything past that. This is what most smart independents run now.

Notice what is missing from all three: a marketplace taking a slice of your food.

The real math on a $50 order

Numbers beat opinions. Same order, three ways.

On a marketplace, 25% commission tier:

Commission: $12.50
Card processing (about 3%): $1.50
Packaging: $2.00
Refund risk (2% average): $1.00
You keep: about $33.00

Direct, your own driver:

Card processing (2.9% + $0.30): about $1.75
Packaging: $2.00
Driver cost for a 15-minute run (wage share + mileage): about $6.00
Guest paid a $4 delivery fee, so $4 comes back
You keep: about $44.25

Direct, on-demand courier at $7.99:

Card processing: about $1.75
Packaging: $2.00
Courier fee: $7.99
Guest paid a $5 delivery fee, so $5 comes back
You keep: about $43.25

That is roughly $10 to $11 more per order. At 20 delivery orders a day, it works out near $75,000 a year. A chef's salary, sitting inside your fee structure.

Funny thing about this industry — owners will argue over a 40-cent case of romaine, then hand over $12.50 an order without blinking. If you want the full breakdown of what the apps really cost, we ran those numbers in detail in the real math of third-party delivery.

Step 1: Give delivery its own hours

Do not run delivery whenever the doors are open. That is mistake number one.

Delivery needs its own schedule, separate from your dining room hours. Most kitchens should shut delivery off during the peak dine-in rush, then open it again after. A ticket that sits in the window for 18 minutes because the line is buried arrives cold, and a cold order is a refund plus a bad review.

Pick your slow-to-medium windows. Weekday 11:30 to 1:00 and 5:00 to 8:00 works for a lot of kitchens. Turn it off on your busiest Saturday night until you can prove the line can handle it.

Step 2: Draw a radius you can actually cover

Twenty minutes, door to door. That is the honest limit for hot food.

In a dense city that might be 1.5 miles. In a spread-out suburb it could be 4 miles. Do not set it by distance on a map — set it by drive time at your busiest hour, with parking, elevators, and one-way streets included. Then drive it yourself on a Friday at 7 p.m. before you promise it to anyone.

A radius that is too big is the single most common reason direct delivery fails. Food goes out great and arrives sad. The guest blames the food, not the distance.

Step 3: Set a minimum order and mean it

A $14 order with a $4 fee is a losing trade. You cannot pay a driver $6 to deliver $14 of food and call it a business.

Set a minimum of $20 to $30 depending on your average ticket. A good rule: minimum order should be about 1.5 times your normal dine-in average ticket. It filters out the trips that lose money and quietly pushes people to add a side or a drink.

Show the minimum clearly on the menu page. Nobody enjoys building a cart and finding out at checkout.

Step 4: Charge a delivery fee that covers the drop

Free delivery is not a strategy. It is a discount you forgot to name.

Own drivers: charge $3 to $5. Your real cost per drop is usually $5 to $8 once you count wages and mileage, so you are sharing it, not covering it fully. That is fine — it is still far cheaper than a commission.
On-demand courier: charge $4 to $6 and absorb the rest. Guests already expect a fee; they see one on every app.
Distance tiers work. Flat $3 inside 2 miles, $6 beyond. Simple, fair, and it steers demand toward your profitable zone.

One thing to check: whether delivery fees are taxable where you operate. In some US states the delivery charge is subject to sales tax, in others it is not, and the rule can change based on whether the charge is separately stated. Confirm with your state's department of revenue or your accountant before you set it up. If you are in the EU, delivery often sits in a different VAT band than dine-in — we covered that split in VAT on dine-in, takeaway, and delivery.

Step 5: Decide who drives

Your own drivers make sense when you have steady volume, a tight radius, and someone to schedule them. Two things you must handle:

Insurance. A personal auto policy usually does not cover driving for business. You need hired and non-owned auto coverage (HNOA) added to your commercial policy. Call your broker before the first delivery goes out, not after the first accident.
Classification. Whether a driver is an employee or a contractor is not your choice to make freely — it depends on federal rules and your state's test, and states like California are strict. Ask a local employment attorney or your payroll provider. Getting this wrong is expensive.

Also reimburse mileage at the current IRS standard rate. Look up the figure for this year — it changes every January.

On-demand couriers make sense when volume is uneven, your radius is wide, or you do not want to run a fleet. You sign up with the courier service directly; it is a separate account from the marketplace app, even when the same company owns both. The delivery arrives in your name, and the order never touches their marketplace.

Start with couriers. Move to your own drivers later, when the volume proves it out.

Step 6: Build a delivery menu, not a copy of your dine-in menu

Some dishes do not travel. Anything crispy, anything with a foam, anything plated to be eaten in 90 seconds — leave it off.

Cut the menu to the 15 to 25 items that survive 20 minutes in a box.
Price delivery items 10% to 15% above dine-in. Packaging and driver time are real costs. Guests accept it.
Photograph everything. Delivery is a photo-driven purchase — good menu photos lift the order value more than any discount you can run.
Test one order yourself. Order it, wait 20 minutes, then eat it. That single test will change your menu.

Step 7: Get one link and put it everywhere

This is where most owners lose. They set up ordering, then hide it.

You need one short link that opens your menu and takes the order. Then it goes on:

Your Google Business Profile. Add it as the ordering link and the menu link. This is your biggest free traffic source by a mile.
Instagram and TikTok bio. One link, no PDF.
A QR code sticker on every takeaway bag and delivery box. Cheapest customer acquisition in hospitality.
The bottom of every receipt.
A table tent and the window. Dine-in guests are your best future delivery guests.

You do not need a full website for this. A digital menu with its own link does the job. Free platforms like Tabres let you switch delivery on per branch with its own hours, maximum distance, minimum order, and fee — no commission and no monthly bill. If you are wondering whether a real website is still worth it, we compared the options in free restaurant website alternatives.

Step 8: Win your app customers back

The apps have your customers. Go get them.

Insert card in every app delivery. "Order direct next time — 15% off, same kitchen, same drivers." Keep it to one line.
A real offer, not a coupon code nobody types. A QR code that lands straight on your menu with the discount already applied converts far better.
Never fund in-app promotions. You are already paying 25%. Adding a "spend $30, save $6" deal on top means you are paying for the privilege of losing money.
Be patient. A realistic goal is moving 20% to 30% of app volume to direct within six months. Owners who push hard on packaging inserts hit it faster.

Keep the apps on for discovery. Just stop treating them as your main channel. Marketplaces for new faces, direct for repeat business — that is the split that works in 2026.

Step 9: Watch four numbers

Every Monday, check these:

Direct orders as a share of total delivery. This is the number that matters. If it is not climbing, your link is not visible enough.
Average delivery ticket. Should be above your dine-in ticket. If it is not, your minimum is too low.
Average delivery time, door to door. Over 30 minutes and you have a radius problem or a kitchen timing problem.
Repeat rate. How many direct customers ordered twice in 60 days? Under 25% means the food is not travelling well.

A weekly ten-minute look at these beats a monthly deep dive nobody does. If you want a fuller list, we put together the KPIs worth checking every Monday morning.

Mistakes that kill direct delivery

Radius too big. Cold food, refunds, one-star reviews. Shrink it.
No packaging budget. A $0.30 container ruins a $22 dish. Spend the money on vented boxes and tamper seals.
No phone answer plan. Direct orders still generate calls. Somebody has to pick up.
Address entered wrong. Collect street, house number, floor, and a note for the driver. "Blue door, buzzer broken" saves ten minutes.
Launching quietly. If you do not tell anyone, nobody comes. Post it, sticker it, print it, say it at every table for two weeks.
Giving up in month two. Direct ordering compounds. The customers you convert in March are still ordering in October.

Direct delivery FAQ

Do I have to leave the delivery apps completely? No, and you probably should not. Keep them for discovery, run direct for repeat customers. Most independents doing this well keep both and shift the mix over time.

How much does it cost to set up direct delivery? Close to nothing if you use a free ordering platform. Your real costs are card processing at 2% to 3%, packaging, and either driver wages or a flat courier fee per drop.

Can I use app drivers without being on the app? Yes. Uber Direct and DoorDash Drive are courier services you contract separately. Your order stays yours; you just rent the driver. Flat fee, no commission on the food.

What delivery fee should I charge customers? Three to six dollars, depending on distance and who drives. Free delivery only makes sense above a high minimum, like $50.

Is my personal car insurance enough for deliveries? Almost certainly not. Personal policies usually exclude commercial use. Talk to your insurance broker about hired and non-owned auto coverage before you start.

Direct delivery is not really a technology project. It is a set of decisions about distance, timing, and what a delivery is genuinely worth to you. Get the radius honest, the minimum firm, and the fee real, and the rest is just repetition.

Start narrow. Two miles, 20 items, your slow hours, one visible link. Prove it works for a month, then widen it. The commission you stop paying does not show up as a headline — it shows up quietly, every single Friday, in the number left at the bottom of the report.