Key takeaways
- Set the outer edge of delivery by drive time on a busy night, not by miles on a map.
- Let your most fragile menu item decide the limit, then pack or cut that item if you want to reach further.
- Use two or three zones, each with its own fee, order minimum and quoted time, so far orders pay for the extra driving.
- Guests rank speed as critical for off-premises orders, so a smaller zone you can serve on time beats a big one you cannot[1].
- Review zones monthly against late deliveries, refunds and driver hours, and shrink them on peak nights if needed.
Why the radius is a food-quality decision first
Most restaurants pick a delivery radius by drawing a circle on a map: three miles, five miles, whatever the last place did. The better starting point is the food. Every dish has a window between leaving the pass and turning soggy, cold or congealed, and your radius should sit inside that window on your slowest realistic night.
Guests notice. In National Restaurant Association research, 94% of consumers said speed is critical when ordering off-premises, and over 9 in 10 named customer service as a top priority[1]. A late, lukewarm order costs you a repeat guest, and on a marketplace app it can cost you a rating as well.
94%
of consumers say speed is critical when ordering takeout or delivery[1]
Delivery is now common enough that a sloppy radius affects a large share of your business. 65% of limited-service operators offer delivery[2], and among the 37% of full-service operators who deliver, 57% say it is a larger share of sales than in 2019[3].
Step one: find your real travel-time limit
Before you touch a map, decide how long each part of your menu can travel. Use the slowest-holding item that sells often, not the sturdiest.
| Menu item type | How it fails | Example travel limit to test |
|---|---|---|
| Fried items, fries, tempura | Steam softens the crust | 15 to 20 minutes |
| Pizza and flatbreads | Crust softens, cheese sets | 20 to 25 minutes |
| Bowls, rice, curries, braises | Holds heat well | 30 to 35 minutes |
| Sandwiches and wraps (cold) | Bread soaks through | 25 to 30 minutes |
| Plated proteins with sauce | Sauce breaks, meat overcooks | 20 to 25 minutes |
| Soups and noodles in broth | Noodles swell if packed together | 25 minutes if packed separately |
Run a simple test: pack three orders the way your line normally does, put them in a car and drive to three addresses at different distances. Open each bag on arrival and score texture, temperature and look out of five. The furthest address that still scores four or better is your travel limit for that dish.
Then subtract time that is not driving. Order handoff, finding parking, walking to a door and waiting in an apartment lobby all count. If your limit is 25 minutes and handoff plus door time takes 7, the drive itself can be about 18 minutes.
Step two: draw zones by drive time, not by miles
A three-mile circle around a downtown restaurant can take ten minutes to cross in one direction and twenty-five in another. Most mapping tools and online ordering platforms let you draw a custom polygon. Use that instead of a radius.
Map drive times at your peak hour
Check typical drive times from your door at 7 pm on a Friday, not at 3 pm on a Tuesday. Mark the line where a driver reaches the edge in your drive limit.
Cut out the slow pockets
Remove areas behind rivers, rail lines, highway interchanges, gated communities or large campuses with hard-to-find entrances, even if they are close on the map.
Add the quick pockets back
Include neighborhoods just past the line if a main road makes them faster to reach than closer streets.
Split into an inner and an outer zone
The inner zone is everything a driver reaches in about half your limit. The outer zone is the rest. Add a third zone only if you have a clear reason, such as an office park you want to reach at lunch only.
Name zones the way drivers and hosts talk
Use labels like Zone A (near) and Zone B (far) so staff can quote the right time and fee on the phone without opening a map.
Step three: give each zone its own fee, minimum and quote
Zones only help if they change something. The outer zone needs more driver time per order, so it should either earn more per order or take fewer small orders.
| Setting | Zone A (near) | Zone B (far) | Zone C (lunch only) |
|---|---|---|---|
| Drive time from door | Up to 9 minutes | 9 to 18 minutes | Office park, 15 minutes |
| Delivery fee to guest | Lower fee | Higher fee | Same as Zone B |
| Order minimum | Low minimum | Higher minimum | Group orders only |
| Quoted time, normal night | 30 to 40 minutes | 40 to 55 minutes | Scheduled slots only |
| Quoted time, slammed night | 45 to 55 minutes | Paused or 60+ minutes | Not offered |
| Fragile items allowed | All | No fried items | Cold items and bowls |
For the fee and minimum, start from what a driver hour costs you, including wages, mileage or reimbursement and insurance, and how many orders a driver completes per hour in each zone. If a far order takes twice the driver time, it needs to bring in roughly twice the delivery contribution. You can get that through a higher fee, a higher minimum, or both.
Example: a fictional pizza and wings shop finds a driver can complete about three Zone A runs an hour but only about 1.5 Zone B runs. It keeps the Zone A fee low to compete, doubles the Zone B minimum, and removes wings from Zone B because they failed the travel test after 20 minutes. Zone B orders fall slightly, but refunds for soggy wings stop and drivers stop running late on Zone A orders while they finish long trips.
Marketplace apps and your own zone map
If you also sell through third-party marketplaces, each app has its own delivery area settings and its own driver network. 68% of delivery customers ordered through a third-party service in the past six months[4], so these settings shape a large share of your delivery orders even if you never see the driver.
Those orders also cost more to fill. Two in three restaurants say they pay platform fees of 15% to 29.9% of the transaction[5], and in a May 2026 survey 47% of full-service and 43% of limited-service operators said they make no profit on third-party delivery orders[6]. A wide marketplace radius can add orders that lose money once refunds and remakes are counted.
Source: National Restaurant Association, 2026[6]
Show the data
| Item | Value |
|---|---|
| Full-service | 47% |
| Limited-service | 43% |
- Set each app's delivery area to match your own Zone A and Zone B, or smaller.
- Remove fragile items from the delivery menu on the app if they fail your travel test, even if they stay on the dine-in menu.
- Ask your account contact which settings you control and which the platform overrides at busy times.
- Compare refund and complaint reasons by distance once a month, using the order reports each app provides.
If you are still deciding between your own drivers and a delivery service, read in-house drivers versus outsourced delivery before you lock in the map, and check delivery driver insurance and liability before you extend drivers further from the door.
When to shrink the zone, even for one night
A fixed map is a starting point. The zone that works on a quiet Monday will break on a holiday weekend. Agree in advance on the triggers that let a manager shrink delivery without a debate.
- A driver calls out and you are down to one person on the road.
- Ticket times on the line run more than ten minutes behind the quote.
- Two or more late or cold-food complaints come in during the same hour.
- Weather slows traffic or makes parking and door drop-offs slow.
- A large catering or to-go order is in the kitchen at the same time.
When a trigger hits, pause Zone B first and keep Zone A open. Change the setting on your ordering site and every marketplace tablet at the same time, and tell the host so phone orders get the same answer. The guide to throttling online orders on busy nights covers how to pace orders rather than turning delivery off completely.
A monthly zone review in fifteen minutes
Put a recurring fifteen-minute review on the calendar. Pull these numbers for each zone and compare them with the previous month.
- Orders and sales by zone, and average order size.
- Share of orders delivered later than quoted.
- Refunds, remakes and complaints, with the reason given.
- Driver hours and mileage, split by zone if your system allows it.
- Items most often mentioned in complaints, and the zone they went to.
If Zone B is late more often and complaints cluster on one or two dishes, remove those dishes from Zone B before you shrink the zone itself. If Zone B is late across the board, pull the line in. If Zone A is growing and on time, consider whether a lunch-only or scheduled zone would bring in office orders without stretching dinner service.
Rule of thumb: a delivery zone is the area you can serve on time with your worst-traveling popular dish on a busy night. Anything beyond that is a promise you will break.
Frequently asked questions
How far should a restaurant deliver?
Far enough that a driver can reach the edge and back within the time your most fragile dish holds up on a busy night. For many restaurants that ends up being a drive-time limit rather than a fixed number of miles.
Should delivery zones be circles or drive-time shapes?
Drive-time shapes. A circle includes places across rivers, highways or one-way grids that take far longer to reach, and it leaves out quick addresses just past the line.
Should far zones have higher delivery fees?
Usually yes. A higher fee or a higher order minimum in the outer zone covers the extra driver time and keeps small orders from far away from losing money.
Do third-party apps use my delivery radius?
Marketplace apps generally let you set or request a delivery area, but they also apply their own rules. Check the settings in each tablet or portal and keep them in line with your own zone map.
When should I shrink my delivery zone?
When late deliveries, cold-food complaints or refunds rise, when you are short a driver, or on nights when the kitchen is already behind. Shrinking for one shift is often better than missing quoted times across the whole area.
References
National Restaurant Association (2025). From Trend to Transformation: Off-Premises Dining Now Essential for Restaurant Consumers, Operators (opens in a new tab) ↑
National Restaurant Association (2025). From Trend to Transformation: Off-Premises Dining Now Essential for Restaurant Consumers, Operators (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑
National Restaurant Association (2026). National Restaurant Association comments to the FTC on the Food Delivery Fees ANPRM (Project No. P267101) (opens in a new tab) ↑