Two orders leave your kitchen at 7 p.m. One goes four blocks, and the courier is back in twelve minutes. The other crosses the highway: twenty-five minutes out, a cold burger on arrival, a courier gone for the whole dinner rush. Both paid the same delivery fee. That flat fee, stretched over a circle on a map, is one of the margin leaks restaurants notice last and fix fastest.
The circle is lying to you
A radius zone assumes your city is a flat, empty plain. It isn't. A 3-mile circle happily includes the district across the river with one bridge, the neighborhood behind the rail yard, and the office park that takes twenty minutes to escape at 6 p.m. A scooter doesn't fly, and it doesn't travel in straight lines, so the circle promises reach the road never delivers.
You pay for that in specific ways. Couriers burn their shift on geometry, where one cross-river order eats two or three near-side deliveries in the same hour. Promise times inflate to cover the worst case, so a customer ten minutes away sees "45–60 min" and orders less. Meanwhile the P&L blends everything together, so you never spot which corner of the map is bleeding.
Polygons: draw the zone the way the city actually works
The fix is to draw shapes instead of circles. A polygon can follow the real streets: it hugs the river instead of jumping it, skips the block that turns into a parking lot at rush hour, and stretches down the avenue where a scooter genuinely moves. A rule that holds up: zone 1 covers everything you can reach in about 10 minutes of real driving, zone 2 in about 20. A pocket that takes 25 minutes at dinner belongs in zone 3, or outside the map entirely.
That's how zoning works on our platform out of the box. In logistics management you draw the polygons on a map, and each one carries its own delivery fee, minimum order, and promised time. Whoever assigns couriers next, a person or the auto-dispatcher, is working with those boundaries already priced in. Restaurants that move from one circle to two or three polygons tell us the same thing: promise times drop in the core, and the far orders start paying their own way.
Bring your address and your delivery map. We'll draw the polygons together on a 30-minute call.
Price each zone on its own cost
With an honest map, you can set honest prices. A courier drop costs real money, and it roughly doubles with each ring out: figure $2.50–3.50 in the core, $4–6 in the middle ring, and $6–9 at the edge once you count the drive back and the orders that courier couldn't take while gone.
Each polygon gives you three things to set. The delivery fee should track the real cost: low or free in the core, where delivery is cheap and the competition is fiercest, and higher out at the edge. The minimum order is what protects the far rings. A $15 ticket to zone 3 loses money at almost any fee a customer will accept, so set the zone 3 minimum where the math actually clears, around $30–40 for most menus. And the promised time is where you stop averaging. Near customers get the fast, honest number; far customers get a real one they can take or leave. Operators keep noticing the same thing: a far customer promised 55 minutes and served in 50 orders again, while one promised 35 and served in 50 doesn't.
Review your zones every quarter
You don't set zones once and forget them. Demand moves, your courier roster changes, a competitor opens two streets over. So once a quarter, pull three numbers for each zone from your reports: margin per order after the real delivery cost, average delivery time against what you promised, and repeat rate. What you're hunting for is a pocket that brings in decent revenue but has ugly margin and repeat numbers. When you find one, shrink the polygon, raise that zone's minimum, or split it off and price it for what it actually costs. The same review grows zones too: when zone 2 keeps beating its promised times, carve out the next ring and see if it earns a place.
There's one more zone worth having ready: a temporary one. Rain, a big match night, or a courier calling in sick cuts your real capacity for a few hours, not for a quarter. If you can switch to a tighter "bad day" polygon and switch back the next morning, your promise times stay honest through the worst shifts, and one chaotic Friday doesn't burn a month of reviews.
If this is your first delivery launch, the order matters: the zone comes before you choose software, hire couriers, or spend on marketing. Our two-week launch checklist walks through the full sequence.
Do this next
Pull last month's orders and put the ten slowest deliveries on a map. If they bunch up in one place, across a bridge, behind a highway, in that one stubborn corner, you've just found the polygon edge your circle was hiding. Redraw the boundary and reprice that pocket this week. It's a change in settings, and you can be done before Friday.
Zoning, courier tracking, and auto-assignment — configured with your map in about two weeks.