Ask ten small fleets how to plan a delivery route and nine will describe some version of the same ritual: a list of addresses, a map, and somebody’s gut feeling about which order “makes sense.” I’ve spent years building routing models for real operations, and the gap between a gut-feel route and an optimized one is rarely small — it’s usually measured in whole driver-hours per day. The good news: planning delivery routes properly is a repeatable process, and most of it isn’t even math. It’s preparation.
Start with clean stops, not clever software
Every routing disaster I’ve seen started upstream of the algorithm. Before you optimize anything, get your stop list honest: full street addresses that actually geocode (apartment complexes and industrial parks love to defeat geocoders), a realistic service time per stop — the minutes at the door, not just the driving — and, if capacity matters, how much each stop takes off the truck. A solver fed vague inputs produces a precise-looking plan for a fictional world. Ten minutes of address hygiene buys you more than any algorithm upgrade.
Decide your constraints before you sequence anything
A delivery route isn’t just an ordering of stops; it’s an ordering that survives your real-world limits. Write them down explicitly: which customers only accept delivery in certain windows, how much each vehicle carries, and how long a driver can legally and humanly be on the road (for commercial drivers in the US, the FMCSA’s hours-of-service rules set hard ceilings). The constraints are where routes actually get hard — sequencing 40 unconstrained stops is easy; sequencing 40 stops where six have morning-only windows and the truck fills up twice is the real problem.
Sequence with math, not with the map
Here’s the part that surprises people: humans are decent at ordering a dozen stops and terrible past that, because the problem explodes combinatorially. Eyeballing also fails quietly — a route that looks tight on a map can hide a river crossing or a left turn across rush-hour traffic (straight lines on a screen are optimists). And the popular workaround has a wall: Google Maps caps you at 10 waypoints, which is exactly when route planning starts to matter.
An optimizer does two jobs at once: it decides which stops belong to which vehicle, and then the order within each route — while honoring every constraint you wrote down. It works from real drive times, not straight-line distance, which matters more in last mile route optimization than anywhere else, because the last mile is where the road network is at its most spiteful.
A workflow you can copy this afternoon
This is the exact loop I’d hand a dispatcher who wants to stop planning delivery routes by hand:
- Put your stops in a spreadsheet: address or coordinates, demand, service minutes, and any time window per stop.
- Add your vehicles: how many, what capacity, when they start.
- Run it through a solver — my free route optimizer takes an Excel workbook or a Google Sheet and comes with a 113-stop sample so you can see the format before committing your own data.
- Read the result critically: check the dropped-stops list, the route durations, and whether the time windows were actually met.
- Download the stop-by-stop CSV and hand it to your drivers — or fix an input you got wrong and re-run. The second solve is where the plan gets good.
The re-run step isn’t an afterthought. In my experience the first optimization mostly reveals data problems (a stop geocoded to the wrong side of the state has a way of announcing itself), and the second one delivers the route you’ll drive.
The trade-offs a good plan gets right
Optimizing delivery routes isn’t one objective — it’s a negotiation between several, and you should know what you’re trading. Fewer trucks means each remaining route runs longer; a shorter total day sometimes costs an extra vehicle. In my own model every vehicle carries a fixed cost and every driving minute a variable one, and moving those two dials against each other changes the answer more than any other setting. Then there’s the uncomfortable question nobody plans for: what happens when a stop simply doesn’t fit? A rigid plan declares the whole day infeasible; a good one prices the miss — my tool charges a penalty per dropped stop, so an impossible customer shows up explicitly on a dropped list instead of silently wrecking every other route to accommodate them.
There’s also a human dial the math won’t set for you: route balance. The cheapest total plan sometimes gives one driver a brutal day and another a short one (drivers notice this long before the spreadsheet does). If you run the same crews daily, it’s worth accepting a slightly costlier plan that spreads the load — turnover is a routing cost too, it just doesn’t show up in the distance matrix.
When to re-plan — and when to leave it alone
Routes decay. Customers move, demand shifts, a new stop lands in the middle of nowhere. My rule of thumb: re-optimize whenever the stop list changes meaningfully, and re-check a “stable” route monthly, because drift is invisible day to day. What you shouldn’t do is fiddle a working plan by hand every morning — that’s how optimized routes quietly turn back into gut-feel routes.
A warning from someone who’s been burned: software fails at the worst moments, so keep yesterday’s plan until today’s is confirmed. My own optimizer once shipped an update that ran perfectly on my machine and hung only in production — the one environment that mattered. The dispatchers of the world deserve a fallback, and so do you.
How to plan a delivery route, compressed to one sentence: clean your stops, write down your constraints, and let a solver do the sequencing. The routing tool is free, needs no signup, and takes the same spreadsheet you’re probably already keeping. Feed it a real day of deliveries and compare the total drive time against what you’d have drawn by hand — that comparison is the whole argument.
Things that I use, like, and am affiliated with:
Mint Mobile offers great cell phone service for $15 flat, get $15 off using the link. Get discounted phones with service activation and no contract.
I never spend money before I check Mr Rebates or Rakuten to get cashbacks, rebates, discounts, coupons or cheaper gift cards.