Routing a delivery van is a shortest-path problem. Routing a snow crew is not — it is a priority problem wrapped in deadlines and reshaped by the storm as it happens. A hospital entrance that has to be open at all costs and a low-priority strip mall that can wait until noon are not two stops to be optimized for distance; they are two different obligations, and the route has to honor that before it worries about miles.
This is how to build a route for 30–40 sites that holds up under an actual storm: sorting sites into priority tiers, clustering to kill the empty miles between them, running one route during the snowfall and a different one for cleanup, and — the part most operations get wrong — keeping the plan somewhere the drivers can actually see it change in real time.
Why snow routing is not package routing
A package route optimizes one thing: total distance to hit every stop once. A snow route juggles three constraints at once. Each site has a service-level deadline (open before 7am, or a two-hour response), sites carry different priorities that override distance, and you may visit the same site several times in one storm. Solve for distance alone and you will drive an elegant loop that leaves your highest-priority client unplowed at opening because it happened to be last on the efficient path.
So the route is built priority-first, then optimized for distance within the priorities. You are not asking "what is the shortest loop?" You are asking "what is the shortest way to keep every tier serviced inside its deadline?" — a different question with a different answer.
Sort every site into a priority tier
Before you sequence anything, tier your book. A simple three-tier scheme carries most operations:
| Tier | Examples | Obligation |
|---|---|---|
| 1 — Critical | Hospitals, 24h retail, life-safety, before-open entrances | Keep passable throughout; open before the deadline, no exceptions |
| 2 — Standard | Offices, most commercial lots, before-business-hours | Serviced by opening; maintained through the storm as capacity allows |
| 3 — Flexible | Overflow lots, low-traffic sites, residential | Cleared after tier 1 and 2 are handled; some latitude on timing |
Tier assignment is a business decision, not a driver decision. Set it when you write the contract — the response-time promise you sold is the tier — so nobody is guessing priority at 3am. A site the salesperson quoted a two-hour SLA on is tier 1 whether or not it "feels" important on the route.
Cluster to kill deadhead miles
The empty drive between two sites — deadhead — is pure cost. No plow is down, no revenue is earned, only fuel and hours are burned. On a 35-site book the difference between a clustered route and a scattered one is measured in hours per storm, every storm, all winter.
Cluster geographically, then honor priority within and across clusters. The tension is real: your tier-1 sites are rarely conveniently grouped, so a pure priority order can send a truck back and forth across town while a pure geographic order misses deadlines. The resolution is usually to assign a truck to a geographic zone and sequence tier-1-first inside that zone, rather than running one truck on a citywide priority list. Fewer, denser zones beat one heroic route.
If your tier-1 sites are scattered across the whole service area, that is a dispatch design problem, not a sequencing problem. Either dedicate a truck to the critical sites or restructure zones so each truck owns a cluster it can keep serviced. No clever ordering fixes a route that spans the whole city.
The storm has two routes, not one
During active snowfall, the job is not to finish sites — it is to keep tier-1 (and tier-2 where you can) passable and open. That means repeated maintenance passes: a lighter, faster loop that hits the critical sites again and again so they never accumulate past safe, accepting that a full clean-to-pavement is impossible while it is still snowing.
After the snow stops, the route flips to cleanup: full clears, widening lots back to their real footprint, cutting back the stacked piles, squaring up corners, and the final salt. This is a different sequence — often you work outward from the priority sites you kept open toward the flexible ones you deferred. Trying to run one route for both phases is why crews either over-serve low-priority sites mid-storm or leave critical ones to drift.
Plow runs and salt runs are different routes
A salt or brine run and a plow run are not the same sequence and often not the same vehicle. Pre-treatment runs before the storm on a route optimized for the anti-icing window; plowing runs during and after; a final salt run chases the cleanup. Bolting salt onto the plow route as an afterthought means either salting too early on sites still accumulating or too late on sites already refreezing.
Plan the salt sequence deliberately against pavement conditions, which is a topic of its own — see when to pre-salt by pavement temperature. The point for routing is simply that "the route" is really several routes for several jobs, and they peak at different moments in the storm.
The plan has to be live, or drivers improvise it away
Here is the failure mode nobody plans for: a route printed in October is wrong by January. Sites were added and dropped, a client renegotiated to tier 1, a truck went down and its zone got redistributed, priorities shifted after a complaint. A driver holding a stale paper route at 3am does the reasonable thing — improvises — and improvisation re-introduces the exact deadhead and missed deadlines the plan existed to prevent.
The route has to live somewhere every driver sees the current version, including which sites are already done and which truck is where. When a driver can see that the tier-1 site two zones over is still open because a truck broke down, the dispatch can be corrected in the moment instead of discovered at 8am. A route is not a document you hand out once; it is a live picture of the storm that the whole crew reads from and updates as they work. That is the difference between a plan and a wish.
Frequently Asked Questions
How many sites should one truck handle in a snow route?
It depends on site sizes, service-level deadlines, and drive distances, not a fixed count. The real limit is whether the truck can keep its tier-1 sites open inside their deadlines during peak snowfall. Build zones around that constraint rather than a target number of stops.
How do I sequence a snow route?
Priority first, distance second. Tier your sites (critical, standard, flexible), assign each truck a geographic cluster, and sequence tier-1-first within the cluster. Optimizing purely for shortest distance will leave a high-priority client unserviced at opening.
Should I always plow the highest-priority site first?
During the storm, keep tier-1 sites passable throughout with repeated maintenance passes rather than fully finishing one site before touching the next. The goal mid-storm is that no critical site accumulates past safe, not that any single site is perfect.
Should the route be different during the storm versus after?
Yes. During snowfall you run a fast maintenance loop that keeps priority sites open; after it stops you switch to a cleanup sequence — full clears, widening, cutting back piles, final salt — usually working outward from the critical sites. They are two different routes.
Is a paper route good enough, or do drivers need it live?
Paper goes stale fast — sites change, trucks go down, priorities shift — and a driver with an outdated route improvises, which re-creates deadhead and missed deadlines. Drivers need the current sequence plus live status of what is done and where each truck is, so dispatch can adjust mid-storm.
Run your snow business on Icey
Dispatch, GPS, salt logs, quoting, and e-sign contracts in one place. Free for solo operators — no demo call, live in about 10 minutes.
Start free