If you run a contract parcel round, the work arrives as a manifest. A list of addresses, sometimes a suggested order, usually a delivery window on some of them. What it is not is a route.
Most contractors sequence that list themselves, either the night before or in the van before the first drop. On a 120-drop round, doing that well is the difference between finishing at three and finishing at six.
Why the manifest order is rarely the driving order
Manifests tend to come out in the order the parcels were processed, or grouped by postcode. Neither has much to do with how you drive.
Postcode grouping in particular looks sensible and is not. Australian postcodes can cover a lot of ground, and two addresses in the same postcode can be on opposite sides of a river, a rail line or a highway with no crossing for four kilometres. Sorting by postcode gets you a list that looks tidy and drives badly.
The constraints that actually bind on a contract round
Delivery windows. Businesses that close at noon. Medical drops that must land in the morning. If your planning tool cannot express a window, you are holding it in your head all day.
Time at the door. A signature-required parcel, a locker drop and a leave-at-door are three different amounts of time. A round planned on drive time alone will run consistently late by mid-afternoon, and you will not know why.
Vehicle capacity. If you run more than one van, what each can carry decides how the work splits. Splitting by suburb and hoping is how one driver finishes early while another is still out at six.
Proof of delivery. Contract work usually requires it. Photos scattered through your camera roll are not an audit trail when a claim comes in three weeks later.
Why 25 stops is a number worth knowing
Most consumer route planners stop optimising somewhere around 25 stops. That limit is not arbitrary and it is not a licence tier. It comes from the mapping APIs underneath, which will only re-order about that many waypoints in one request.
Past that, apps generally chop your day into chunks and optimise each chunk on its own. The chunks are fine. The joins between them are where the doubling back happens.
For a contractor doing 80 to 200 drops, that limit is the whole problem. It is worth checking directly before you commit to any tool: put a real 100-stop manifest in and see whether the order that comes back makes sense end to end, or only in patches.
What RouteMate does with a contract round
RouteMate runs a vehicle-routing solver rather than a waypoint re-orderer, so the whole manifest is sequenced in one pass. We measured a 498-stop route solved in 8.7 seconds against production, over real Australian roads.
Practically, that means:
- Paste or import the manifest, or photograph labels for the ones not on it
- Mark the drops that have a window, and how long you actually spend at each type
- Optimise once, for the whole round, not in chunks
- Assign runs across vans by shift hours and starting depot rather than by suburb
- Proof of delivery attaches to the stop, so the audit trail builds itself
Being straight about the limits
RouteMate does not integrate with Australia Post's systems. You bring the manifest across, by CSV, by scanning, or through the API if you have something that can push it.
And route optimisation does not fix an overloaded round. If the work genuinely does not fit in the hours, better sequencing narrows the gap but does not close it. What it does do is tell you honestly which drops cannot make their windows, before you drive, rather than after a customer rings.
Where to start
Take one real round, the messiest one you have, and sequence it. Compare the order that comes back against what you would have done yourself. If it does not beat your own judgement on a round you know well, no feature list is going to make up for that.