“Deliver by 11” sounds like a trivial request. Multiplied by fifty customers, each with its own slot, it becomes the constraint that breaks any plan built by hand. Delivery time windows are the point where the planner’s experience is no longer enough: too many combinations, too many interlocks, too little room for error.
Let’s look at what they really are, why they’re so hard to manage, and how to meet them without stretching the routes.
What delivery time windows are
A delivery time window is the time slot within which a customer can or wants to receive goods — for example “8:00-12:00” or “only after 15:00”. They can be hard (binding, a miss means a failed delivery) or soft (preferences), and a single customer may have more than one in a day.
They’ve become increasingly common: grocery chains impose unloading slots, pharmacies receive at precise times, restaurants don’t want deliveries during service, and in cities low-emission zones add access windows. The result is that each stop no longer has just a location, but also a when.
Why they break manual planning
Without windows, building a route means ordering stops by proximity. With windows, every choice about order ripples through all the others: moving one customer earlier can make it impossible to serve another in time. The planner ends up solving a puzzle in their head where:
- Each stop has one or more allowed slots.
- Travel times between stops change with traffic.
- Service times (unloading, signing) eat into the window.
- Vehicle capacity limits how many stops fit in a route.
Beyond a few dozen stops, finding by hand the sequence that respects every window and minimizes distance is practically impossible. You end up accepting routes that “work”, not optimal ones.
What window mistakes cost
Getting a window wrong isn’t free:
- Failed delivery: the customer isn’t there or won’t accept off-hours, and the parcel returns to the warehouse — a second trip that wipes out the delivery’s margin.
- Waiting and idle time: arriving too early means waiting; too late means losing the slot.
- Penalties and disputes: in grocery and structured B2B, missed slots have contractual consequences.
- Domino effect: a delay early in the route propagates to every stop after it.
Managing windows is therefore a cost issue, not just a service one.
How to manage time windows while optimizing routes
- Collect the real windows for each customer, distinguishing hard from soft.
- Estimate service times by stop type: not all deliveries take the same time.
- Use realistic travel times, ideally with traffic, not straight-line distances.
- Let optimization fit the stops together: the engine builds the sequence that meets the windows and minimizes km.
- Flag conflicts in advance: when a customer asks for a window incompatible with the route, it’s better to know before, not on the road.
It’s exactly what good route optimization software does: it treats time windows as a constraint of the calculation, alongside capacity, priorities and urban restrictions. For the full picture see the guide on how to optimize delivery routes.
Frequently asked questions
What is the difference between a hard and a soft time window?
A hard window is binding: delivering outside that slot counts as a failed delivery. A soft window is a preference: meeting it improves service, but it can be violated if necessary, usually with a “light” penalty in the calculation. Distinguishing the two lets optimization build realistic routes without being needlessly rigid.
Can you meet time windows and reduce kilometres at the same time?
Yes, and that’s exactly the job of route optimization. The two seem to conflict, but an optimization engine looks for the sequence that satisfies the windows while minimizing total distance. Manual planning tends instead to sacrifice one for the other.
How do you handle customers with several windows in a day?
Good software lets you attach several slots to the same stop and automatically picks the one that fits best in the route. It’s a frequent case (e.g. morning or afternoon) that is very costly to evaluate by hand.
Do low-emission zones count as time windows?
In practice yes: access to a restricted zone is only allowed in certain slots, so it behaves like a window that applies to the area rather than the single customer. Zone-aware optimization treats them together with customer windows.
Want to see how optimization fits your time windows together? Discover OptivoRoute or book a demo.