
The last fifteen metres: why the handover is the bottleneck
Between the kerb and the flat door sits the part of delivery that no datasheet describes. Intercom, door release, stairwell, parcel box and proof, and why that is an integration item.
Every discussion about delivery robots is about locomotion: wheels or legs, kerb or stairs, range and payload. That is understandable, because it films well. The part where projects actually get stuck comes afterwards, and it begins where the vehicle stops.
Key Takeaways
- 1The stretch from kerb to door is rarely longer than fifteen metres and contains almost every open question in a delivery.
- 2No delivery-robot manufacturer publishes a protocol for intercom and door-entry systems or names a door-system partner.
- 3The same applies to lifts. Wherever robots run inside buildings, lift integration is routinely a per-property item.
- 4Every documented delivery pilot ran with a human on site. That points at the handover, not at locomotion.
- 5If you cannot find the handover in the offer, the offer is not complete.
What happens in those fifteen metres
A courier stops, takes the parcel, walks to the door, finds the right bell, waits, speaks, is let in, climbs to the second floor, hands over or leaves the parcel, gets a signature and walks back. In daily practice this takes a few minutes and contains half a dozen separate operations, all of which assume a human being who can improvise.
For a machine, each of those operations is a separate interface. The intercom is a third-party system. The door release is a third-party system. The lift is a third-party system. The parcel box is a third-party system. The proof of delivery has to flow back into your dispatch. And the weather cares about none of it.
This is exactly why the published pilots are so instructive: they all ran with a courier alongside. Not because the robot could not drive, but because a human operates those six interfaces without a project.
The gap nobody documents
Put delivery-robot datasheets side by side and you will find payload, speed, step height, range and ingress protection. What you will not find is any statement about how the machine rings a doorbell.
That gap is remarkably consistent. No manufacturer publishes a protocol for intercom and door systems, and none names a maker of such systems as a partner. We know the same pattern from inside buildings: no robot manufacturer publishes a lift protocol or names a lift partner either.
This is not a criticism of the manufacturers. It is in the nature of the thing that a series product cannot make a statement about an installation that differs in every building. But it is a clear statement about where the work sits, and who does it.
Why this is a per-site item
A product feature is something you buy with the machine. An integration item is something that has to be built once for your site. The handover belongs in the second category, for a simple reason: there is no standard intercom, no standard door release and no standard parcel box.
For your business case that has two consequences. First, every offer needs a line for the door side, separate from the hardware. Second, the effort scales with the number of different entrance types, not with the number of robots. An operator with thirty identical entrances has a different project from one with thirty different ones.
Separate this early and you get a defensible number. Hide it in a flat rate and you will meet it again later.
The pragmatic route: start where the door is yours
On your own ground, a large part of the problem disappears. On a campus, a works site, hospital grounds or a new quarter, the entrances, the installations and the rules belong to you. You can integrate the door release because you own the door release. You can place a parcel box because you decide where it goes.
This is not the small solution, it is the right order. The handover you solve there is the same one that arises later in the public realm. The reverse is not true: a pilot on a public pavement teaches you little if the door question stayed unsolved inside it.
It has a second advantage: on your own site the approval question is answerable, whereas in the public realm it is a project of its own.
What to insist on in an offer
Ask specifically about the door side. Which intercom is installed, and how is it addressed. Who supplies the door release, and who is liable for it. What does the proof of delivery look like, and which system does it land in. What happens when nobody answers. What happens in rain and in snow.
Ask about remote supervision as well: how many units does one operator handle, and what does that person cost. It was present in every documented pilot, and it is a real operating-cost block.
A vendor who answers these precisely has understood the project. A vendor who points at step height instead has sold you a machine, not an outcome.
FAQ
- Why is the handover harder than the drive?
- Because the drive is a closed problem and the handover is an open one. On the route the robot meets surfaces and obstacles that can be measured. At the door it meets third-party installations, third-party rules and people who are not at home. That is why every documented pilot ran with a courier on site.
- Is there no standard for intercom and door systems?
- Not in a sense a robot could rely on. There are many systems from many manufacturers, some decades old. No robot manufacturer publishes a protocol for them or names a partner. That is precisely where the integration work arises.
- Is a parcel box better than door integration?
- Often yes, and in many projects it is the most pragmatic answer. A defined drop point resolves the bell, the door and the presence question at once. It does shift the problem to proof of delivery and to who supplies, maintains and empties the box.
- How much effort is the door side?
- It depends on the number of different entrance types, not on the number of robots. Thirty identical entrances are a small project, thirty different ones a large one. This is why the item belongs in the offer separately rather than inside a flat rate.
- What does werob actually do here?
- We specify the handover as its own work package: a survey of the installations, integration of intercom and door release or of a drop point, proof of delivery, and the feedback into your dispatch. Manufacturer-independent, because the installations on your site are a given and not ours to choose.