Handover Signatures

A captured signature in AWRA is three things stored together: the hand-drawn mark, the printed name of the person who signed, and the moment they signed it. The drawing supplies the human moment — somebody physically marked this, in front of somebody else. The name supplies an address you can act on. The timestamp places it against the delivery. Remove any one of the three and what is left does not settle an argument.

The reason to labour this is that the drawing is the part people think they are buying, and it is the weakest of the three. Nobody can read a signature; that is close to the definition of one. On its own it tells you that a person was present and willing to move a finger, which you already believed, because the stock is on the shelf.

This is why the printed-name field is not optional in practice even though it is optional in the form. A storekeeper who collects beautiful signatures and leaves the name blank has produced a folder of anonymous marks. When the supplier queries an 84-unit delivery three weeks later, "Grace Achieng signed for 84 at 14:32" ends the matter in one phone call; an unnamed squiggle starts a fortnight of messages instead.

The pad appears inside the movement you are already recording: a stock check-in (including receiving against a purchase order, which runs through the same screen), a stock check-out, an asset custody movement, or a stock transfer. Complete the movement first — item, quantity, warehouse or location, reason — then hand the device to the person receiving or releasing the goods. They draw with a finger, a stylus, or the mouse on a desktop. You type their name. Save.

It works with no signal, and this is a design decision rather than a happy accident. A loading bay is usually a metal shed with a concrete roof, which is exactly where the handover happens and exactly where coverage fails. The mark travels inside the movement record rather than as a separate file upload, so it rides the offline queue and posts intact when the device finds signal on the way out of the yard.

Worked example: a delivery arrives at a site with no coverage. You record the check-in for 84 units, the driver signs, you type "Grace Achieng", and you save. The phone shows the work as queued. Twenty minutes later the vehicle reaches the main road, the queue flushes, and the receipt posts with the mark on it. Nobody has to find Grace again, and nothing was stranded waiting for an upload to retry.

A signature that can be added at any time is not evidence, because the most tempting moment to add one is after the argument starts. So every mark has exactly one window in which it can be taken or replaced, and that window closes when the record is decided. A goods receipt or stock issue is signable while it is still pending; once approved, adjusted or rejected, somebody has relied on what it said and the mark is fixed. An asset custody handover behaves the same way, closing at approved, completed, rejected or cancelled.

A stock transfer is stricter, because each of its two marks has exactly one honest moment. The dispatch signature can only be taken while the transfer is pending — a dispatch mark added after the lorry has left is backdating, whatever the intention. The receipt signature can only be taken while the transfer is in transit — signing for goods that have not arrived is a promise, and a promise filed as a receipt is worse than no record, because it looks like one.

The practical consequence for an operator is simple: if the pad is not showing on a record you expected to sign, check the status before raising a fault. A closed window is the usual reason and it is deliberate. The operational discipline that follows is to collect the mark at the moment of handover rather than at the end of the shift, because by the end of the shift the record may well have been approved.

One test decides where a pad belongs: is a person physically present, and is something changing hands? Deliveries, stock issues, asset handovers and both ends of a transfer pass it. Expense approvals, budget sign-off and procurement authorisation do not — nothing is changing hands and nobody is standing there. Those decisions are made by an authenticated account against a permission that had to be granted, and written to the audit trail, which is a stronger record than a drawing and does not need one added on top.

Adding a pad to an approval would actively weaken it. A drawn mark carries cultural weight; people treat a signed page as more binding than a clicked button. Putting one on an approval screen teaches everybody that the drawing is where the authority lives, moving the perceived control off the mechanism that is really enforcing something. Then somebody notices the drawing proves nothing and concludes the approval was theatre — when it was the pad that made it look that way.

The other line to hold is between signing in person and e-signing at a distance. A vendor acknowledging a purchase order, a candidate returning a contract, a supplier completing prequalification — these want a real signature from somebody who is not in the room, which means a document sent, tracked and returned. That is remote e-signature through a connector, and it is not interchangeable with a pad. Understand both, and a policy that "requires a signature" can be answered with the right mechanism instead of the nearest one.