I know it left Sheffield last night. I know it got sorted in Birmingham. I know it got to Croydon. I know it’s on a van on its way to me. The only thing I don’t know is the one thing I want to know: when is it going to get here?
That’s partly because there is something genuinely unknowable about this: predicting where a van touring the back streets of south London is going to be at any point of a journey stretching over many hours is a mug’s game, even assuming that the drivers solve the travelling salesman problem in their heads each time they drive out of the depot.
But I suspect that there is another reason too: the reason I am interested in parcel tracking is not the reason the delivery company is interested in tracking. I want to know to know as precisely as possible when it’s going to get here and don’t very much care where it’s been along the way. The delivery company wants to keep track of it along the route, knowing where it is and who is responsible for it at each stage, but it doesn’t matter to them precisely when it arrives within a particular delivery window. Not very surprisingly, their logistics systems are good at answering the questions they are interested in and not very good at answering the questions I am interested in.
There’s one more thing which is interesting about all this: I am receiving a service, but I am not the customer.
The customer is the company which sold me the stuff which is now either just round the corner or somewhere the other side of London. They can influence the delivery company’s behaviour by their purchasing and contracting decisions. Presumably one factor they take into account is any complaints they get about delivery issues, but unless the delivery company screws up pretty comprehensively, that is unlikely to weigh very heavily.
So what I get to see is some of the internal tracking information the company has anyway. That doesn’t cost them much to implement but gives the appearance of being an added service. I have no means of giving signals to delivery companies which encourage them either to create better quality information or even to let me see or use information they might already have but either haven’t worked out that I might be interested in or haven’t bothered to make available. I am not sure what that might be – I can’t know what I don’t know – but some combination of GPS tracking of the van and position in sequence (if mine is delivery 50, then knowing whether the most recent delivery is 1 or 49 tells me quite a lot) might give me a much clearer sense of what is going on – and much greater confidence that something actually is going on.
There is, of course, a public service moral to this fable. Making the data you have available is a good thing. It’s also relatively easy, so there is no reason not to do it. Building services which make use of that data is also a good thing. But even those two together don’t necessarily produce the optimum result, because the data may not have been what was most wanted or needed in the first place. And even with the extraordinary creativity of the people who have been turning open data into applications, there are still not enough ways for service users to act as customers, and still not enough being done to compensate for that by involving them much more directly in the process.
And I am still waiting for my parcel.