HERE'S WHY service doesn't get better - systems that managers install with the aim of improving service actually make it worse, as in the case of interactive voice response that we highlighted last week.
Take the case of a leaking water main experienced by an Observer reader. Repairing a leak (or anything else) is a common and straightforward type of service - something breaks, the customer asks for service, the company fixes it. But it's frequently not that simple. In this case, a new customer rang the water company to ask it to mend a leaking main (she knew it was leaking because the meter kept spinning when the stopcock was shut).
The company replied that someone would phone her back (within five days according to the customer, two according to the company) to arrange an appointment. No, she told them, it's an emergency: having just moved into the house, we don't know where the main lies and the leak may be damaging the foundations. Anyway, aren't we supposed to be conserving water?
No, replied the firm, an emergency is something like Boscastle it doesn't apply to domestic repairs. We'll phone you back to fix an appointment - which we guarantee will be within the following five working days. In other words, there could be a seven- or 10-day wait, depending on the versions, between reporting an incident and the first visit.
The first visit, because when the man from the water company arrived after what most of us would call two weeks (these are working days), he was a surveyor, not a plumber: a second visit would be necessary to fix the leak. In any case, he thought the leak was underground and outside. Underground meant excavation, and excavation required another operator. So ... you need to ring the company. They'll ring you back to make an appointment. That's right, within five days (or two). And the excavator? Five days. Guaranteed.
By this time, with the meter still running more than two weeks after the leak had been reported and, she thought, potentially another fortnight in the offing, the customer quickly had the leak - which turned out to be under the house - fixed by another plumber.
Let's be clear what this story is about. It's not about incompetence or ill-will. There's no reason to doubt the company's worthy intentions about water conservation and customer service on its website. What interests the customer - and me - is the extraordinary discrepancy between what the company insists is 'good service' and her personal experience that it is not.
How can a company construct a system that allows it to live with such opposites, claiming in good faith excellent customer service and conservation at the same time as it leaves the water running for a month, give or take a few days? That requires a minimum of four transactions with the company, up to six if the leak is underground (work it out), to fix a domestic leak?
The answer is, very easily. This is the norm, rather than the exception, in service companies. You do it by managing the trees and ignoring the wood, in the name of efficiency breaking up the process of serving the customer into individual activities (phone the customer, send a surveyor, send a plumber) and managing the activities according to detailed specification. The underlying purpose gets lost in the process.
The purpose of a water company is, presumably, satisfying customers by delivering water as reliably and economically as possible. The repair ought to serve that purpose, so the quicker the better. That's certainly what matters most to the customer.
You'd think, then, that the company would measure and manage the process from beginning to end: the complete repair cycle time from first phone call to the plumber's van pulling out of the drive. But you'd be wrong. Not only does the company not manage overall repair times, it doesn't even know what they are. It can tell you how well it meets the specifications for making an appointment or sending a surveyor or plumber, but not how long the overall process takes.
In other words, it has no way of measuring how well it is meeting its purpose of carrying out the repair. Result: it is perfectly possible, as in this case, for an urgent repair to take the best part of a month and everyone to meet their specification . So, again: because everyone is meeting their specification, service is impeccable - we can prove it. The customer must be mistaken. No one knows anything is wrong.
Does this sound familiar? It should. Fragmentation and specification are everywhere, and everywhere they have the same dire effects of building in inefficiencies and waste. Because of the inefficiencies - think of the wasted time, transactions, phone calls, mileage, not to mention water, and the large possibility of error in the multiple hand-offs - poor service is always more expensive than good, although managers (and politicians) can't see this either.
Specifications and targets put in place to improve service are in fact incompatible with it, because they subvert purpose. Instead of minimising customer inconvenience, water company employees strive diligently to meet the specification - phone back or get someone out within the specified time. This isn't their fault: they are doing what they are told. To a degree, this is true of the company as a whole. Incredibly, among the water regulator's eight indicators of service levels, one is the time companies take to answer the phone - but not how long it takes to fix a broken main.
As always, there is an alternative. Stop measuring activities that are part of the whole, get rid of the accompanying specifications and targets, and manage the work as a piece from beginning to end. Put people to work solving customers' problems rather than meeting specifications. Keep the water flowing - and staunch the flood of leaking customer confidence.
The Observer, 5 September 2004