go back

12/31/2008
Better Thinking about Demand

One of the fundamental mistakes in call center design and management is to treat all demand as units of production. Call center managers are preoccupied with ‘how many people do I need?’ and ‘how many calls do they have to take?’ Managers are regaled with data about demand. The modern ACD systems and associated software can tell managers how much work there has been to do by hours of the day and days of the week.

Managers use call volume data to plan resources, but this is to ignore why their customers call from the customers’ point of view. This simple and fundamental mistake can and does add alarmingly to costs. For example, one financial services organization I worked with planned three call centers and, because of higher than anticipated demand, got to five before the chief executive started to express concern about costs. Managers rationalized the reason for the unanticipated volume of calls, comparing their experience to the unexpected volumes of traffic on the highway: because it was there, they claimed, it was attracting more traffic. But in fact the managers knew nothing about what was causing the increased demand. Analysis of the calls showed they had caused the problem themselves, for calls included an enormous volume (46% of the total) of what I call ‘failure demand’.

The distinction between ‘failure demand’ – demand caused by a failure to do something or do something right for the customer and ‘value demand’ – what the call center exists to provide, is a distinction that few call center managers make. I find that many call center managers do try to determine the reasons for a customer call, but they do so with an ‘internal’ – ‘what we do with it’ – perspective; when I look at call coding I find the codes make no distinction between value and failure demand. Worse, call coding is often ‘compulsory’ – it is a forced part of the wrap-up. This only encourages operators to put in any code that will move them on to the next call (after all, too much time in wrap-up will mean you get paid attention to).

In some cases I have found managers who are wise to the enormous potential for working on removing failure demand. But what means do they have at their disposal? Report writing and meetings across the hierarchy – probably the best way to prevent any learning and change. Working in the right way with demand data is just one component of a better system of managing call centers – roles have to change from managing through the hierarchy to managing ‘flow’, a subject I shall return to in a later issue.

Is there an economic prize associated with the removal of failure demand? Consider the following statistics: In financial services I have found failure demand to run from 20 to 50% of demand; in police forces I have found failure demand to run as high as 70%. Imagine the cost associated with such levels of waste – for that is what this represents. Top management will often be heard to say things like ‘if only we could reduce our average call length by 30 seconds, we could improve our bottom line by…’ It is to focus on entirely the wrong thing. Of course, top management has no other view than cost data.

It is from a cost view that managers find out-sourcing an attractive idea. To hand over calls to an agency, whether in the same country or, more lately, in other countries is to out-source waste. These organizations are paying someone else to work on their ‘muda’ (a Japanese expression for waste). Moving calls to another country just makes the understanding of what is going on all the more difficult. Should we praise these leaders for operating in a global economy or embarrass them for missing the obvious?

It is not only managers of call centers that treat failure calls as part of the normal workload. So do IT consultants. In one organization I was working in, a major consulting firm was conducting an analysis of call demand. I asked how much failure demand they had found and the answer was none. The consultants didn’t think that way, they took an internal focus on the nature of demand, being concerned about what agents had to do with calls, rather than what the calls represented to the customers. In fact failure demand was actually running at a predictable 65% of call volume. What were they paying for such misleading guidance?

The better way to work with demand

The first step in working with demand from the customers’ point of view is to establish type and frequency of demand. Type – why do customers call in from their point of view? Frequency – how many of such types do we have and how predictable are they? The predictability question is crucial. If demand is predictable, you can act and you act in different ways with each type of demand.

With failure demand you act to turn off the cause(s). The impact can be seen immediately on measures of productivity and customer service.

With value demand you act to design against it. You want customers to experience ‘pulling value’ from the organization. If you design the work to do the ‘value work’ and only the value work, costs will fall as service improves. Something that most call center managers find counter-intuitive, for their whole world is based on production assumptions – equating service with cost.

Articles were written by John Seddon (Managing Director) and Vanguard Consulting Ltd.  He is an occupational psychologist, author and consultant.  John describes his work as a combination of systems thinking – how the work works, with intervention theory – how to change it.  This article has been edited by the people of Bryce Harrison Inc. (USA).  The Bryce Harrison website is www.newsystemsthinking.com.