11. Part A. Eliminate numerical quotas for the workforce.
I find this probably the most confused part of Deming’s thinking. Carefully reading Out of the Crisis (at pp70-75) Deming’s attack is not on standardised work, that is advocated as central to his message, but against specifications for the volume of work: calls answered per hour, finished parts per day.
Deming recognises management’s need to predict costs and revenues but condemns quotas as destructive of achieving productivity.
Deming also deprecates such quotas as corroding workplace pride. I shall return to that in Point 12.
Deming’s criticism of work quotas goes as follows.
- Some individuals may achieve them easily and their productive capacity will then stand idle.
- Some individuals may struggle and suffer poor moral.
- Some individuals may compromise quality so as to make a quota or so as to make it sooner.
- Achievement of quotas may be frustrated by faults in “the system” which are outside the individual worker’s control.
Deming gives the following example of how he would advise financial planning in a call centre of 500 people (at pp73-74).
- Set a preliminary budget.
- Make it clear to every one of the 500 that their aim is to give satisfaction to the customer, to take pride in their work.
- Everybody will keep a record of calls made.
- Customers with special problems will be referred to the supervisor.
- At the end of each week, sample 100 individuals’ record and summarise the data.
- Repeat steps 2 to 5 for several weeks.
- Analyse the data.
- Establish a continuing study following the above steps but on a reducing basis.
- Use the data to predict costs.
Now there is much merit in forecasting costs based on actual data. Further, improving performance based on the relentless criticism of historical data is essential. However, I think Deming’s prescription naïve and idealistic. The trick is to extract the ideals and industrialise them.
Planning
The simple matter is that any new enterprise has to be established on the basis of a robust business plan. There is competition for resources: people, capital, infrastructure … and everyone has to make their case. It is impossible to do that without judgment. No matter how much historical data or even qualitative experience is to hand we cannot simply project it into the future without establishing further conditions (RearView). It is unlikely this can ever be done exactly in a new establishment.
That competition for resources then prevents us from taking an overly conservative view of what can be achieved. Setting the bar too low for call centre operators starts off from an uncompetitive position. Further, the modest answering rate in the plan has to be resourced with infrastructure. Intentions to improve the answering rate post-launch are all very well but what will happen to the personnel and materiel that we bought in to accommodate the unambitious start-up?
Sometimes work needs to be set at a rate that is recognised by a team of co-workers and other parts of the organisation. Excess production is as contrary to the philosophy of lean operations as is shortage. The idea of takt time allows production lines to be balanced, receipts and deliveries co-ordinated, stock turns to be minimised and cash flows improved. In many situations that is sufficient to answer Deming’s fears about individuals distorting production to bank an accomplished target.
Stretch
What is now proved was once but imagined.
Is it so wrong to set a target that nobody involved has seen achieved before? Deming would say that it was fine so long as there was a plan defining the means by which this could be achieved. There are many compelling stories from sports science telling how records have been broken by incremental improvement (e.g. Dave Brailsford and the GB cycling team).
But what about setting an ambitious stretch target without a plan for achieving it? That would be brave indeed. It would be based on no more than an exhortation to the call centre operators to work more furiously, more furiously than anyone had ever done before. I cannot say that would never work. In my athletics days I ran some of my best times when team mates were urging me on from the sidelines. However, as a business strategy it faces the social realities of employees’ collective ability to resist quietly that to which they do not assent. With a carefully recruited and motivated team it could work. It would certainly require a high degree of collective problem solving and improvement by the operators. But of all strategies for operational excellence it looks the most limited and the most risky. There is no obvious Plan B.
The Ringelmann effect
There is a tension between unrealistic stretch targets and a further problem that Deming ignores entirely, the Ringelmann effect. It may sadden the hearts of those who believe in the inherent fulfilling joy of work and best intentions of workers to do a good job but evidence is overwhelming that there are situations where individuals exert less effort in a group environment than they would if acting individually.
In 1913, Max Ringelmann conducted experiments that showed that individuals pulled less strenuously on a rope when pulling in a group than when pulling alone.
A realistically set and communicated takt time can assist in concentrating effort and communicating common work standards and the expectations of peers.
The poor supervisor
If Deming was so pessimistic as to believe that workers would sacrifice quality to hit targets then they would surely be more than happy to shunt enquiries off to their supervisor in order to post commendable performance. All that Deming’s proposal does is to divert the whole problem of difficult calls to the supervisor who, presumably, is either beset with his own performance problems or operates outside business measurement.
Unanticipated Knowledge 