The future of lean
In my last newsletter I referred to Dan Jones’ recent comments on the future of lean. Jim Womack (co-author with Jones) wrote about the same in his recent newsletter. I thought I’d reproduce elements of Womack’s newsletter and add some comments:
Womack wrote: “In The Machine That Changed the World in 1990, Dan Jones, Dan Roos and I made an effective case that lean thinking can be applied by any company anywhere in the world but that the full power of the system is only realized when it is applied to all elements of the enterprise.”
I regard The Machine That Changed the World as a ‘must-read’ book. That the enterprise should be understood and managed as a system was the book’s central message and at that time we (Vanguard) took up the challenge to translate that idea for service organisations. It did not occur to us to look for a set of tools, we were trying to understand how to think of service organisations as systems, and we quickly began to realise that thinking in this different way was throwing up significant challenges to management conventions (as indeed it did for Ohno).
But Womack and co took a different tack: “As this view became accepted, the focus turned to how organizations everywhere could transform themselves from mass producers into lean exemplars. Given the magnitude of the task and its many dimensions it’s understandable that lean tools came to the foreground: 5S, setup reduction, the five whys, target costing, simultaneous and concurrent engineering, value-stream maps, kanban, kaizen. Indeed, I think of the period from the early 1990s up to the present as the Tool Age of the lean movement.”
This is a rationalisation. People are finally learning that applying the tools doesn’t do the trick (change the system) but the tools are (sometimes fairly) attributed with having done some good. Crosby made the same mistake. His approach (generally labelled ‘TQM’) taught tools and techniques. I remember Crosby’s response when asked what, after having done all the tools training, one should do when the initiative fails (as most did). He reply was to do it all again. It set us thinking that we should find out why the initiative failed (an intervention question), which is what led us to the conclusion that the theory implicit in the tool box was diametrically opposed to the theory-in-use in the organisation. Hence the challenge was to articulate these two ‘theories’ or ways of thinking about the design and management of work.
As Womack admits: “The attraction of tools is that they can be employed at many points within an organization, often by staff improvement teams or external consultants. Even better, they can be applied in isolation without tackling the difficult task of changing the organization and the fundamental approach to management.”
And this, I think, is the reason tools took off: it’s the kind of intervention that appeals to the current management mind-set and, moreover, offers no challenge to it. Sold to managers as a ‘training and projects’, something to ‘get your people to do it’, the tool heads do not appreciate that this is a change the system problem. Tools won’t change a system. Tools are manifestations of problems people solved (in manufacturing) having first taken a different view of the work (as a system). Ohno insisted we shouldn’t teach tools, for it is the different view – the way you conceptualise the problems – that is paramount. For these reasons we knew we had to develop a method for helping managers take a different view of their work. In time it became the Vanguard Method.
Womack’s prescription for moving ahead with the next ‘age’ – ‘lean management’ is for organisations to have ‘value stream managers’ – people who have responsibility end-to-end services but no authority, for the organisation remains hierarchically designed. In Womack’s view, functional hierarchical management is an inevitable feature of organisational life. We saw how this was tried and failed in the TQM era. How would you expect managers of functions to behave at quarter-end if their numbers are jeopardised by the priorities as seen by the value stream manager?
We have learned it is entirely possible and incredibly powerful to manage services end-to-end, removing functional hierarchical designs, replacing them with system designs. Central to the changing the system problem is changing roles and measures.
As Womack points out: “If managers focus on their process, the performance metrics will come right; but if managers focus on their numbers, the process is likely never to improve.”
But actually the key is to remove ‘performance metrics’ that undermine performance and replace them with measures that are more useful for understanding and improving the work. Managers have to ‘un-learn’ old thinking about measurement and then learn how to work with measures that are more valid and useful. It also leads to a complete re-think of the budgeting process, which is at the heart of the ‘command-and-control’ disease. Womack and co do not seem to get this; the ‘value stream manager’ idea will be disabled by the current system. The point is to change the system.
A reader writes to tell me of his experience being taught lean ‘tools’:
“I was sent on a two day workshop to learn about lean principles – guess what – we had tools to measure process cycle efficiency, KPIs to report into the management factory, averages coming out of your ears and ….. NO VARIATION discussed at all. When I asked a question regarding the lack of focus on variation I was told that we might cover that in a future course!”
It is indicative. The tool heads don’t question management’s current measures; their belief is you can employ the tools to improve anything. They cannot see that current measures are part of the problem and prevent people seeing what needs to be seen for things to be understood and improved.
He continues: “We had to work through a painstaking simulation over a day and a half to ‘fix’ a broken process managing requests for bank loans using all of the tools and measures we had been taught – I could see all this coming so after about half an hour on day one I asked why we didn’t just move all the work to the Customer call centre, and eliminate all the nonsense failure demand we were trying to manage and resource!
I pointed out that there was no such thing as good or bad performance if we measured by average and by individual agent without knowing about variation. But this did not go down too well either – I was politely asked to stop being disruptive.”
As I guess he knows, you always learn that the major causes of variation in the workers’ performance are down to the system – the way work is designed and managed. And that is where attention should be focused. But the ‘tools teacher’ is there to teach tools, not how to think; so any distraction is unwelcome; hard questions especially so. As he says: “my real concern was that most people left the course thinking that all they had to do was ‘measure’ a process and ‘improve’ it.”
Courses like this just breed more tool heads.
One of our students on the Cardiff Lean Service MSc had been applying lean tools to his work for some time. He was dissatisfied with the results he was getting and was being told by his ‘tools experts’ to keep going. [A sure sign of madness?] His is an intricate business involving the maintenance and repair of complex equipment.
We (Richard Davis and I) suggested he drop his tools and instead study the system; we recommended he took successive repairs in order to study whether things that went wrong were predictable, if they were he should identify their causes (‘system conditions’ in Vanguard-speak). Like most managers he thought that so many things could go wrong that it would be most unlikely that anything could be predictable.
Having studied his work as a system, he wrote to us:
“Finally the mist has cleared and I have begun to see the ‘systems’ way….. I owe you both an apology, having argued vehemently that ‘my’ system was not predictable, after some research, I now realise that it is!”
So now he knows the good news, for if the problems weren’t predictable he couldn’t do much about them. And consider this: The tool heads would tell him the first step in improvement is to MAKE the work predictable, that is, standardised. How would this play out as an intervention compared to seeing what he has seen – the type and frequency of predictable problems and their systemic causes? By standardising the work as the first step in change, he would create more systemic causes of waste and, worse, it would make them harder to see.
It is also worth pointing out that he didn’t ‘get it’ in the classroom, he rationalised like we all would, for in the classroom you are trying to take on new ideas while your perspective on the work is based on your current ideas. He ‘got it’ when he did it. As did Ohno.
By the way, if this MSc is something that you might be interested in, contact gardnerca@Cardiff.ac.uk for information.
During the Christmas break I finally got around to responding to Ruth Kelly’ s White Paper on public sector reform. There is a lot to say about the White Paper so for the sake of getting something to her, I chose to write about the targets problem first. In essence the White Paper suggests that having fewer targets is the right way to go. But, as you may know, doing less of the wrong thing is not doing the right thing.
This paper can be downloaded from the web site (for free) at:
If public sector reform matters to you please get involved in the discussion.
It is of no surprise that the White Paper places greater emphasis on local authorities sharing services. Vanguard’s Shared Services events continue to sell out. We are scheduling more: January 18th, Manchester. January 25th, London. March 22nd, Leeds. If you want one in your area or want to host one please let us know.
One of the topics we discuss is the idea that the private sector is ‘better’ at management. If something can be done better, it will be a matter of method, not ownership. And if we can get massive savings at the same time as improving public services (which we know is achievable) why wouldn’t we want taxpayers to reap the benefits rather than shareholders?
The profit motive drives providers to offer services that put the provider in a position where they can focus on how to get more and more revenue from the captive authority. Many of the contracts are based on transaction costs so any waste (usually lots) associated with poor service at high costs is locked in. Ticks in the boxes for call centres, CRM systems, front- and back-office designs, partnering and shared services, but worse service at higher costs.