You are here:Forum » Why is there still so much interest in balanced scorecards?
Why is there still so much interest in balanced scorecards?
.jpg)
Hi everyone,
Just signed up but my path crossed Andy Neely's when he invited me to be one of the speakers at his First International Conference on Performance Measurement in Cambridge back in 1998 (where I spoke about the maturity of organisations in terms of how they managed their people). Since then I have been teaching many people (including my MBA class) that Kaplan and Norton admit that 70% of Balanced Scorecards don't work and that this (approximate) percentage seems to apply to a lot of management fads - Champy admitted 75% of Business Process Reengineering projects failed and only yesterday I had a flyer in my mail from Cranfield School of Management pointing out that "68% of projects fail" and offering "a new and exciting programme".
I am not an academic myself but this percentage seems to be so constant that it makes me suspicious that there must be an underlying trend here. So it leads me to ask - if this statistic is so constant - is the 'technique' at fault or is it the organisation that fails to get the most out of it? In relation to the Balanced Scorecard my answer is that the model itself has always been seriously flawed. In particular, why only 4 perspectives (what about system and culture as separate perspectives, for example?). Norton himself admits that, in his experience, organisations are particularly bad at the people measures (learning and growth) but does not then admit that a 'three quarters' scorecard cannot be a balanced scorecard at all. Moreover, as someone who is still predominantly interested in innovations in thinking in individual and organisational performance, the fundamental scorecard concept of 'balance', in practice, just results in more mechanistic measurement that is just as disconnected as before but looks on paper like a whole system. Shouldn't the fundamental concept be one of 'holistic systems' where any performance measures used are checking that the whole system is working effectively and integral to such a concept is the notion of effective human systems (e.g. how does an organisation measure whether it has a human, learning system in place?) I look forward to an interesting debate.
best regards
Paul Kearns
Let's just imagine for a moment that there really is a fairly consistent failure rate across management improvement projects of different kinds - or indeed projects of all kinds.
Why might that be? Perhaps it has nothing much to do with the value of the management methods and everything to do with a consistent level of optimism and overconfidence demonstrated by people when they decide to start such a project.
Even if a method can be made to work in every situation people will still underestimate the resources and time needed, leading to failure. On top of that I suspect that many management methods actually do not work well in some situations, so optimistic thoughts that methods will work even outside their best applications might also account for some failures.
On this model, no matter how good the management method, 70% of the projects started to apply it should not have been started at all, at least not with the meagre resources at their disposal.
Having said that, the failure rate of projects is not consistent. Large projects fail more often than small ones, and very large projects fail even more often than that.
Well, this thread has grown and extended even further, I see! I think Paul (K, of course) deserves great credit for opening up this debate in the first place. Some of the comments - like Phil's - are fascinating, in part because they come from a quite specific background and draw on a particular body of experience. I think, Phil, that you mentioned time spent in the nuclear power industry, as well as exposure to OR? Well, I think that's highly relevant. It is precisely in industries like this that quite enormous expenditures have gone inot rendering potentially unstable (runaway, China Syndrome) core processes in a stable form; harnessing the atom, as we used to say! I too had exposure to nuclear fission a long time ago, working in the (long-dead) Department of Energy in England and the experience is a powerful one. Work clearly has a profound effect (maybe in the longer term) on how we think about ourselves and our world. I simply ask: would someone with a lifetime of experience in human services feel the same way about what seem to be smiilar things? The things that go to make up an effective performance in one situation may look the same (and be treated with the same words) in another context, but I suspect they only rarely are, on close examination. The peril that this presents, as far as I'm concerned, is that using a template (that is frequently literally interpreted) in differing contexts may actually inflict damage to the people and things within that situation. This possibility has been repeatedly raised by scholars in this field, relating to behaviors, for sure, but also to a wider class of practical actions as well. In OR, Checkland's interpretation of Soft Systems goes a long way to trying to capture this inherent multiplicity in life. It seems, though, that this line of reasoning on what we my be doing to the world through premature generalisations has to be made not once, but many times, in a debate that will continue, I suspect, for a very long time.
Kind regards to you all
Paul D
Paul Davis
University of Worcester Business School
I have to agree with Paul Kearns and Francis A-K on this one.
As a programme director who has run commercial procurement and more lately Turnaround programmes (in banking and the health sectors) it has been absoultely essential that a system of measuring performance was developed and implemented.
The simpler the system, the better (a simple system is less open to having its results interpreted incorrectly, than a complex system).
That people have tried to develop a "holistic and balanced philosophy" (thanks Paul) around a BSc, was trying to make the BSc into something it wasnt intended to be.
John Halsted
Paul
You say that "N&K admit that 70% of balanced scoeracrds do not work". Do you have a reference for this statement please? I would like to see the original context in which either David or Bob said this. Thanks
Phil
... it was in a conference pack of Norton's own slide presentation a few years ago at a Balanced scorecard Collaborative event but I'm afraid I binned it. Regardless of attribution and context though I would argue that Norton himself knows that 100% of balanced scorecards must fail if any part of the scorecard fails - here's a quote by him from his Foreword in "The HR Scorecard" (Ulrich et al 2001) "..the worst grades (for executives) are reserved for their understanding of strategies for developing human capital. There is little consensus, little creativity, and no real framework for thinking about the subject. Worse yet, we have seen little improvement in this over the past eight years". He adds "Several facets of the book will make lasting contributions. First, the development of causal models, which show the relationship of HR value drivers with business outcomes, takes the Balanced Scorecard to the next level of sophistication". Unfortunately he backed the wrong horse - Ulrich's book does not produce a causal model and Ulrich's own research over many years is not causal (for attribution I would refer you to a phone conference I personally had with Ulrich himself who eventually, after much pressing, agreed that correlations never prove causation). Furthermore the advent of Kaplan and Norton's 'strategy alignment maps' were akin to over-engineering a structure built on very shaky foundations and an admisssion that the BSC is, at best, a very flawed model. Not surprisingly, I see no likelihood of things improving in the next eight years because until we educate executives about the real meaning and management of human capital the concept of the BSC cannot pertain. Hence my original question - why is there still so much interest in such a poor model?
Of course I shouldn't criticise unless there is a better alternative - so what better alternative is there than to ask every employee whether and how they are adding value? To which the only answers are - I am increasing output, reducing cost, improving quality or raising revenue? I think in terms of simplicity this beats the BSC hands down.
Paul Kearns
Director
PWL
Paul, there is an expression I fall back on in these circumstances. "Those that say it cannot be done should not get in the way of those that are doing it". Sorry, but that is how i feel about what you have put.
Clearly we have quite different experiences of balanced scorecards (I worked for N&K for over 4 years) and have been designing them since 1996.
A few points to explain my perspective:
1) Yes 70% of balanced scorecard initiatives probably do fail, but as McKinsey and Hackett cite, 70% of so called balanced scorecards fail the most basic test of being balanced. Qed they fail as a tool of startegy - they are just bicycles claiming to be cars.
2) A brief reading of any of the books would have shown that the learning and growth perspective includes the potential to put objectives about culture, technology, or anything else concerned with developing and growing the capability of the organisation. No other perspectives are needed. Systems and culture already have a space awaiting them.
3) If you do add extra perspectives then any semblence of learning and growth driving the performance of what you do to deliver better for your customers and therefore improving finances will be destroyed.
4) Yes organisations are poor at measures in the learning and growth perspective for three reasons (at least)
a) they treat it as the people perspective and then measures staff turnover, sickness, etc etc. Not pieces that drive behaviour. Underlying problem: They do not consider the right question.
b) They start by thinking what can they measure, not what do we need as an organisation to imperove our capability and competency in this market. Underlying problem: They start with the wrong problem in mind.
C) They are not thinking about hat they want to achieve before they start to ask, how to measure it. Underlying problem: They have a poor measure design methodology.
All these are fixable - and my clients do all the time.
5) You are right correlation does not imply causality. (I spent 3 years in operational research in the nuclear industry doing statistical modelling and analysis). But causality will imply a correlation. Correlation might merely be simply that - a correlation. And yes its a systems thing - chicken and egg - do high performing organisations create a positive culture or vice versa? But the point is this: If you ask a series of managers what they believe they need to address within their culture, technology, skills etc. What they have to learn and grow and develop as an organisation to deliver specific parts of their strategy, they will tell you. They will tell you what they believe and what they want to change and why they believe thatwill make a difference.
This is where the strategy map comes in. It asks, what are the few things we need to focus on to deliver our strategy.
6)The balanced scorecard is not a flawed concept that uses a startegy map to patch it up. I recall seeing early startegy maps, then called performance driver models (PDMs), on the servers in Renaissance from projects dated 1994. The thinking of causality that strategy maps use is deeply embedded in proper Balanced scorecard thinking. (Just read the preface to teh very first book and you will find references to it). The problem is that many people still use the discarded 1992 cruciform model which has no causality. They then search for measures for these perspectives and any sense of startegy, performance and driving changed has disappeared.
Just because some practitioners do not understand how to make them work, and many organisations use them simplistically, does not mean all are flawed. It merely means that teh principles are not understood properly, nor followed properly.
Might I say I am concerned that if you are passing this sort of message to a community of MBA students then you are pedaling a worrying message. One that does not represent good practice but feeds off poor practice. Cars crash therefore we shoudl dump all cars and travel by train. That is a flawed correlation and causality model. Like I say, I have clients who use strategy maps to drive their balanced scorecards, they work really well.
To be frank i will tell what the biggest problem is. The tough questions about causality between objectives in perspectives get avoided. Either by practitioners who are afraind to ask them or managers who do not wish to answer them. When this happens the resulting scorecard degenerates into a set of whatever you have operational measures - and they fail.
I have had nmy say., As i say I believe we have completely different experiences of this, though I too have seen many organisations who get it wrong, apply it simply and fail to get the benefits they should really get. But I blame poor quality of thinking and conversation based upon a mis-understanding of the principles.
Sorry - but we stand in quite different camps. Phil
Hi Phil, quite happy to accept we are in different camps. The K&N approach, regardless of whether it is coherent or not, strikes me as an unnecessarily, over-complicated way of doing things and your own admission of a 70% failure rate seems to bear that out (regardless of who or what you might try to attribute that to). Of course, organisations can be helped to improve their performance and no doubt you have helped some to do so but you didn't need a BSC to do it - it was an unnecessary invention and does not get an organisation from A to B as quickly and efficiently as they could. Toyota did very well out of using TQM for example without a scorecard in sight and will always use the simplest methods available. The brief alternative I suggested is a great deal simpler and can be explained to anyone in just a few words - no need for maps or any of the other paraphernalia. There is an expression I always fall back on in these circumstances - I could cut my lawn with a pair of nail scissors, but I wouldn't recommend it.
Paul Kearns
Director
PWL
Paul, you are choosing to mis-interpret me. I said that 70% of things that the research says that get called balanced scorecards are failing. The research says that these are not balanced scoreacrds (even though they get called such). They are merely scorecards. Nor balanced. So to say that I said 70% of balanced scorecards fail is to mis-interpret what was being said.
An important thought: If 70% are failing, 30% are succeeding. My approach is to ask "What are the 30% doing that the 70% should learn about" That is the approach I take. That is where I have spent my time.
Render unto Caesar... You can do a lot of operational performance management without a proper balanced scorecard. Many organisations do.
You are right, asking people to step up to the plate and change is a good idea. However you might ask why this has not changed the public sector much in the last 20 years?
You can implement strategy without one as well. If you want. You just have to put other compensating management mechanisms in place. On the other hand if you properly design a strategic balanced scorecard aimed at strategy implementation..... you will not cut your toes off.
I am sorry Paul, but it is clear that you have either being poisoned by a very bad experience or have something fundamental against this approach. I am never going to convince you otherwise. If you want to ignore the experiences of organisations who have found it a valuable tool, that is fine by me. But I do think it is poor show to tell people it is a load of nonsense whilst at the same time seeming to mis-interpret the approach.
Phil
Paul Kearns
Director
PWL
Hi Paul (Kearns)! Afraid I only partially agree with you on this! I do agree that deconstructing strategy is a task that is on a hiding to nothing. I also feel, though, that business strategies generally do not readily withstand the tests of reality, especially where a ‘context’ is quite volatile. Research in this field is, I think, problematical by design. Where circumstances (or situations) really test a strategy and the often precarious alliances that underpin it, it seems to have a habit of unravelling fairly rapidly! Things like the Balanced Scorecard go rapidly down the plughole of pragmatism, I suspect.
An holistic approach is a good idea in principle, though defining the ‘holon’ is itself very challenging; it seems to move hither and yon, again with a high degree of contingency. The concern that many scholars put to me on this is what things go to make up this whole? People like Latour have pointedly observed that we have to include machines and other non-human things in the mix, if we are to make sense of current trends. What flows from this fairly comfortable idea is a job-lot of actor-network theorising, which I find quite persuasive, to be honest. One of the things that object-oriented theorists teach me is to beware of taking purely human metaphors as the basis of thinking about organising. Always provisional, often treacherously detailed and annoyingly localised, these materialist views provide a new (ahumanistic) challenge to artefacts like the BSC. I’m still thinking about what they might mean to approaches that talk about optimising around a singular idea of ‘a performance’. I’ll get back to you in about 20 years, when I have some initial conclusions!
Kind regards
Paul Davis
University of Worcester Business School
This is an interesting thread. The BSc was fleetingly fashionable, but then again, so have so many other such 'models' over the years. I think of the EFQM/'Excellence' model, which was truly de rigeur not so long ago. There are Prisms and other images of performance(s). Then there was MBO., with an even longer pedigree... and so on. In each case, there was a ballooning in research and practitioner interest, then relative quiet, based on generalising disappointment, followed by a period of sustained amnesia. In fact, it's remarkable how completely each successive claim has been erased from memory and artifacts redeployed or erased. Surely, how these things are lost is a very interesting question in itself, but one with many disciplinary answers.
There are numerous readings of exactly why this recurring process of failure has occured. To consider only two: causality; and ontology. On causality, there is always an inferred, usually linear, causal mechanism underpinning these models. Linear causality is so rare in all socially significant areas that experts effectively ridicule it. The evaluation community has started to take stock of a steady growth in causal mechanisms and these rapidly become quite complex. This development says to me that a performance model that either does not directly address causality (usually in the name of heuristic or intuitive accessibility) or more usually, assumes an indefensibly linear interconnection among things really will not hold the line against empirical inquiry or the rigours of practice. Ontology is a second big issue, to which an earlier respondent in this thread obliquely refers, under the currently highly fashionable label of a 'system'. Yes, it looks to me to be another case of fashion: people like Seddon in England are trying to impose their own idiosyncratic definition of a system. If one scratches the surface and looks, for example at the established body of work surrounding self-production and autopoietic systems, the terms of a system change markedly. so too do the associated causal mechanisms that, in one reading, define these systems. Is the evaluand then the organisation, the institution, the institution in its 'environment', the institution organising its environment, groups of institutions in a sector or industry, or indeed, entire poplations? Without specifying the ontological level, I would argue that all models are far too underspecified to be convincing.
Thanks for rebuilding the site - a real and needed contribution!
Kind regards
Paul Davis
University of Worcester Business School
Hi Paul
A very interesting set of exchanges. Don't propose to muddy the waters further on the original topic, but noted a couple of points in this posting that might be worth separate consideration.
Hi Francis,
I wonder - are the proponents of BSC's really just proponents of simple performance measurement rather than a truly holistic, balanced philosophy? Just as proponents of Six Sigma are not really proponents of the total philosophy of TQM?
Paul Kearns
Director
PWL
The balanced scorecard intorduced a new focus to performance measurement and management, that is the inclusion of non-financial performance metrics in the overal performance measurement framework, by so doing it has achieved for itself a firm place in this area.Over ther years the authors of the balanced scorecards have also strived to expand the focus of the tool to include strategic performance measurement among others. So I think the reason for the wiespread use of the balanced scorecard will be its status perhaps as the first comprehensice technique in this area, its continued relevance due to attempts to expand the initial perspective of the tool but also its adoption is partly attibutable to issues of fashion, fad and legitimacy due to the huge public image it has and continues to be given by its proponents.
Francis Ashong-Katai