Target and the Targeteers

This blog appeared on the Royal Statistical Society website Statslife on 29 May 2014

DartboardJohn Pullinger, newly appointed head of the UK Statistics Authority, has given a trenchant warning about the “unsophisticated” use of targets. As reported in The Times (London) (“Targets could be skewing the truth, statistics chief warns”, 26 May 2014 – paywall) he cautions:

Anywhere we have had targets, there is a danger that they become an end in themselves and people lose sight of what they’re trying to achieve. We have numbers everywhere but haven’t been well enough schooled on how to use them and that’s where problems occur.

He goes on.

The whole point of all these things is to change behaviour. The trick is to have a sophisticated understanding of what will happen when you put these things out.

Pullinger makes it clear that he is no opponent of targets, but that in the hands of the unskilled they can create perverse incentives, encouraging behaviour that distorts the system they sought to control and frustrating the very improvement they were implemented to achieve.

For example, two train companies are being assessed by the regulator for punctuality. A train is defined as “on-time” if it arrives within 5 minutes of schedule. The target is 95% punctuality.
TrainTargets
Evidently, simple management by target fails to reveal that Company 1 is doing better than Company 2 in offering a punctual service to its passengers. A simple statement of “95% punctuality (punctuality defined as arriving within 5 minutes of timetable)” discards much of the information in the data.

Further, when presented with a train that has slipped outside the 5 minute tolerance, a manager held solely to the target of 95% has no incentive to stop the late train from slipping even further behind. Certainly, if it puts further trains at risk of lateness, there will always be a temptation to strip it of all priority. Here, the target is not only a barrier to effective measurement and improvement, it is a threat to the proper operation of the railway. That is the point that Pullinger was seeking to make about the behaviour induced by the target.

And again, targets often provide only a “snapshot” rather than the “video” that discloses the information in the data that can be used for planning and managing an enterprise.

I am glad that Pullinger was not hesitant to remind users that proper deployment of system measurement requires an appreciation of psychology. Nobel Laureate psychologist Daniel Kahneman warns of the inherent human trait of thinking that What you see is all there is (WYSIATI). On their own, targets do little to guard against such bounded rationality.

In support of a corporate programme of improvement and integrated in a culture of rigorous data criticism, targets have manifest benefits. They communicate improvement priorities. They build confidence between interfacing processes. They provide constraints and parameters that prevent the system causing harm. Harm to others or harm to itself. What is important is that the targets do not become a shield to weak managers who wish to hide their lack of understanding of their own processes behind the defence that “all targets were met”.

However, all that requires some sophistication in approach. I think the following points provide a basis for auditing how an organisation is using targets.

Risk assessment

Targets should be risk assessed, anticipating realistic psychology and envisaging the range of behaviours the targets are likely to catalyse.

Customer focus

Anyone tasked with operating to a target should be periodically challenged with a review of the Voice of the Customer and how their own role contributes to the organisational system. The target is only an aid to the continual improvement of the alignment between the Voice of the Process and the Voice of the Customer. That is the only game in town.

Borrowed validation

Any organisation of any size will usually have independent data of sufficient borrowing strength to support mutual validation. There was a very good recent example of this in the UK where falling crime statistics, about which the public were rightly cynical and incredulous, were effectively validated by data collection from hospital emergency departments (Violent crime in England and Wales falls again, A&E data shows).

Over-adjustment

Mechanisms must be in place to deter over-adjustment, what W Edwards Deming called “tampering”, where naïve pursuit of a target adds variation and degrades performance.

Discipline

Employees must be left in no doubt that lack of care in maintaining the integrity of the organisational system and pursuing customer excellence will not be excused by mere adherence to a target, no matter how heroic.

Targets are for the guidance of the wise. To regard them as anything else is to ask them to do too much.

The elephant in the room – proving a negative in litigation

File:African Bush Elephant.jpgThe apocryphal story goes around that Ludwig Wittgenstein challenged fellow philosopher Bertrand Russell to prove that there wasn’t an elephant in the room in which they were sharing afternoon tea.

It’s a fairly self-indulgent challenge between intellectuals but it does highlight a feeling we’ve all had. It’s easy to prove that there’s an elephant there, if there is, by pointing to it. Proving that something isn’t there is more problematic. You have to point to everywhere that it isn’t.

Former Shell Legal Director Peter Rees QC recently observed that litigation and compliance are the most significant risks currently facing corporations. In litigation, defendants sometimes find themselves in the position of having to prove that something didn’t happen against an allegation from a claimant that it did. That always puts the defendant at a disadvantage. The claimant will give evidence of what they say happened. What evidence can the defendant give?

This asymmetry will be all the more keenly felt in England and Wales following the recent Jackson reforms to personal injury litigation. The former control mechanisms have been swept away and the Ministry of Justice believes that this is likely to result in more claims against businesses. Claims that would have previously been screened out will now be run because of the economics of the restructured claims environment. All my instructing solicitors are now confirming this to me.

Ironically, the instrument of this upwards pressure on claims risk is Qualified One-way Cost Shifting (QOCS). QOCS also pretty much prevents a business who successfully defends a claim from recovering legal costs against the unsuccessful claimant. In any event, legal costs are likely to be dwarfed by irrecoverable costs to the business from having key people distracted from the value-creating process.

All that means that businesses need to get better at stifling spurious claims at the outset. The twin keys to that are process discipline and record keeping.

It always saddens me when I have to advise businesses to settle doubtful claims simply because their record keeping was not capable of setting them up to rebut an allegation.

There are three principal elements to staying ahead of the game:

  • Ensuring that risk assessment identifies where record keeping would support the organisation’s narrative of prudent operation and regulatory compliance;
  • Implementing a system of process surveillance to foster process discipline; and
  • Building a document retention system that ensures that such a record can be interrogated to provide a compelling picture of conscientious management and risk mitigation.

A well designed document retention system is a key part of managing risks.

I find it instructive and encouraging that in Divya & Ors v Toyo Tire and Rubber Co. Ltd & Ors, Toyo Tire managed to persuade the court that it was very unlikely that a road traffic accident could have been caused by a manufacturing fault in their tyre.

I do not advocate rigorous process management as a net cost motivated by defensive operations aimed at providing a patina of compliance. That is not what succeeded at Toyo Tire. Rigorous process management reduces waste, improves socially recognised customer reputation and streamlines cashflow. Its potency in litigation is a bonus.

Deconstructing Deming I – Constancy of Purpose

File:W. Edwards Deming.gifMy 20 December 2013 post on W Edwards Deming attracted quite a lot of interest. The response inspired me to take a detailed look at his ideas 20 years on, starting with his 14 Points.

Deming’s 14 Points for Management are his best remembered takeaway. Deming put them forward as representative of the principles adopted by Japanese industry in its rise from 1950 to the prestigious position it held in manufacturing at the beginning of the 1980s.

Point 1. Create constancy of purpose toward improvement of product and service, with the aim to become competitive and to stay in business, and to provide jobs.

In his 1983 elaboration of the point in Out of the Crisis, Deming explained what he meant. Managing a business was not only about exploiting existing products and processes to generate a stream of profits. It was also about re-inventing those products and processes, innovating and developing to retain and capture market. Deming was fearful that management focused too much on short term profits from existing products and services, and that an effort of leadership was needed to reorient attention and resource towards design and development. The “improvement” that Deming was referring to is that through design and redesign, not simply the incremental improvement of existing value streams. Critically, Deming saw design and redesign as a key business process that should itself be the target of incremental continual improvement. Design and re-design was not an ad hoc project initiated by some rare, once in a generation sea-change in the market or motivated by a startling idea from an employee. It was a routine and “constant” part of the business of a business.

Some of Deming’s latter day followers sometimes deprecate the radical redesign of processes in approaches such as Business Process Re-engineering, and instead promote the incremental improvement of existing processes by those who work in them. That is exactly the approach that Deming was warning against in Point 1.

It is worth recalling the economic and geographic climate within which Deming put forwards this principle. During the early 1980s, the US and Western Europe suffered a significant recession, their populations beset with the dual evils of unemployment and inflation. The economic insecurities were aggravated by social unrest in the West and the intensification of the Cold War.

In 1980 Robert Hayes and William Abernathy, academics at the Harvard Business School, attacked US management in their seminal paper Managing our Way to Economic Decline. They found that fewer and fewer executives were from engineering and operations backgrounds, but increasingly from law and finance. Such managers had little understanding, they said, of the mechanics of the businesses they ran or the markets in which they competed. That in turn led executives to tend to pursue short term profits from existing value streams. These were easy to measure and predict on the visible accounts. However, managers were allegedly ill placed to make informed decisions as to the new products or services that would determine future profits. The uncertainties of such decisions were unknown and unknowable other than to a discipline specialist. Franklin Fisher characterised matters in this way (1989, “Games economists play: a noncooperative view” Rand Journal of Economics 20, 113):

Bright young theorists tend to think of every problem in game theoretic terms, including problems that are easier to think of in other forms.

This all appeared in contrast to Japanese manufacturing industry and in particular Toyota. By 1980, Japanese manufactured goods had come increasingly to dominate global markets. Japanese success was perceived as the (Lawrence Freedman, 2013, Strategy: A History, p531):

… triumph of a focussed, patient, coherent, consensual culture, a reflection of dedicated operational efficiency, or else a combination of the two.

Certainly in my own automotive industry days, my employer had come to see its most successful products literally as commodities. They belatedly realised that, while they had been treating them as a mere income stream, admittedly spent largely on unsuccessful attempts to develop radical new products, Japanese competitors had been filing dozens of patents each year making incremental improvement to design and function, and threatening the company’s core revenues.

But did Deming choose the right target and, in any event, does the exhortation remain cogent? It feels in 2014 as though we all have much more appetite for innovation, invention and product design than we had in 1983. Blogs extol virtues of and strategies for entrepreneurship. Slogans proliferate such as “Fail early, fail fast, fail often”. It is not clear from this web activity whether innovation is being backed by capital. However, the very rate of technological change in society suggests that capital is backing novelty rather than simply engaging in the rent seeking that Hayes and Abernathy feared.

In 2007 Hayes reflected on his 1980 work. He felt that his views had become mainstream and uncontroversial, and been largely adopted in corporations. However, information and globalisation had created a new set of essentials to be addressed and to become part of the general competencies of a manager (“Managing Our Way… A Retrospective by Robert H. Hayes” Harvard Business Review, July-August 2007, 138-149).

I remain unpersuaded that there has been such a broadening in the skill set of managers. The game theorists, data scientists and economists seem to remain in the ascendancy. Whatever change of mind in attitudes to design has taken place, it has happened against a background where CEOs still hop industries. There are other explanations for lack of innovation. Daniel Ellsberg’s principle of ambiguity aversion predicts that quantifiable risks that are apparent from visible accounts will tend to be preferred over ambiguous returns on future inventions, even by subject matter experts. Prevailing comparative advantages may point some corporations away from research. Further, capital flows were particularly difficult in the early 1980s recession. Liberalisation of markets and the rolling back of the state in the 1980s led to more efficient allocation of capital and coincided with a palpable increase in the volume, variety and quality of available consumer goods in the West. There is no guarantee against a failure of strategy. My automotive employer hadn’t missed the importance of new product development but they made a strategic mistake in allocating resources.

Further, psychologist Daniel Kahneman found evidence for a balancing undue optimism about future business, referring to “entrepreneurial delusions” and “competition neglect”, two aspects of What you see is all there is. (Thinking, Fast and Slow, 2011, Chapter 24).

In Notes from Toyota-Land: An American Engineer in Japan (2005), Robert Perrucci and Darius Mehri criticised Toyota’s approach to business. Ironically, Mehri contended that Toyota performed weakly in innovation and encouraged narrow professional skills. It turned out that Japanese management didn’t prevent a collapse in the economy lasting from 1991 to the present. Toyota itself went on to suffer serious reputational damage (Robert E. Cole “What Really Happened to Toyota?” MIT Sloan Management Review, Summer 2011)

So Deming and others were right to draw attention to Western under performance in product design. However, I suspect that the adoption of a more design led culture is largely due to macroeconomic forces rather than exhortations.

There is still much to learn, however, in balancing the opportunities apparent from visible accounts with the uncertainties of imagined future income streams.

I think there remains an important message, perhaps a Point 1 for the 21st Century.

There’s a problem bigger than the one you’re working on. Don’t ignore it!

Richard Dawkins champions intelligent design (for business processes)

Richard Dawkins has recently had a couple of bad customer experiences. In each he was confronted with a system that seemed to him indifferent to his customer feedback. I sympathise with him on one matter but not the other. The two incidents do, in my mind, elucidate some important features of process discipline.

In the first, Dawkins spent a frustrating spell ordering a statement from his bank over the internet. He wanted to tell the bank about his experience and offer some suggestions for improvement, but he couldn’t find any means of channelling and communicating his feedback.

Embedding a business process in software will impose a rigid discipline on its operation. However, process discipline is not the same thing as process petrification. The design assumptions of any process include, or should include, the predicted range and variety of situations that the process is anticipated to encounter. We know that the bounded rationality of the designers will blind them to some of the situations that the process will subsequently confront in real world operation. There is no shame in that but the necessary adjunct is that, while the process is operated diligently as designed, data is accumulated on its performance and, in particular, on the customer’s experience. Once an economically opportune moment arrives (I have glossed over quote a bit there) the data can be reviewed, design assumptions challenged and redesign evaluated. Following redesign the process then embarks on another period of boring operation. The “boring” bit is essential to success. Perhaps I should say “mindful” rather than “boring” though I fear that does not really work with software.

Dawkins’ bank have missed an opportunity to listen to the voice of the customer. That weakens their competitive position. Ignorance cannot promote competitiveness. Any organisation that is not continually improving every process for planning, production and service (pace W Edwards Deming) faces the inevitable fact that its competitors will ultimately make such products and services obsolete. As Dawkins himself would appreciate, survival is not compulsory.

Dawkins’ second complaint was that security guards at a UK airport would not allow him to take a small jar of honey onto his flight because of a prohibition on liquids in the passenger cabin. Dawkins felt that the security guard should have displayed “common sense” and allowed it on board contrary to the black letter of the regulations. Dawkins protests against “rule-happy officials” and “bureaucratically imposed vexation”. Dawkins displays another failure of trust in bureaucracy. He simply would not believe that other people had studied the matter and come to a settled conclusion to protect his safety. It can hardly have been for the airport’s convenience. Dawkins was more persuaded by something he had read on the internet. He fell into the trap of thinking that What you see is all there is. I fear that Dawkins betrays his affinities with the cyclist on the railway crossing.

When we give somebody a process to operate we legitimately expect them to do so diligently and with self discipline. The risk of an operator departing from, adjusting or amending a process on the basis of novel local information is that, within the scope of the resources they have for taking that decision, there is no way of reliably incorporating the totality of assumptions and data on which the process design was predicated. Even were all the data available, when Dawkins talks of “common sense” he was demanding what Daniel Kahneman called System 2 thinking. Whenever we demand System 2 thinking ex tempore we are more likely to get System 1 and it is unlikely to perform effectively. The rationality of an individual operator in that moment is almost certainly more tightly bounded than that of the process designers.

In this particular case, any susceptibility of a security guard to depart from process would be exactly the behaviour that a terrorist might seek to exploit once aware of it.

Further, departures from process will have effects on the organisational system, upstream, downstream and collateral. Those related processes themselves rely on the operator’s predictable compliance. The consequence of ill discipline can be far reaching and unanticipated.

That is not to say that the security process was beyond improvement. In an effective process-oriented organisation, operating the process would be only one part of the security guard’s job. Part of the bargain for agreeing to the boring/ mindful diligent operation of the process is that part of work time is spent improving the process. That is something done offline, with colleagues, with the input of other parts of the organisation and with recognition of all the data including the voice of the customer.

Had he exercised the “common sense” Dawkins demanded, the security guard would have risked disciplinary action by his employers for serious misconduct. To some people, threats of sanctions appear at odds with engendering trust in an organisation’s process design and decision making. However, when we tell operators that something is important then fail to sanction others who ignore the process, we undermine the basis of the bond of trust with those that accepted our word and complied. Trust in the bureaucracy and sanctions for non-compliance are complementary elements of fostering process discipline. Both are essential.