Why benefit management fails in the NHS

Posted on January 21, 2010 by Phil Jones

The NHS IT delivery programme has a dreadful track record. It is not surprising when you look at their methodology and the techniques the NHS and the ISIP programme recommend for benefit mapping, benefit management and benefit realisation. They are fundamentally flawed.

Here are some of the many reasons why they fail:

  1. They ignore the landscape and context, confusing benefits at different levels

  2. They ignore impact and consequence

  3. Being uncluear where change needs to happen

  4. They treat enablers as the same as impact

  5. Projects and enablers are confused

  6. It makes following the timing and tracking deliverables difficult

First lets look at the NHS benefit management material. You can either look at the approach in the NHS ISIP benefits mapping or look at the alternative approach in the ISIP Benefit dependency network masterclass material.

1) Understanding the landscape and context

The first thing you will notice is that although the talk about benefits to particular parties, they are not specific about those parties and how they relate.

Imagine you are in a PCT. You are delivering technology to GPs so they can access a more complete view of patient records for a particular care pathway. It is important to realise that there are dependencies across that pathway that will involve other healthcare professionals such as tests in pathology labs, x-rays, hospital discharge records, and perhaps even records in Social services and district nurses. It is only when you have all of these in place that you will start to give benefits to patients. Until then you need to be clear who is involved and how they interact. Without this complete view of the landscape, the next step is impossible.

What you have is a sequence: PCT, GP, patient. What you need is to be clear about the benefits at each level.

2) Distinguishing between impact and benefit.

The second mistake that the NHS benefit approach makes is not to distinguish the difference between impact and benefit at each level. So for a GP, you can implement the system but other pieces of work will be necessary to ensure that the benefits for the GP and at the GP level, are implemented. So you actually need two levels of “benefit” for each body: “How am I affected/What is changed for me” and (separately) “What is the benefit for me”.

Notice there is ALWAYS a timing difference here. You implement a system and the benefits of that implementation at the point of implementation come later (if at all).

It is the same at the patient level. There are always two “benefits”. The first is the impact, “I can now make my own booking” and the consequence “Which means I can choose when I want and appointment”. These are often assumed to be the same but they are not. The patient may be able to make their own booking but something in the system may stop them getting the appointments they want (or even getting through). If this is the case the “benefit” of being able to make your own booking becomes more of a nuisance than not being able to.

It is “impact” and “consequence” at each level, for each party, that needs to be considered. The models within the ISIP and NHS benefit maps do not make this clear. Again you have to have the NHS landscape clear here as well, to do this.

3) Enablers vs impact.

This is a really important distinction. Again confused by the way the material is presented and the lack of context in any examples.

If I am in a PCT and am introducing a Long Term Conditions system across multiple healthcare providers then I am putting in place some enablers of this technology, perhaps software, training, new hardware or infrastructure, interfaces between systems, protocols, new working practices, that are used across multiple healthcare professional groups (Acute, GPs, Social services, community nurses, polyclinic, etc.) .

These are enablers that work across the various parties involved, BUT ARE NOT directly impacting those involved. For instance the link between the GPs systems and the central Long Term Care system is an enabler. The impact is that GPs can then see the more complete pathway and other Health care professionals can also see what the GP has put on their referral forms more easily (provided you have also rolled it out to all the GPs, given them access and trained them).

It is an enabler. Think underlying capability or common infrastructure.

4) Being clear where change work needs to occur

Now everywhere there is an arrow between these enablers, impacts and benefits and then onto further impacts and benefits, you need action and change.

So why does the ISIP benefit dependency network go:

Enabler -> Projects (and Actions to change) -> Outcome -> Benefit -> Priority objective?

It does not make sense. Projects deliver enablers. Actions turn outcomes into benefits. All operate at different levels.

What we have been drawing is clearly a set of dependencies where actions for change are needed at many levels across the NHS landscape and environment. In fact everywhere an arrow exists, some change intervention is required. If you don’t think about this you are at best leaving things to chance and at worst destroying any chance of the project actually delivering the benefits at any level.

5) Projects and enablers are confused

Lets be clear here. Projects do need enablers (give me some funding, resources, technology and an enthusiastic clinician to make things happen). But the enablers are different at different levels. This whole model seems confused about this.

However it is useful to talk about enablers. It is the Infrastructure or common enablers that are delivered by projects. The project will also deliver impact on the health care professionals or care pathways and help then enable their own benefits.

Another example of strange labels is Priority objectives. The way they draw the sequence of boxes there is “priority objective” at the end. In other words the sequence should ultimately lead to this priority objective. I agree, but that is not how to represent things.

What you will have when you develop these benefit maps across the landscape is perhaps several for the Long Term Care priority objective. One set of projects for Diabetes, another for heart conditions, another for COPD etc etc. What we are doing here is explaining how a priority is delivered by the chains of enablers, impact, immediate benefit across the various levels . By the time you reach the benefits for the patients you may have 4-6 ultimate significant benefits (that result from the impact of the project on their care). Now you don’t want to have to draw a link from each of these to a box called “priority objective”.

The whole benefit map is a description of how the priority objective will be delivered. So it should be a label on the benefit map, rather than a linked ultimate outcome. Think of the wrapper around the benefit map, or simply its title.

6) The implication for phased delivery and timing

So this is a more structured approach that is sensitive to how the impact and benefits work at any level of the health landscape. This is where the real advantage comes in. With this richer picture you can track the stated benefits at any level across the landscape you have drawn.

First have you got the enablers in place? Then have you put in place that which will immediately impact the care professionals? What are you doing about the benefits you promised for them? Have they occurred or are they not using the system because it is a pain, slow or “just another system to log out of and into”. And when will it be delivered?

Having this clear cause and effect model in your benefit map enables you to clearly show what has been delivered and what is holding up progress. Having a muddle that mixed outcomes and benefits at each level makes it impossible to draw, impossible to use and impossible to manage.

I have been using such benefit maps in the NHS for over 5 years now. Whenever i present them I first get them to check the landscape before showing them the benefit model laid over the top. This way they agree the landscape before they start to think about the benefits, consequences and impact of the project. It works.

Shame they did not do this before they wasted $6bn on the programme for change and subsequent NHS IT programmes.

By the way, you will have realised that this approach is basically strategy mapping. Therefore it is really easy to make the step to “How would we measure this” and developing a modern NHS balanced scorecard for the operation of the care pathway or service. Really easy, but to find out how to do this you’ll have to talk to me.

Phil Jones

Posted in ISIP programme, NHS Service improvement, NHS balanced scorecard, NHS benefit management, NHS benefit maps, NHS benefit realisation, Strategy Map, nhs targets | Leave a comment

End or Means?

Posted on January 5, 2010 by Phil Jones

At this time of goal setting and new year resolutions we often face a problem. We set a goal or objective, (stop smoking, make more money, lose weight, change job, etc etc) and within a few days or weeks, the goal has been dropped. Good intentions melting away like snow when the sun rises.

Good intentions have been sacrificed on the altar of expediency and reversion to existing practices and patterns of behaviour (I’ll just nip to the fridge, have another cigarette, etc).

The problem is often that we have put our attention on the end rather than the means. It is far more important to concentrate on the change that is required in behaviour rather than the goal. Its about how we behave on a day to day basis (walking instead of using the car, drinking some water instead of heading for the fridge, eating consciously rather than whilst reading a paper).

This is true for organisational goals as much as individual ones. Think of the organisation’s strategy as a persistent pattern of behaviour. It is this persistency of behaviour that is creating results. Therefore to change strategy you don’t simply change goals – you have to address the behaviour and ensure the new behaviours persistently replace those of the old strategy.

This is one reason why the learning and growth perspective is so important in the balanced scorecard. It is about identifying and then signalling to the organisations that behaviours need to change. Then actively ensuring that the new behaviour is encouraged and the old ones are dropped. I use the expression “permission” at this point. People, individually and in their social groups, have to be given permission to drop existing behaviours – ones that might have served them well for a long time, and be allowed to use, and learn how it feels to use, new ones.

Take flexible working and working from home, for example. The idea has been around for ages. the technology is readily available and has been available for ages. It is not a constraint. No, the constraint is often the reluctance to let go the command and control associated with “trusting” people to work. It is no longer about time served, but tasks achieved. The staff of the Director magazine tried home/flexible working and were surprised how hard it was to let go of the existing practices. In the same edition they interviewed Steve Shirley (Dame Stephanie) who set up FI in the early sixties so that women with children could still have a career in computer programming and yet work from home. As she put it, the issue is as about trust and control. People have to learn new behaviours and norms. The objective is not to have people working from home, but to learn new ways to help people be productive by trustiung them and giving them the chance to use their time to in the most appropriate way.

Sure, think about the end, but concentrate on the means. And signal that new behaviours are acceptable, appropriate and now the expected norm.

Phil Jones

Posted in Balanced Scorecard, Goal setting, Objective setting, Persistent pattern of behaviour, ends and means, flexible working, new behaviours, permission for change | Leave a comment

Judgement and Evidence: Audience Q&A

Posted on October 19, 2009 by Phil Jones

Using Judgement and Evidence in Modern Balanced Scorecards: the use of subjective and qualitative measures.
Audience questions and responses

I presented a BetterManagement webinar on this subject, on 14th October, and it received a lot of interest. You can listen again to the webcast on the better management site (See link at the end)

Here are all the questions that those listening asked at the time, but not all were answered in the time available. I have published them here so you too can get the insight these questions raise.


Atena asked:

Q: If staff are not trusted on qualitative measures, wouldn’t this be the same for quantitative measures?

A: The premise in your question is not that the data is wrong, but that people are giving you false information. There are documented examples in UK hospitals where staff simply reported better information to central government than was actually the case to hide problems and avoid punishments. Of course you will need to sort out this trust questions, before you address judgement. But I would ask you some questions:
a) Is it the data you don’t trust, or the people? And why?
b) Why don’t you trust them? Do you trust them for other things? How do you know you can’t trust them on this? And why this?
c) Is it possible that they are using their judgement? The information you are asking for may not reflect the real situation: perhaps they are using their judgement. I am trying to expand your thinking here – give it some thought.
d) What is it about the way these people have been managed or are being managed that is encouraging them not to tell you the truth or the whole picture?

Fadi asked:

Q: I do a lot of business intelligent software, and the success of these application implementation depends a lot on making the collection of data as automatic and with minimal effort as possible, and that is very hard when collecting judgemental data, how can I deal with that?

A: Absolutely right. Business intelligence is about collecting large volumes of data and analysing it. It provides the evidence. Judgement is about how you are interpreting the data and what you are doing about it. I am sure you know that the balanced scorecard is about decisions and performance management – not just measurement. Let me explain with an example.

First do not be afraid of volume: in the case study each service may have been evaluated on up to 100 items once a year. So on that basis there were 3,000 (100×30) potential items of data, for each of the 14 regions. So first, remember you do not need a judgement for every single piece of evidence. We have an overall assessment and judgement score for all the statutory reports for a region across their 30 services. The assessment and judgement came from the regional director looking across all these and asking, “How well have we done and what do I need to pay attention to across all my services?” A manager of an individual service could do the same for the 100 items within their control. Judgement operates at a higher level and so there is less to capture.

Secondly we are looking at change, over time. That statutory analysis was done once a year, but the Regional director was working through the year to address any issues. So you are looking also at how well the actions that RD has put into place are working and their judgement of whether these are fixing things, before the next assessment. So the judgement could be about, “Is our response to the evidence, effective”

Just on example, but all this is easy to record manually, is relatively low volumes of written assessments and should be being assessed and discussed each month anyway.

Valerie asked:

Q: Is data available to mine (Data Mining)?

A: Typically the data is not available to mine because it is not in computer systems, it is in people minds. However their minds are available to mine. How do you do this? You can use the sort of techniques I explained (eg score out of 10 followed by questions) to understand what people are basing their judgement on.

If you collected a large number of people’s assessments, over a period of time, you would have some information on which you might be able to mine using textual analysis. But I would not recommend this above simply talking to people.

Se also my answer to the BI question from Fadi above.

Barbara asked:

Q: How do you address subjective measures in a culture that primarily values what you can measure quantitatively (financial, output, etc.)?

A: There is nothing wrong with valuing what you can measure quantitively. Remember it is judgement AND evidence. Two thoughts though:

First, sometimes, a strength overdone, can become a weakness. I did work for an organisation with a background from pharmaceuticals that was extremely rigorous in its analysis of information and markets. However they sometimes over-analysed things and missed opportunities because it was over-lengthy and over-rigorous. They didn’t trust their judgement to go with what they had, test it and learn. They wanted it “Right”.

Secondly I would say, go with what you have and exploit it. You say ”they primarily value…” but I suspect they “they also value…” judgement, and are unlikely to dismiss it altogether. Your staff will be making judgements based on the evidence all the time, and they will naturally be thinking the about capabilities, culture, skills, knowledge and behaviours of people that drive performance and outcomes. It may be that management are focusing attention of outcomes, quite rightly and for good reason. The reporting systems are not looking (explicitly) at the culture, skills and behaviours (Leading indicoto4rs or drivers of change). Evidence dominates judgement in discussions and decisions. So, go with what you have and, build on it. Perhaps start by capturing commentaries and assessments of the information they have looked at. This way you start to expose judgement, based on evidence. Remember someone is making a judgement about what is good evidence. It is always there.

Valerie asked:

Q: How long has the process of Judgement analysis been available?

A: Well people have always been using judgement and evidence. So it is not new at all.

It is not new to explicitly look for judgement statements and elicit them to create a conversation, but it is less common in many organisations. Remember in this case study we were pushing against an open door – they trusted their staff to use their judgement, and like any organisation recruited, developed and rewarded good judgement.

It is certainly far less common to find any use of it in the performance management, management information or software solution. In fact let us go back a step. Since 1994 balanced scorecards have been designed with objectives, before measures. Yet so many systems are incapable of expressing an objective and its measures. Yet the approach for measure design has always been, objective first, then characteristics, then, what to measure and how to measure it. But the systems concentrate on the answer – and record just the measures.
So I believe the thinking and approach has always been there. The software providers and designers of performance measurement and management approaches, in general, tend to ignore or dismiss it. Good managers naturally encourage it and develop it.

Paul asked:
Q: Doesn’t the use of subjective measures require the alignment of judgments?

Do you mean before or afterwards? It is absolutely not a pre-requisite. In fact the advantage of the approach is that it exposes differences and helps people share and align their judgement. Think of it as encouraging a conversation where people can express their judgement, learn from other’s views, discover why they think like they do or believe what they do, in a safe environment.

The Regional Directors here (and managers in every workshop on strategy and balanced scorecard development I have ever run, all say that the exposure of the deeper thinking of people, the sharing of ideas and the discussion was the most important part, because we came out aligned and understanding where and why differences occurred.

I have done this in an organisation where the people we brought together were from quite different departments and had never met, yet all tried to support the same outcomes. Obviously we were staring further back there but the conversation still helped understanding and eventual collaboration and progress.

If there was alignment of judgement everywhere you would have an organisation of clones. Just think about the credit crisis – Hardly anyone one questioned the situation with excessive exposure to hedge funds and potentially defaulting mortgages. You had a form of social acceptance of the norm, without question.

You want to encourage debate, and this approach does that.

Emre Asked

Q: Even if there is conversation between people, they may not be in the same opinion about the subjective judgements. In this case the disagreement may result in reactive behaviour of the people who was evaluated negatively. What is the correct approach to ensure smoothness of operations after the evaluation?

This question follows on nicely because it raises the topic of the culture that surrounds the conversation, the context, and what people might believe about the motives of people and consequences.

Isn’t it a shame that people feel they cannot speak openly because they will be criticised! In the presentation I talked about the difference between a context where failure was punished, (which will cause people to not say anything) vs a context where a failure to cooperate and learn, was punished (which is designed to encourage conversation and learning).

I go back to the two thoughts of “Explicitly gibing people permission” and “persistently signalling a difference in the culture”. Remember what you see is a learnt behaviour, and therefore it can be un-learnt and re-learnt. But that requires work, explicit change, repetition, practice and time.

Jean-Pierre asked:

Q: What relationship do you see with installing “culture of performance”?

I see what we are discussing as one of the ways to encourage the development of a culture of performance. But we should be clear what we mean by that phrase. I have a longer definition of what I mean by a culture of performance.

A visible and explicit pattern of behaviour, actions and values…
Working to achieve the organisation’s overall objectives…
That encourages honest evaluation, feedback and appraisal…
And informs decision making…
Built upon collective and individual responsibility…
Responsive to changing circumstances…
That encourages self regulation, trust and learning…
Given this, you can see where I am coming from. This is entirely about creating a culture of performance.
If you missed the webcast you can listen again here via the Bettermanagement website http://www.bettermanagement.com/seminars/seminar.aspx?l=15139
Comments welcome
Phil
Posted in Fourth generation balanced scorecards, Modern balanced scorecard, judgement and evidence, learning organisation, qualitative measures, subjective measures | Leave a comment

The jargon of CSF, KPI, and Measures

Posted on October 15, 2009 by Phil Jones

Clean up your jargon.

Are you, like me, confused by the way people use KPI, PI, Measure, CSF etc apparently interchangeably. I am.

People often ask, what is the difference between KPI, CSF , measures, performance indicators? Frankly people use them interchangeably and loosely. I recently saw a definition of a KPI on a forum that made me angry and so posted this reply: A posting for all those literalists out there!!!!!

“KEY PERFORMANCE INDICATORS: Key Performance Indicators (KPIs) represent a set of measures focusing on aspects of organizational performance that are the most critical for the success of an organization.” It was actually quoted from David Parmenter’s book on KPIs’ which is actually quite good. But this definition is terrible.

You see, here is a classic problem with the definition cited: “Key Performance Indicators (KPIs) represent a set of measures…” ie it defines “an indicator” as “a measure”. In cact a set of measures. Ummm?

For goodness sake – what is the difference? Why not call them “key performance measures” then? And frankly what is the difference between KPIs and good old “Critical Success Factors” which is what he later goes on to define them as, as well!!!! This seems like muddled thinking to me and the source of all these problems.

Try this. Go back to basic plain English, Please!!!!!

1) If I “measure” something I have a tool for measuring and units. Think of a ruler and a table. Its 3 foot by 6 foot using a ruler. I have measured it.

2) In contrast in common parlance, we use “Indicator” in two ways:

a. If I have an “Indicator” then that is “indicating” something. Its indicative – (ie it points towards something, but is not necessarily accurate). So an “indicator” of the size of the table would be that I could fit a small pool table on it or it is large enough to seat six people. It is indicative of its size, but not a measure.
b. Likewise we have “Indicators” on cars to show where we are about to go, rather than where we are. So an indicator can be used to mean “it indicates direction”, or intention.
c. In other words, indicator, is not the direct measure but an approximation, surrogate or precursor to the real measure. And in all these cases is clearly distinct from a direct measure. I can tell whether I have a measure – or an indicator. I can test the difference.

3) Factor: An influence. One of a number of influences on something that makes up the whole. A variable, from a set of variables, under examination in a study. As in “which factors will improve sales and which of those factors are most critical to success? – Ah ah, they must be the most critical success factors”. Neither an indicator nor a measure. Clearly something different. Which oddly enough is why, in English, we have three different words for these three ideas. Surprise, surprise.

Unless we can get back to this simple form and stop cluttering up the conversation with jargon that is self referral, (measure = indicator) then we will all be in a mess.

As I always say in these situations, “Jargon should be between consenting adults in private”. Clearly we are neither all consenting, nor in private in this forum.

I was once asked to talk on “Strategic KPIs” Yes “Strategic key performance indicators” Those KPIs that are strategic (as opposed to tactical), from amongst those Performance indicators that were key (to something unspecified), as opposed to those that were not key (to something again unspecified). See what I mean. And we haven’t even got to performance, indicator and measure yet. I spent the first 10 minutes pointing out the stupidity of the title of the talk, which went down well.

Plain English please – it saves a lot of problems with jargon.

Finally a useful linguistic trick. Given anything “jargonised phrase” such as “Key performance indicator”, it always makes more sense if you turn it into a proper sentence with verbs and subjects and objects. eg A Key performance indicator is an indicator of performance that is more key (in some way) that others. Or CSF – A factor that is critical to success. See what I mean – it makes the meaning much clearer, already.

Hail Plain English, sorry, I mean “English that is plain (and uncomplicated)”

Phil

Posted in Critical Success Factor, Key performance indicator, environmental KPIs, jargon in performance management, measures and indicators | Leave a comment

What is the correct balance of measures across perspectives

Posted on October 12, 2009 by Phil Jones

A recent forum posting asked “What is the correct balanced of measures across the four perspectives of a balanced scorecard?” They wanted to see if any academics had researched whether it should be 25% each or some other ratio.

This is a massively mistaken question which goes straight to the heart of why many balanced scorecards fail and it is easy to end up with many measures that are not relevant to driving performance.

The purpose of balanced scorecards is not balance.

I’ll say that again. The purpose of a balanced scorecard is not “balance”. Rather, “balance” is a consequence of getting the thinking right.

The purpose of the thinking underpinning a balanced scorecard is to define causality and drive performance (and resulting balance of measures across the perspectives is a consequence).

So for example you would say that improving this skills and knowledge improves the process which gets better results for our customers and therefore better money for us. If you only measures the consequences, and not the drivers, you won’t be balanced.

Therefore an academic would look at an organisation that was measuring those drivers of change (in the learning and growth perspective) and be asking
a) Are they drivers of performance
b) Are they the right ones
c) Did they help bring about change.

The fact that you end up with a balanced set of measures is a side effect. So a sensible academic would not investigate this issue. Yes having 25% in the L&G perspective is more balanced but are they the right 25%? That is the real question. And so they drive performance so that there are changes to the higher perspectives?

Don’t confuse “the balance of a scorecard” with the causality it contains and the purpose of the measures . You are trying to get better at identifying (and measuring) what drives good performance and then influencing it. It is not about having a balance. Its about cause and effect which shows up as balance.

It is this sort of mistaken thinking that causes people to rush around trying to “Balance” the lower perspective by finding a few more learning and growth (usually people) measures to fill it up. Of course when the criteria is, “Are they in the perspective” rather than, “Are they useful in influencing performance” you end up in a mess and don’t actually improve performance.

So please don’t fall into this trap.

If you need it, have a look at this article for an explanation of how cause and effect works.

http://www.excitant.co.uk/pages/bsc_balanced_history.html

Phil

Posted in Alliance balanced scorecard, Balanced Scorecard Perspectives, Balanced measures, Performance Management, Process Perspective, cause and effect, drivers of performance | Leave a comment

How you frame performance management discussions

Posted on September 21, 2009 by Phil Jones

Do be aware how you frame a performance management discussion

I often encounter clients whose experience of performance management discussions is that they are a game of scoring the maximum amount and justifying a high score or protecting against the potential consequences of a low score.

The conversation is dominated by the consequences.

In contrast, when working with a recent client implementing fourth generation balanced scorecards, at this stage, we encouraged the teams to score their performance and discuss why they had scored them as they did. Not so much justify – as explain. In doing so they learnt about what each other were thinking and also how they were doing relative to others, given the contexts they were each in. As a result some moved scores up – others down and others were pleased to find their judgement confirmed by others.

It was about quality of conversation and learning. The teams were doing this in “learning sets” rather than “performance reviews” which created a different frame around the discussion. Of course they later used it to explain their performance to themselves, the people they supported, their staff and the management and the board. However by then they were confident that they knew how they were performing relative to peers and why, so these explanations were much more that-an explanation – rather than a justification which is what many can become.

So as much as anything it is how you “Frame” the discussions that influences what goes on within them.

So do think about how you frame performance management discussions. What hidden agenda or even explicit consequence is sitting behind the meeting?

Even if you do not have it, others may perceive you have it from past experience. So be aware of these and explicitly mark out ways in which you are breaking this thinking and encouraging conversation and learning, rather than evaluation and game playing.

Phil Jones
Excitant Ltd
Where balanced scorecards work properly and make a difference

Posted in Balanced Scorecard Ownership and Implementation, Culture of Performance, Fourth generation balanced scorecards, evaluating performance, performance management discussions | Leave a comment

Rapidly evolving and changing strategy

Posted on September 15, 2009 by Phil Jones

How do you handle rapidly evolving and changing strategy with a balanced scorecard?

Today I opened my email and found out that my “BT yahoo search is now powered by the nations favourite search engine – Google!”. Yes Yahoo uses Google. It is hard to believe that these rivals from only a few years ago are now in such close cooperation. But that is the nature of Internet technology.

The environment and the strategy are changing rapidly and you need an approach to strategy planning, communication and implementation that can accommodate this rate of change.

Likewise with the credit shortage (I refuse to call it a crisis). Again it came upon us quickly and probably more severely that we expected. We also still have the uncertainty of when it will recover and how it will recover. Again thick business plans will be hard to change.

This is why I like strategy maps. They describe the strategy in short (1 page) documents. You can easily put a business plan behind them that sets out the details of the financials, markets, competitions, competitors and what you are going to do.

When I was on the management team of a dot.com the strategy completely flipped inside six months. We moved from a net market providing auctions for commodities in an industry, to a collaborative supply chain solution in the same industry. The strategy map could see that coming: On the main strategy map for the net market auction model, we introduced the supply chain capability, initially as a small theme to the right. Over six months the auctions business slid into a thin sliver of a theme on the left as the collaborative supply chain solution took hold with the customers and became the focus of attention (and the source of money).

The business plans, finances, resources, staff all followed behind as the organisation rotated through 90 degrees to provide a completely different solution to the same customers.

Strategy mapping helps you explain startegy as it evolves. It provides a quick and effective way to describe the strategy, that is rich and powerful in its description yet at the same time can also change and evolve far more easily than a written document.

That is why strategy maps are often used to describe alternative strategies when they are being tested and thought through. It is also useful to develop a strategy map when acquisitions are on teh horizon. It means thatyou concentrate on the main business whilst developing your startegy for the acquisitions and can then feed the integration often two businesses into the strategy map.

Having the staretgy map allows you to then develop the measures, targets, projects, actions and responsibilities (The balanced scorecard) beneath the startegy map. That way as the strategy map evolved you can more easily refine the balanced scorecard with reference to the strategy map.

For more on strategy maps and balanced scorecards, read our introduction to strategy maps on the main website http://www.excitant.co.uk/pages/strategy_mapping.htm

Phil Jones

Posted in Fourth generation balanced scorecards, Strategy Map, rapidly changing stratgey, rapidly evolving stratgey | Leave a comment

The importance of accountability

Posted on August 22, 2009 by Phil Jones

The importance of accountability.

I am a great fan of the “Undercover Economist” series of articles by Tim Harford, that appears in the Saturday Financial Times. This week Tim asks why would people so poor that their weekly wage would not even cover the cost of the newspaper put their children into private school, when there is a free public school. Also why would poor people in poor and developing countries with free healthcare choose to pay to go to private doctors that are in these countries less qualified than their public sector free ones.

Seems bizarre. The answer according to studies by Das and Hammer that Tim Harford quotes is that these doctors give their patients much more time and attention. In other work by James Tooley, the teachers pay more attention to their pupils. In contrast some public sector teachers were asleep in the classroom when a film crew arrived, even though they knew the film crew were coming.

The answer they say, is accountability. If you are paid directly by your patients or pupils’ parents, even though the amount is really small, you have to demonstrate care and attention, otherwise they will not come back or use your service.

In contrast, the disconnection between the source of the income and the people they serve in these free public sector leads to a lack of accountability. The conclusion from the article is that a little accountability goes a long way. It also concludes that low-cost private sector services should be nurtured.

If we consider how this maps across to the UK public sector we get an interesting contrast. The accountability to the UK government departments is often about penalties for failing to reach targets, rather than accountability to the people being served. This is being redressed through surveys from the populations of users, but this is not accountability, but feedback. The accountability is still through a third party (or third parties) rather than to the “customers”.

If we look at private sector responsibilities, there is or course an accountability to management and the owners (shareholders) but that is only served if the public themselves are served as well and continue to come back and use the services.

So think about the links you can make between people’s actions and their accountability to the
people they serve: The public, the customers and even other parts of the organisation, for back office functions.

Contrast the thinking of “Service level agreements” vs “service” and “accountability”. SLAs may encapsulate part of the issue but certainly not all of it.

And think through the thoughts of the private sector doctor in these poor nations who are taking the time to listen to their patients, understanding their needs and helping them as much as they can. In contrast a UK NHS Hospital Consultant I spoke with was extremely frustrated to be told that he had only 5 or 7.5 minutes in which to conduct an appointment.

Where is the accountability in that?

Where is the accountability in your services?

Phil Jones

Posted in Culture of Performance, Performance Management, Tim Harford - The Undercover Economist, accountability | Leave a comment

The failure of single point targets: A tragic NHS story

Posted on July 30, 2009 by Phil Jones

A tragic story of targets and “Incentives”

When explaining how poor targets and measures destroy common sense, I often cite the case of Accident and Emergency (A&E) times in UK NHS hospitals.

The government (through the Department of Health & Social Services DHSS) were concerned that patients were having long waits within A&E. So, to reduce the time, they imposed a target of 4 hours waiting on all hospital A&E waiting times. Hospitals that failed to meet this target were to be penalised.

Now you can see the logic: We want to reduce waiting times for patients, so lets create a target and incentivise the hospitals to achieve it (or threaten them with punishment for not achieving it.

However, what happened? Well arrivals at A&E are very volatile. Sunday lunchtime is a lot quieter than Saturday night. Its in the nature of A&E that emergencies happen. Demand and supply don’t fit well together. So, to avoid punishments and to achieve the targets with the limited resources they had available, the highly rational doctors, managers, nurses in A&E resorted to ways to achieve the targets (and avoid the punishments).

I am sure there were many rational moves to achieve the targets which were not reported. (The BMJ reports on some short term actions in this article). To avoid the punishments some seemingly irrational behaviours occurred. One solution was to cancel operations to ensure beds were available for A&E patients. Another was to have a patient to stay on the ambulance instead of being admitted. That meant that the clock did not start counting. Of course, it tied up a valuable ambulance, but someone judged that penalty was less that the penalty from the DHSS.

Another solution was to address the discharge end of the problem . You see you need to discharge a patient from A&E (either sending them home or being admitted into a bed) to stop the clock. If beds are not available the patient waits in A&E and the (arbitrary) clock keeps ticking. The solution: Put the patient on a trolley but take the wheels off it and designate it a bed. They are still in A&E but effectively not according to the target. Creative yes. Bizarre no.

I have told these stories many times,. They are rational people trying to do their best in a ridiculous situation that does not consider the whole problem – that of patient care.

However last week I was chatting to an ex-nurse. I asked why she left the NHS and she told me a story that make this even worse.

She said that whilst on a ward she had two critical patients who needed a Doctor’s (Consultant’s) attention. They were critical. Life threatening. However, the doctor choose to serve a non-critical patient in A&E, instead of seeing to the critical ones on the ward, because he was driven by the Governments A&E targets (and punishments) to save a target rather than save a life. She was so appalled by this that she left the NHS.

There are some important things to realise about this story:

a) All the actors in this story are rational and have good intentions.

Yes. I know it sounds daft, but they are. The problem is that the context and situation they have been put in has forced them to act irrationally when the bigger context is considered.

b) Each is trying to improve the piece of the picture that is in their control.

Each player has limited resources, influence and control. The DHSS believe that all they can do is impose targets and penalties. It is all they have in their control and influence. They also believe targets work and change behaviour – and they are right!

The Doctors can only work within the context of A&E and can’t magic beds up. The hospital has only so much time and resource available.

c) They all care desperatly about the patients.

Yes, they all care about the patients. It might not seem it, but they do. Think of the dilemma that that Doctor was in: The hospital loses a large amount of its funding if a target is not met which will affect many patients, for a long period, (and that Doctor gets the blame as it was on his shift) or two patients wait for 45 mins whilst a person is A&E gets attention. Think about it.

Again he is acting rationally in the context in which he has been put.

It is not the hospital that has created this situation, it is the DHSS and to some extent the Hospital management. Perhaps the hospital management could have addressed the problem more effectively earlier – we don’t know.

The others players in this story are doing the same

So what is going wrong and what are the solutions?

Have athink while you wait for a later posting, where I will reveal my analysis.

Phil Jones, Excitant
Balanced Scorecards that work

Posted in Fourth generation balanced scorecards, Performance Management, balanced scorecard design, incentives, meaningful measures, measuring performance, motivating change, nhs targets | Leave a comment

How Incentives and punishments affect measures

Posted on June 26, 2009 by Phil Jones

A recent forum discussion I was involved in centred around how you should engage people with measures and the effect of incentives, on the culture of performance.

In my experience there are several aspects that alter the culture of performance, of which incentives (at the normal levels in organisations) is one of the least influential. One of the more dominant is:

Punishments, penalties and the culture of failure:

People will distort measures, or resort to dysfunctional behaviour, to avoid poor ones if there perceive (rightly or not) that they will be punished for them (and whether the punishments are implicit or explicit). It also applies when the messenger gets punished. When the need/urge/peer pressure/management pressure to avoid these penalties supersedes their common sense, you get the sort of silly situations that the NHS had with patients waiting in ambulances to stop the watch starting on 4 hour waiting times.

They will hide information, distort it, be economical with it, or report it differently to avoid this.

A (not THE only) way around this is to retrain the behaviour and address the underlying belief about failure and punishment. If the staff realise that they can learn from the information and it is a failure to learn that will be seen as poor performance you are starting to change the game. Again this relies in explicit messages and actions from management.

Again it comes back to local utility and understanding its value at that level, so people can make informed, sensible judgement and decisions with the information (not necessarily measures) they have at their disposal. They can also refine their judgement as they learn from the situation.

In other words, all of this is about the culture, messages and context that surrounds the discussions about measures. Understanding what is the current norm. HOW you invite people into the discussions, the baggage they bring, the context they have learnt their existing cultural norms from and the extent to which you can signal changes.

For this reason I have no “favourite way” to engage teams in measure discussions. Only questions I ask, and signs I look for, that tell me what the situation might be like. In many cases it can be excellent – in others, more problematic, and I apply different tools in different situations.

Phil Jones
Excitant Ltd
http://www.excitant.co.uk/

Posted in Alliance balanced scorecard, Choosing measures, Culture of Performance, Poor measures, designing measures | Leave a comment