Instruction to Deliver: Fighting to Transform Britain's Public Services Paperback – 22 May 2008

Be clear what you are trying to achieve, have clear milestones and monitor progress - the relentless means to achievement. Good read as mixes the personal perspective with the technical how to.

Great book. You can dip in and out of this book and you always find something new. Very easy to read.

Arrived and was in the condition as advertised. No complaints here.

Getting stuff done is difficult in any organisation. But if you're trying to reconstruct the UK public services, then it's about as tough as it gets.Barber does a great job in revealing the detail and struggle. It's an excellent book.It's equally appropriate for business, public sector and third sector.Somewhere in the book he talks about "Grind, not Grand" and that about sums it up.

When you intervene by target-setting then you may make the target - but have you actually improved things?Mr Barber has no doubt that he has - his world-view depends on it.There are, however, many (including myself) who would say that this is a dangerously wrong-headed way to attempt to improve complex systems.So what's the problem?What could possibly go wrong with putting a metric on elements of a system (for instance, literacy levels at age X in each school), and then judging each element of the system (each school) on how well they do on improving their individual metric (literacy levels at age X) relative to the target level?Well, whether you think this is a sensible way to proceed depends on where you think the variation in the metric comes from. Is is primarily down to the individual efforts of each element concerned, or is it primarily coming from the characteristics of the system that the element is part of?Barber's assumption is that it comes primarily from each element. I think that he is wrong - that an awful lot of each element's performance is down to the system.And if it *is* primarily down to the system, then metrics such as literacy outcomes need to be analysed from the point of view of the system, rather than used as a direct lever of intervention on each system element. The primary effect of pushing down directly on each element's metric will be the destruction of the metric's informational content (so it is harder to understand and correct the problem that the metric measures), together with predicatble distortions of behaviour as people focus on the metric instead of real improvement.Barber would contend that this kind of problem can be fixed by being careful about what targets you use and how you monitor them.I think that he is wrong - that driving improvement in a system by direct pressure on element-level metrics will always go wrong in this fashion. Barber is making a fundamental logical error - he thinks that because a metric can be expressed on a per-element basis, it can therefore be effectively controlled on a per-element basis (rather than from a more system-level standpoint).If Barber's approach has a basic logical error, then why do people do it?The tragedy is that this kind of target setting plays well politically. This is why it's so popular. Not because it works! After all, the result of applying pressure on a metric is that the metric improves. Result! So it *looks* lilke you are doing good. But in fact all you are doing is making the target setter look good.If anyone thinks Barber's ideas make sense, then I would ask them to read something by Deming and see if they feel differently afterwards. The system-level view is not new - it's decades old. We should not be making mistakes like this in how we attempt to drive improvement in public services - or anywhere else for that matter.

Sir Michael's nostrums fall into the old trap of 'if it can be measured it can be managed' so beloved of management consultants. Never mind what you are managing or whether it can be properly measured. Barber's platitudes are beloved by politicians who continually call for 'evidenced-based policies' which -- although they have their its place once a strategy has been determined -- are a good excuse for avoiding action about what is important and concentrating on the trivial.This is old stock dressed up as new. It stems from a micro-managing centralisation of the worst Blaire-ite type brought about by the failure of a Civil Service disembowelled by the New Administrative Management theories.