Where our taxes get wasted - Government IT projects and its official
It is not the first, not the last…. But this heart-warming story of £390 million needlessly thrown away on a case management computer system for the UK courts identifies in important change in attitude by the Government.
Big scary monsters, out of control
Government projects are big, high risk and high profile – and have a high price tag. Over the last 10-15 years the Government has tried various ways of controlling cost and mitigating risk – by engaging the big System Integrators with proven expertise, by PFI (Private Finance Initiative) where the system is not paid for upfront, but charges are based on some usage metric, and more recently by employing a high profile CIO to drive a hard-bargain throughout the bidding process.
All these rather miss the point – and at last the Government seems to be recognising that.
Focus on the delivery of the software
All the to date effort has been on making sure that a 3rd party, lets call them the System Integrator (SI)) delivers a system to the specification it has been given. But the real business benefits come from 2 areas which are outside the control of the SI:
- the ability to transform the operation – the system is being put in place to drive up productivity and drive down costs. Therefore it is critical that the end-users create a new streamlined business model which will define the system that the SI is required to deliver. Only the business users, the end users, can define this. It needs to be an end-to-end view of the business – not a list of requirements which can be misinterpreted. The SI is contracted to deliver a system to support this business model. If it is ambiguous, wrong, or fluid then the SI will inevitably make a killing out of change requests… especially if they have been beaten down on cost in the up-front procurement exercise and are desperate to make the project profitable.
- the workforce adopts the new ways of working – the benefits are based on the entire workforce understanding how their behaviour and day to day activities will be different – and better. And that these new ways of working will be supported by the new system. This is NEVER part of the SI’s brief and is fundamental to make the most of system that they are being paid to deliver. It is NOT just running some training courses.
Is it only the Government who get this wrong. Of course not. It is just the projects are such high profile and the job of the National Audit Office is to expose them. The track record however is not good – a recent survey found that of the £10 billion worth of UK Government IT projects only 13% were on time, on budget and to specification. So what percentage actually delivered any value?
In the commercial world that are major failures. Fox-Meyer Corp. A bungled ERP forced the drug distributor into bankruptcy. Hershey spent $112m on a system roll-out and had a $100m drop in sales in the quarter after go-live. And the list goes on.
So the basic issue is engaging the end-users in upfront business design and in the roll-out. That is the responsibility of the Government department. You cannot delegate that, or outsource it.
Libra - £390m thrown away to date: the post mortem
Now I said that the Government is realising the implications of this. The recant NAO post mortem for Libra, the failed (failing?) £390million case management system was analysed by ComputerWeekly, the first key lesson was:
standardised computer systems need standardised business processes…. The Department for Constitutional Affairs (the client) “sought automation as a priority before questioning the existing processes”, therefore it was “difficult to obtain a single view of IT requirements across the various committees”, and not “did not have the power to impose the business process change on the independent Magistrates’ Court Committees”.
Bear in mind… THEY HAVEN’T DELIVERED THE SOFTWARE YET – so they haven’t struggled with the business change / adoption issues yet. The full 12 years of history, and the 3 previous attempts makes painful reading.
So what is changing?
These conclusions in the post-mortem are far more realistic – identifying the real cause of the project – which mostly lay at the feet of the Government Department. The problems of previous project foul-ups have always made the SI the culprit.
Ian Watmore has joined as the new IT tsar for Government, leaving a high profile job as UK Managing Partner of Accenture. He is to form a super-group of top information officers across the public sector to help streamline central departments, improve take-up of online services and reduce IT-related failures.
"Simplified business processes and new IT will be required to make possible savings recommended in the Gershon efficiency review, published in July. The review will force departments to effect important reforms", said Watmore.
As a taxpayer, I hope he can make it work.
Big scary monsters, out of control
Government projects are big, high risk and high profile – and have a high price tag. Over the last 10-15 years the Government has tried various ways of controlling cost and mitigating risk – by engaging the big System Integrators with proven expertise, by PFI (Private Finance Initiative) where the system is not paid for upfront, but charges are based on some usage metric, and more recently by employing a high profile CIO to drive a hard-bargain throughout the bidding process.
All these rather miss the point – and at last the Government seems to be recognising that.
Focus on the delivery of the software
All the to date effort has been on making sure that a 3rd party, lets call them the System Integrator (SI)) delivers a system to the specification it has been given. But the real business benefits come from 2 areas which are outside the control of the SI:
- the ability to transform the operation – the system is being put in place to drive up productivity and drive down costs. Therefore it is critical that the end-users create a new streamlined business model which will define the system that the SI is required to deliver. Only the business users, the end users, can define this. It needs to be an end-to-end view of the business – not a list of requirements which can be misinterpreted. The SI is contracted to deliver a system to support this business model. If it is ambiguous, wrong, or fluid then the SI will inevitably make a killing out of change requests… especially if they have been beaten down on cost in the up-front procurement exercise and are desperate to make the project profitable.
- the workforce adopts the new ways of working – the benefits are based on the entire workforce understanding how their behaviour and day to day activities will be different – and better. And that these new ways of working will be supported by the new system. This is NEVER part of the SI’s brief and is fundamental to make the most of system that they are being paid to deliver. It is NOT just running some training courses.
Is it only the Government who get this wrong. Of course not. It is just the projects are such high profile and the job of the National Audit Office is to expose them. The track record however is not good – a recent survey found that of the £10 billion worth of UK Government IT projects only 13% were on time, on budget and to specification. So what percentage actually delivered any value?
In the commercial world that are major failures. Fox-Meyer Corp. A bungled ERP forced the drug distributor into bankruptcy. Hershey spent $112m on a system roll-out and had a $100m drop in sales in the quarter after go-live. And the list goes on.
So the basic issue is engaging the end-users in upfront business design and in the roll-out. That is the responsibility of the Government department. You cannot delegate that, or outsource it.
Libra - £390m thrown away to date: the post mortem
Now I said that the Government is realising the implications of this. The recant NAO post mortem for Libra, the failed (failing?) £390million case management system was analysed by ComputerWeekly, the first key lesson was:
standardised computer systems need standardised business processes…. The Department for Constitutional Affairs (the client) “sought automation as a priority before questioning the existing processes”, therefore it was “difficult to obtain a single view of IT requirements across the various committees”, and not “did not have the power to impose the business process change on the independent Magistrates’ Court Committees”.
Bear in mind… THEY HAVEN’T DELIVERED THE SOFTWARE YET – so they haven’t struggled with the business change / adoption issues yet. The full 12 years of history, and the 3 previous attempts makes painful reading.
So what is changing?
These conclusions in the post-mortem are far more realistic – identifying the real cause of the project – which mostly lay at the feet of the Government Department. The problems of previous project foul-ups have always made the SI the culprit.
Ian Watmore has joined as the new IT tsar for Government, leaving a high profile job as UK Managing Partner of Accenture. He is to form a super-group of top information officers across the public sector to help streamline central departments, improve take-up of online services and reduce IT-related failures.
"Simplified business processes and new IT will be required to make possible savings recommended in the Gershon efficiency review, published in July. The review will force departments to effect important reforms", said Watmore.
As a taxpayer, I hope he can make it work.