| Previous Section | Index | Home Page |
Mr. Mark Todd (South Derbyshire) (Lab): I am delighted to contribute to the debate and I start by congratulating the Committee on the quality of its report. I say that as someone who 10 years ago served as an IT director in a large company and therefore had to be familiar with the principles of how large projects should be run and, in many cases, how they fail. The Committee's report was excellent and demonstrated to me that even my 10-year-old knowledge is not completely out of date. The report addressed many issues with which I was familiar when I worked to try to deliver major projects. The Committee has done everyone a considerable service, and I share the disappointment of its Chairman at the quality of the response, which did not attempt to address many of the core issues in the report.
As I mentioned in an earlier intervention, I took an interest in the Office of Government Commerce gateway process at an early stage, because with my background in the business I welcomed the attempt to take a much firmer grip of technology projects, to address the risks that might be involved in delivering them and to discover what could be done to control those risks to achieve the desired outcome. I obtained an early gateway review of a change programme in the Department for Environment, Food and Rural Affairs, which was reassuring. I thought that some further steps could be taken, but I am sure that the gateway process has altered since then. I was impressed by the initial start.
I have asked for gateway reviews on other occasions—several times for the project we are discussing—but the Committee's report contains various phrases that echo precisely previous parliamentary written answers that I have received on why disclosure was judged to be inappropriate. That serves us ill in many ways, including in the performance of our role as guardians of the public purse. These are important projects. I recognise that many of the issues involved are substantially beyond the
9 Dec 2004 : Column 1308
ken of virtually all Members of Parliament, but nevertheless we can take advice on the content of documents and they should be available to us so that we can make our own judgments.
I recognise the merit in removing some of the personal comments that might be made that could be attributed to individual members of staff or those involved in the supply of the system. However, enough would remain to be publishable after that process and allow the documents to be made more widely available. Indeed, by coincidence, I received a written answer today to a question to the Chancellor—because the OGC process is the property of the Treasury—as to whether he had reviewed the gateway process following analysis of failures in the CSA system and process change programme. I was concerned to learn that there has been no such re-examination "as a direct result" of the CSA system and process change programme. Perhaps there has been a re-examination that was not a direct result. We await changes in the way in which the OGC operates and the disclosure of its activities in the future. I hope that that will happen soon.
The principles of delivering a project of the kind we are discussing are outlined in some of the papers that the Committee has produced. The first point, which is worth repeating to those who are not comfortable with technology, is that we should forget about the technology. The issue is how to make the human processes more efficient. Unnecessary elements in those processes should be removed and the most sensible way to deal with a particular task should be devised. Then the technology should be applied. All too often, those who commission a programme start from the other end. They say, "Well, this is a major technology project. We can see all sorts of problems with it." The first problem is to define the most effective way to do the job to be done. I admit that I did not see a huge amount of evidence, from what I could tell anyway—we would have liked to have greater disclosure—of that process having been properly undertaken within the CSA before the project started. But the revisioning of how to do the job in question is much the most important part of a project.
The second element is investment in the specification of how to do the job so that it can be put into a system. Again, in complex projects that element is often elided with others. People say, "Well, we need to map this down accurately, but no doubt there will be opportunities to amend it at some future stage." That is wrong, because those opportunities should not occur. The specification should take some time. A large proportion of the project's life should be spent defining the system properly so that it can then be turned into code.
That investment in time, however, should be balanced by rigid discipline. There are references in the report to project management and skills. IT project managers are strange people, and they are not comfortable to deal with, in my experience, because—I would use a rather unparliamentary term to describe the characteristics that many of them have—they are extremely detail-oriented if they are doing their job properly. They are highly protective of the discipline of the project, and will constantly remind people who attempt to depart from it of the consequences. It is therefore not easy to deal with a robust project manager, but their approach is
9 Dec 2004 : Column 1309
valuable, because it is important to resist the desire for change while development is under way. That requires tremendous restraint. We all recognise that human beings make errors, and in the process of designing a major system, people may leave something out or fail to spot something else of importance. The specification programme should remove as much of that risk as possible but, nevertheless, something is bound to be missed. The project manager's task is to say, "I am sorry, we will not change this now. It will be changed in a properly planned process in future. The first task is to deliver a stable system that functions according to the original spec."
There were worrying signs in the project that people were all too free to suggest that something had been missed, something else could be added and so on. It was hard to work out the exact nature of the changes—I can understand the problems faced by the Committee, as Doug Smith gave some rather opaque answers—but I am concerned about changes of any substance whatever. Robustness is therefore needed from the outset. When one is commissioned to produce something that is highly complicated and of an extremely high risk, one must be disciplined and exercise restraint.
Mr. Paul Goodman (Wycombe) (Con): The hon. Gentleman appears to have taken a close interest in the project, so presumably he is aware that Mr. Smith told the Select Committee that about 50 change requests—
Mr. Goodman: I am grateful for the exact figure. Mr. Smith said that the CSA had made 55 change requests. Tony Collins of Computer Weekly, however, said that the figure claimed by EDS is about 2,500.
Mr. Todd: Those statistics demonstrate that this is a hazy area. There was a casual approach to counting the changes, some of which were claimed simply to be screen changes, but they can be of material significance to the system. To be too casual about managing change in the process of development shows either an improper grasp of the way in which a complex project should be run or a lack of transparency.
The Select Committee highlighted the critical value of honesty about progress and deadlines. When I was responsible for a project I was keen, like everyone else, to deliver systems rapidly. Early in my career, I attempted to talk up the progress in delivering one. However, I was firmly told that that was not acceptable. People had to attempt to remove areas of risk in the programme, and time had to be set aside to make sure that the system worked properly. I was not a politician at the time, but I was told that I was driven by the political purpose of persuading my business colleagues that I was doing my job correctly. That mirrors almost precisely the temptations in the DWP project. People felt that they had to say that the programmes would be delivered rapidly when, frankly, the complexity of the task should have made them much more hesitant about such ambitious deadlines.
9 Dec 2004 : Column 1310
I will not dwell further on the gateway process. It is imperative not just for this project but for any other major system project in government that we are given appropriate disclosure, although I accept that there may be constraints on the advice that project leaders receive.
There is a tendency, particularly in political circles, to be obsessed with the peripherals of an IT project or front-end considerations, such as how it appears to the citizen or whether they can go online and do certain things. I am disappointed by some of the presentations on the use of IT in government, as the importance of such things has been greatly overestimated. The critical elements are the infrastructure and the core systems, which are incredibly boring, but are the foundations on which the service is delivered. The appearance of the screen or the way in which citizens can access the system and get the information that they want are, frankly, secondary considerations that are some way down the line in a project. That is a generic issue, and does not apply to the DWP project because, thank goodness, there has been no attempt to allow people to access the system and look at one another's requirements
Finally, as the hon. Member for Roxburgh and Berwickshire (Sir Archy Kirkwood) rightly said, we are undertaking a huge number of complex IT projects at the moment. I firmly believe that we need a holistic view of the use of technology in government that is driven by the concept of priorities. We must consider where the resources most need to be applied. It is sometimes forgotten that they are finite and that we do not have limitless numbers of skilled purchasers and project managers to do such tasks either within the industry or, critically, in government. We must apply those resources to ensure that we deliver our objectives properly, which requires a willingness to say, "I am sorry, that must wait for some time while we complete a number of other projects." Ministerial leadership is needed, and I am concerned that sometimes Ministers distance themselves from the process. I have made private remarks to Ministers about the CSA systems and, given my background, mentioned my willingness to talk to some of the people involved. Almost invariably, however, Ministers were uncomfortable with the technicalities. We need a ministerial focus of some depth on such subjects to lead a strategic process of change in government. It would be too optimistic to rely on individual departmental Ministers, given all their responsibilities, to try to understand and fix these things. A more strategic and holistic view is therefore needed and, above all, a more transparent process. I, too, was delighted to read some of the industry's remarks that suggested that it was not particularly worried about transparency, but were rightly concerned about the effect of some of the projects on its reputation. It is keen to achieve better quality outcomes, and some of the proposals that I have made this afternoon would help to achieve that.
2.38 pm
| Next Section | Index | Home Page |