Monday
Sep 08th
| Successful Management of IT Infrastructure Projects |
| Written by David Rippon Director of ULCC | |||||||
Page 4 of 5 1. Keep it SIMPLE Complex objectives lead to complex systems. Clear, simple and easily understood project objectives lead to simple and cheap systems. The Entitlement Card consultation paper was superficially proposing the relatively simple objective of issuing a smart card continuing a biometric identifier to every UK resident. The reality however was that there was a multiplicity of objectives: • Passport card production • Driving Licence production • Proof of Age • Electoral Register • New ways of voting • Personal medical information • Reduction of crime and reduction of some administration processes in the Police force • Access by third parties to use the smart card chip to store and retrieve data • Access by Customs and Excise and the Police to assist in cases of serious crimes This multiplicity of objectives will significantly increase the risk of failure for this £1.6 Billion project and the BCS recommendation was to have only one objective: The reduction of identity fraud through the use of Entitlement Cards 2. Keep the team SMALL Big teams make communications and peer review difficult. Poor communications and an absence of peer review lead to divergence in objectives and system design criteria. Both of these spell failure. In a major development like the Entitlement Card project the concept of having a small team seems an impossible dream. However, there is no reason why the project cannot be split into smaller projects each with its own sub objective and small project team. The overall project then becomes a programme management task. 3. Implement change frequently in small increments One enormous implementation after a multi year development process produces an indigestible lump of change for the User. Frequent implementations of small amounts of change create an environment where change can be managed and, perhaps even more importantly, an environment where change is accepted as a normal part of working life – the true utopia. For the Entitlement Card project there is the obvious question: ‘How do you make a project with 60 million users small?’ This is obviously a challenge but it is one that must be addressed if the risk of failure is to be reduced. There are several ways that can be investigated: • phasing functionality • phasing geographically • Delegating decision making capability to the individual organisational units (perhaps the best solution). 4. Manage user expectations If User expectations are allowed to grow beyond what can be achieved within the project budget the project is doomed. All members of the project team must have the same view of what the project will deliver. In the Entitlement Card proposals this is clearly dependent upon the simplicity of the objective being retained. 5. Manage the suppliers Supplier or vendor management is often a prime cause of failure. Care must be taken that the right relationship is created which plays on the Vendor’s strengths and minimises the weaknesses. Vendor management is NOT about obtaining the lowest possible cost from your supplier – it is about forming win-win partnerships based on mutual respect and governed by fair and equitable legal agreements. Supplier involvement is essential to most IT projects. If they are treated as enemies or scapegoats and their profit margin is constantly attacked then their motivation to succeed for the project team will obviously be diminished. They have to be welcomed into the project team as a fully participating member. Central Government projects, such as the Entitlement Card project, are usually dependent upon a small number of key suppliers and representatives of these suppliers often become the only IT expertise on the project. This is clearly a high risk situation which, quite apart from any other issues, leads to a significant conflict of responsibilities for the supplier representatives on the project team. The key recommendation of the British Computer Society is that IT professionals with significant experience of managing IT development and service delivery in end User organisations should be employed to provide independent and professional input to the project teams and to manage the IT vendor organisations. 6. Use a standard development methodology If each IT member of the project team uses a different process for developing their part of the system then the Users in the project team will quickly retreat in confusion. Speaking the same language is the key to integration across the IS and User functions. And a standard systems development methodology is fundamental to this. This also applies to the clerical aspects of project management. However, care should be taken to correct the commonly held belief that the use of MS Project as a time recording system constitutes effective Project Management. True project management is about the sensible use of risk management techniques to allow early identification of issues (such as changes in functionality) so that the impacts can be assessed and communicated BEFORE the problem actually appears. In the Entitlement Card project one of the key responsibilities of the independent IT professional would be to ensure that all parties to the project were actually talking the same language. 7.Use walkthroughs Peer review at an early stage in the development process is an essential mechanism for cost reduction and quality improvement. The earlier mistakes are caught, then the cheaper the cost of their resolution. Again this would be a key responsibility of the independent IT professional on the Entitlement card project team. 8. Use proven technology Do not under any circumstances use leading edge technology – let some else bear the cost of debugging the latest release of (for example) the database management system. If it is absolutely essential that the project must use the latest release of some third party software then adjust your project costs accordingly – whatever your initial cost estimate is you should probably triple it. The Entitlement Card project will be based upon iris scanning, very much a leading edge technology. Unfortunately the need to use a biometric such as iris scanning means that this risk cannot be avoided. This remains the single biggest risk to the project. 9. Use a component based development process The concept of re-usable objects (initially called modules) has been around for well over 30 years. Nowadays modules are called components but they perform essentially the same purposes of cost reduction and quality improvement. This a systems development standard that the IT professional on the Entitlement Card project should ensure is adhered to. 10. Ban the ‘Not Invented Here’ syndrome Using other people’s ideas and learning from other peoples’ mistakes are some of the biggest cost saving ideas ever. Re-inventing the wheel is a sure fire way of needlessly spending excessive amounts of money. It always amazes me how many normally sane and sensible people refuse to take advantage of the experience and learning examples of the people who have gone before! I will be following the progress of the Entitlement Card project with interest to see if the recommendations made by the British Computer Society are adopted. If they are not then this £1.6 billion project, which is already scheduled to take 9 years from initiation to full implementation, could become one of the biggest IT fiascos ever! |
|||||||
| < Prev | Next > |
|---|
FOUNDING PARTNER
PARTNERS
OOPS. Your Flash player is missing or outdated.Click here to update your player so you can see this content.
News Feeds
Copyright © 2005 - 2008 LeadSure.
Comany No 06216310
