Government And IT - "A Recipe For Rip-Offs": Time For A New Approach - Public Administration Committee Contents

Examination of Witnesses (Questions 141-222)

MARTIN RICE, DAVID CLARKE MBE, JANET GROSSMAN, AND SUREYYA CANSOY

15 MARCH 2011

Q141   Chair: May I welcome you all to this evidence session on the use of IT in government? May I just alert you to the fact that we have a very busy morning in the House of Commons this morning? Several members of the Committee unfortunately have other duties in the House, specific debates allocated to their name, duties on other Committees and so on, but don't let the thinness of attendance lull you into a false sense of security. I notice the cameras are not here, so do speak freely although you are on the record. To start with, will each of you identify yourselves for the record?

Sureyya Cansoy: Sureyya Cansoy from Intellect, the trade association for the technology industry in the UK, representing 780 technology companies.

Janet Grossman: I am Janet Grossman, and I am the chair of the Intellect Public Sector Council. In my day job, I work for a company called CSC and I should also tell you that I am a former civil servant.

David Clarke: David Clarke, Chief Executive of BCS which is the Chartered Institute for IT, equivalent to the Institute of Chartered Accountants, for example.

Q142   Chair: The British Computer Society?

David Clarke: The British Computer Society, although we actually use Chartered Institute for IT rather than the full words of BCS these days. We are an impartial organisation, which is totally independent and self-funding. We have 70,000 members, who include world-class members on pretty much any subject in IT. I think you will find that most of the well­known people in IT around the world are fellows of the BCS. What we do is offer good, professional advice, impartially with really no representational links at all. We are the professional body for the IT profession.

Q143   Chair: What rather confirms all of one's worst prejudices about the IT industry is that your name belies what you actually do.

David Clarke: I don't want to go into that too much but when BCS got a Royal Charter in 1984, it was in the name of BCS and we would have to go through changing the Royal Charter to actually change the name of the chartered institute, so it is a bit of a project for us there.

Q144   Chair: Again, I will refrain from drawing attention to the parallels. Mr Rice?

Martin Rice: Martin Rice, CEO of a software company called Erudine that specialises in agile technologies and agile service delivery. I am also co­founder of an organisation called UK Innovation Initiative and a former vice­chair of the Intellect SME group.

Chair: Well thank you all very much for joining us, it is very much appreciated. Mr Halfon, who has to leave very shortly, is going to ask the first question.

Q145   Robert Halfon: Thank you, I do apologise, I have a debate on something in my constituency downstairs. Why do you think that such a small number of companies are awarded the majority of Government IT contracts?

Sureyya Cansoy: Shall I attempt to respond to that? First, we need to recognise that it is very difficult for smaller companies, and indeed new entrants to the public sector market, to win business in the public sector market. We are very encouraged to see that this Government are taking the SME agenda very seriously. We saw some important announcements made by Francis Maude at the Treasury just last month, attempting to open up the market to smaller companies, social enterprises, charities etc. As a trade association, 60% of the companies that we represent are actually SMEs—small and medium-sized enterprises—so we are very much encouraged by that. Having said that, yes, there is an issue and there are practical things that could be done to address it. The most important thing is to look at how procurement currently works. The current procurement process in the UK Government space does not help smaller companies or new entrants coming into the market. By improving the way procurement works, we can open up the market to not only smaller companies but all sorts of other organisations in this space. So yes, there is an issue—we recognise that—and the measures that the Government have taken so far are very encouraging in terms of addressing those, and we are really looking forward to being able to work on some of the details of those new Government initiatives.

Q146   Robert Halfon: But do you think some of these big IT fat cats have too close a relationship with the civil servants and, because it has gone on for so long, it is part of the de facto system?

Sureyya Cansoy: The key to a successful project or programme is a real partnership between a customer and a supplier. In a sense, you would want a strong partnership­based relationship between a customer and a supplier, albeit not a cosy one, and perhaps we need to accept that. There is some comfort zone issue in that procurement customers at times might prefer companies that they know well and that they have an experience of working with. I will pass on to my other colleagues to add to that.

Chair: May I just interrupt for a second? I forgot I wanted to place on record a potential conflict of interest. The Chairman of Fujitsu Europe has been a family friend for a great many years and our two families know each other extremely well. I just wanted to put that on the record. Carry on.

Janet Grossman: I have been on both sides of the equation as a procurer in Government of very important IT systems that deliver citizen outcomes for the poorest in society, including pensioners, and I am now with a supplier. I will tell you that, as a small and medium player with innovation, the cost of entering a procurement cycle can be life threatening. As the Government contracts tend to be very, very well­prescribed, very detailed, long and very big, if you are a small innovator and you want to do something radical or even a bit different, it can be very hard for you. As a taxpayer, as someone who wants to deliver citizen services and all those good things, what we need to do is broaden the ecosystem however we can. That means that we help make it easier for them to enter the cycle, that we encourage the big guys to partner and change with them and that we look at the very, very best in the world to bring to the UK. It is radically shifting but not fast enough and we all stand for a lot of change in that area. In terms of the cosiness, when you have very long­term contracts you can become too familiar, that is a fact. However the fact that the economy is suffering and people are having to get out of their comfort zones, is helping people to look at each other in the mirror and, on both sides of the aisle, challenge each other to get better value out of these contracts so I think we are at a quite important tipping point.

David Clarke: When you work with people over a long period of time, clearly then relationships form. There is some evidence of that, but I agree with my colleagues here that the root cause is the procurement programme, because it is very expensive to bid for large projects. The bigger the project, the more expensive it is to bid, and because of the sheer scale of a lot of the Government contracts, that excludes all but the very largest companies from actually being able to afford to bid. The UK rigorously follows the EC procurement directives—more rigorously that anyone else in Europe—and its aim is to be more transparent and to have more competition, but the effect is that the procurement processes are much longer, which makes it more expensive and less people want to bid. There is a fundamental issue about the procurement process that needs to be fixed first and, if that isn't fixed, nothing else will be fixed. That makes it very difficult for smaller companies and even some very large companies who look at whether it is worth spending that money to bid—it is only worth it if they win. If they are new into this area for the first time, it costs them more than someone who is doing repeat business—they will know how it works—so it is a lower cost to the people who are already in place. So the whole process actually works in favour of the existing suppliers, which is the fundamental problem.

Martin Rice: I am always interested in root cause analysis rather than how we deal with a symptom of what is going on, and my understanding is that in 1918, the Haldane report for the Ministry of Reconstruction made the decision to structure in vertical silos for each Department, which has carried on. Now we have Departments who produce what the Department needs, so we don't get reuse. If we think in terms of verticals, you will always buy IT in the big, but you are reinventing the wheel all the time. So something being developed in that Department is being done there, and the problem has been solved. If you start dividing it horizontally, which is what is happening in the cloud in the rest of industry, you can start procuring the services in much more subset parts and you can assemble services so that you procure IT in the small. I also feel that because of the vertical alignment of buying IT in the big, it has created an oligarchy, which is a very dangerous situation. It is much worse than just a cosy relationship, and what needs to happen now is that the oligarchy has to be destroyed. You don't deal with an oligarchy through talking; you change the environment and you destroy the oligarchy.

Q147   Chair: So which is the oligarchy—the industry, the Government or both?

Martin Rice: No; with the suppliers who have the majority of the work, it's the tail wagging the dog now. It is not the companies that are at fault; the Government procure badly—they procure in the big, and they allow these contracts to be let. There has to be a move to what the rest of industry is doing, which is buying horizontal services, such as ata centre services, and people use them.

Q148   Robert Halfon: Is there an artificial cartel with the big companies crowding out the smaller ones?

Martin Rice: It is dangerous for me to say yes, but I understand what you mean and it is close to that.

Q149   Chair: You can't be sued in here.

Martin Rice: Then I believe it is.

Q150   Chair: You believe that it is a cartel.

Martin Rice: Yes. I believe that everybody knows they will win a proportion of the work, and they are careful what they bid for.

Q151   Chair: Do you think they talk to each other unofficially?

Martin Rice: Unofficially, yes.

Q152   Chair: And does that mean they actually decide, "You go for that contract and we'll go for this one"?

Martin Rice: I don't know if it is as much as that but I know that if they win one, they will bid for other ones knowing that they will lose some and they will not put as much effort into the bid.

Q153   Chair: Well it is very refreshing to have such frankness and, if I may, on behalf of the taxpayer and the public, there is a very strong suspicion that this is the case and when you say, Mr Clarke, that the system militates in favour of the large companies, that is basically what you are saying, isn't it?

David Clarke: I have no evidence to say to you whether there is a cartel or not, I am afraid. I cannot tell you but the system certainly militates in favour of that.

Q154   Chair: To an outsider, frankly it looks like a racket, because the taxpayer is losing billions—we are spending billions of pounds on systems that don't work properly—and the sector is as profitable as ever. We are going to come on to the shortage of skills later on, but why is the industry so adept at exploiting the lack of skills in government and making money out of it. That is what you are doing isn't it? Intellect.

Janet Grossman: I will take that one. It was by design, if I may say so.

Q155   Chair: By whom?

Janet Grossman: In the '90s when the big outsourcing contracts were let, the business case relied on transfer of intelligence and knowledge into the supplier community and it went too far. Government did not retain enough balance on that side and through successive Governments and budget pressures etc it has not been improved.

Q156   Chair: So it's the Government's fault that you make profit out of failed systems.

Janet Grossman: Oh absolutely not, no, absolutely not. We take full responsibility where we have failed, make no mistake about that.

Q157   Chair: So you lose money.

Janet Grossman: Sometimes we do, sometimes we don't.

Q158   Chair: Take the identity card system: who were the main contractors on the identity card system?

Janet Grossman: My company was one of them.

Q159   Chair: Which is?

Janet Grossman: CSC, but I was not involved in it—it was before my time so I will be limited in what I can tell you.

Q160   Chair: Okay. Do you think your company lost money on that contract?

Janet Grossman: I don't honestly know.

Q161   Chair: Well they are not going to advertise if they made money, are they, because it is a bit embarrassing?

Janet Grossman: I will tell you that it is in the public domain. You mentioned your counterpart at Fujitsu. Fujitsu lost money out of a contract and exited. I am sure all the big boys have done it at some point or another, and they have also made money on contracts, so it is a mixed bag.

Sureyya Cansoy: Perhaps I can add something on the ID cards. As Janet has said, the industry takes responsibility for the mistakes it has made in the past. As a trade association, we have both large and small companies in our membership so we are able to see different sides of the argument as well. I attended the Institute for Government event a couple of weeks ago where they launched their latest report on Government ICT and, as you may know, Ian Watmore, the Government's chief operating officer, was the keynote speaker. He made a comment about ID cards which I found really striking. He said that we always talk about the Government ICT failing but what we don't really consider is that often ICT is there to implement a Government policy and a Government business change programme. He was referring to the policy decision to introduce ID cards and the policy decision to cancel IT cards, and the technology elements were independent from that process. There is something to think about there—it is not only technology. We need to think about the wider policy and why that has an impact on how the Government do technology.

Q162   Chair: I think we all accept that policy churn has an effect on the cost of IT. But the ID cards project was way off target before it was cancelled, wasn't it?

Sureyya Cansoy: I don't know the details of the project.

Q163   Chair: And there are plenty of projects, like the Rural Payments Agency and the Child Support Agency, for example.

Janet Grossman: As a taxpayer it is a bugbear of mine as well; RPA started with a policy that could not be implemented. Should the ICT industry have raised its hands sooner? Absolutely, and the Child Support Agency is a similar thing. We both have to sit down and look at the outcome to the citizen or the taxpayer first, craft policy and delivery mechanisms that are deliverable and then put the IT around it, not the other way around.

Q164   Chair: So the industry is becoming aware that the perception that you are exploiting the dumb customer is not acceptable any more.

Janet Grossman: It is not acceptable to us either.

Q165   Chair: And is that going to stop? How are you going to stop it?

Martin Rice: Identity is a really interesting one. I agree with you; I think the IT industry should publically apologise to the citizen for the rip-offs of the last 10 or 20 years. The Martin Read report in 2009 said something like: the UK is being charged 23% more than our peer nations for no discernable benefit so we are ripping you off as an industry. I feel that very strongly. People only started making any noise about it after the Government had the strength to propose a moratorium that scared the willies out of the industry, and it was a good thing. Identity is a very interesting one; you can go to IT companies and say, "Can you do identity?"—and this is where Government is not a good customer—if you ask for something, industry will happily tell you that they can charge you a lot of money to deliver it. Facebook deals with identity for a 12^th of the world's population and they did not have anything like the budget the Government has to deal with 70 million people. But we are not bringing the learning of these paradigms to bear; we are reinventing the wheel each time and it should not be allowed. As a taxpayer, I am very angry about this and it should just not be allowed. A lot of these problems have been solved; they are not being brought to the Government because of the oligarchy. It is not in a profitable interest to bring you these paradigms. That is why I feel the oligarchy has to stop and Government has to start looking at how we can learn from these organisations: very clever 24 year old people, dealing with 500 million people regularly. In another real case which was interesting, The Guardian run 'Hack the Government'[1] in which, basically, geeks get together and do clever things; four people in two days produced the equivalent of a multimillion pound DWP website for Jobcentre Plus. In two days they had a globally scalable website that you could use to find out what jobs were in your area. It was a better experience for the user. They couldn't keep it going because the Post Office wants to charge too much for the lookup of the postcode. The DWP know about this but they haven't adopted it. It cost two days, four people, and delivered a better experience but they would rather carry on going to the same supplier. It is criminal.

Q166   Chair: But the industry has locked the Government into these very large supplier contracts. You insist on these exclusive arrangements, don't you?

Martin Rice: Stop them. As an intelligent customer, just stop placing them.

Janet Grossman: It is multidimensional. The contracts are let over a long period of time because of the cost to the Government's civil servants to procure, evaluate and all that, moreover they get the best value if the cost and the application is spread over time. That is changing dramatically as we go to software as a service and things are more spotty and dynamic, so it is a two­way street. I will be honest with you; yes, we get better return on our investment, as anyone would, over a period of time. It suits the Government as well because they get to lock in a procurement and don't have to repeat that cycle over and over again. It is a two way street and we both have to address that.

Q167   Chair: Anybody else?

Martin Rice: I feel nobody is fully to blame here; generally as an industry we have taken advantage of a non­intelligent customer who made a quick saving outsourcing everything a decade or two ago. The fact that the industry continues to take advantage is wrong. There are paradigms out there that we, as an industry, should be bringing; the Government should be listening and the Departments will not let those paradigms in. The cosy relationships exist and it will keep those out and I personally believe that they cannot let one success story get through because it will open the floodgates. You only need one or two successes and it will open the floodgates.

Q168   Chair: We are going to get to more specific questions about Open Source and agile later so we will deal with that then. Mr Clarke?

David Clarke: I am struggling a little bit because, if you take the difference between the Government and the private sector, this would never happen in the private sector. It will not happen in the private sector, because those companies will not let it happen—the customer does not let it happen. There is a fundamental piece here where the Government have to get the skills to stop this happening. I do not represent the suppliers at all, but they are commercial organisations and the Government have to be in a position to manage that, with the skills to make sure that doesn't happen. That is where the fundamental issue is. That whole skill set was outsourced in the '80s and has not been properly replaced, and that, to me, is fundamentally what needs to change.

Martin Rice: I agree with what you are saying but for any industry in a supply chain, if it is a professional supply chain, everybody has a duty of professionalism not to take advantage of who is below them in the supply chain, or take advantage of the customer or the person above. We all have a duty to educate each other so I do think that the Government are lacking the skills and that they need to learn more. It is our duty as an industry not to take advantage of it while it is vulnerable. What I am seeing at the moment is the Government making a lot of noise—which is good— and I don't mean noise in rhetoric; they want change. I am seeing rearguard actions being fought everywhere and contracts being extended to get 10% savings. I believe the HMRC has an extension to 2017 because it reduces costs, which precludes innovation into HMRC. We need to stop this, we need to educate Government and we need to bring in the paradigms.

Q169   Chair: So you think the nature of the contracts that the large companies negotiate with the Government are protectionist by nature?

Martin Rice: Yes

David Clarke: Absolutely.

Q170   Chair: Well then, why does the industry go on insisting on them?

Sureyya Cansoy: Can I come into the debate here? The industry understands, as I said before, some mistakes have been made in the past but it also takes two to tango. As industry takes responsibility, Government also need to take responsibility for some of the mistakes. However, what we are now seeing is a real window of opportunity to do things differently. We talked about Government's initiative launched last month about making it easier for smaller companies and other types of organisations like social enterprises etc. to do business with government—we will come on to areas such as Open Source and agile, as you have suggested. We do need to look at this in a positive manner. Industry is committed to working with Government to make this work and the industry understands that its success depends on making Government ICT work, and there is also an element in this for us as citizens. I cannot see my life as a citizen working without the contribution of technology. Every service that I use in the private sector is delivered to me by the help of technology. The citizen expects the Government to deliver the same standard of service to them through the use of technology. So we really need to start looking at this in a positive light rather than thinking about the past too much.

Q171   Chair: Well I do not want to dwell on the past, except to say that our customers, our voters, do expect us to ask about this. We are going to be just as beastly to the Government, I can assure you; as you say it takes two to tango. But there has been an inevitability about this for the last decade, every time the Government embarks on a large IT programme like the National Health Service patient record system—the Government has cancelled that one too, but it was just doomed to fail wasn't it? Why didn't the industry just say, "Government, you have the wrong idea, go back to the drawing board because different ways of thinking about this would produce much cheaper solutions," instead of going for large mainframe databases.

David Clarke: We submitted three reports to the Government and NHS saying exactly that over a period of about six years.

Q172   Chair: And they were ignored?

David Clarke: Totally.

Q173   Chair: Well I think a note about that would be jolly useful to put in front of the Government. I would be grateful for that.

Martin Rice: I concur with that as well. The Government talked to certain companies and individuals for advice and there is a wealth of advice out there in the industry where people will tell you, "This is the wrong thing." I am not an advocate of large companies or SIs, as I am small business, but I do have to interact with SIs if I am to deal with any sort of large contract. If a systems integrator puts in what is called a "non­compliant bid" to Government, they are discarded, it is not listened to. So if the Government asks for something and you do not comply with that bid, as an SI, you will not be considered because if you start saying, "We think it is flawed", you will not get the work. I have never been in that tendering system because I am so far down the supply chain, but these are the apocryphal stories that you hear. There is a wealth of advice given to Government that is not listened to and it really should be.

Janet Grossman: I have been in the position many, many times where they say, "You are missing a trick here, why don't you do it this way? Banking does it this way, why don't you?" and you get the "Oh, oh, tender's out, we can't talk to you, we can't listen, we don't want to hear it". So there has to be a way to have an interaction about what is best before Government makes up its mind, particularly before the policy is nailed to the wall.

Q174   Chair: But the industry will be the first to litigate if they feel the contracting process has not been adhered to. Isn't that why the Government are so defensive?

Janet Grossman: It's a cop out to be honest.

Sureyya Cansoy: There is engagement that needs to happen before that situation happens. One of the main reasons for projects going wrong, identified 10 years ago, is lack of engagement with key stakeholders within the industry before the procurement process starts. So what we have been telling our colleagues in the Government is, "Talk to the industry collectively; we understand you might not be able to talk to individual companies because of competition rules etc. but please talk to the industry and use their collective knowledge and experience". At Intellect, we have a service called Concept Viability, which we launched with the support of Peter Gershon, who was the first Chief Executive of the Office of Government Commerce. The idea is that for any given project or programme in Government, the customer can come to the industry and test the viability of their thinking before they put out the contract notice. The industry has the opportunity to tell the Government department or agency honestly whether the project or programme is designed well, whether it would work, whether the commercial arrangements are the most appropriate ones, whether their budgets are realistic etc. And we are urging more Government departments to use services like Concept Viability and to talk to the industry collectively.

Q175   Chair: On the question of SMEs particularly, how can we get more SME involvement?

Martin Rice: I am passionate about SMEs but I don't think they have a right to work. It is dangerous to say that 25% goes to them; they should only get the work if they offer the best value. John Suffolk started the G-Cloud strategy a year ago; we put a full­time person on it for free, gratis, because it was a looking at how we break it down horizontally and how we combine the small. As soon as Government starts buying IT in the small instead of the big the best person with the best value will win because there are much smaller contracts for each small subset part. You do not start buying £100 million, £200 million or £300 million systems that are never going to work. You start saying, "I need a computer to process x", and it will cost threepence. Anybody who needs to use that uses it, and you can work out all these different services. Then you allow the SMEs an open playing field.

Q176   Chair: But there is a conceptual conflict here, isn't there? On the one hand you think, 'Well buying bulk must be cheaper'.

Martin Rice: I didn't say buy, it should be a pay­as­you­go system; the world is going to pay­as­you­use. If I get a BlackBerry, I do not pay £300 to £400 for the physical phone. I get a phone, on a contract and I pay an amount of money per phone call. If I use it a lot, I pay for more time and if I don't use it, I don't get charged. I don't pay for the infrastructure of the phone network—I could phone Australia—it is a pay­as­you­go. The rest of the world is moving to a pay­as­you­go service and the Government isn't. You shouldn't be paying the capital expenditure, you should be specifying a requirement and the industry should be building it at their expense, and it should be multiplicity of supply for each service. You should be able to bring the competitive market into no­locking contracts. There is so little that Government is doing that couldn't be delivered that way and if you stop buying large capital expenditure projects and say, "This is how we are going to do it", you are forcing industry to have to interact that way. While the DWP and HMRC are happy to spend £2 billion at a pop, industry is still going to keep coming—they're big boys—and taking your money. You will not move to where the rest of industry is going. You have the ability to do it. You own the cheque book, close it.

Q177   Chair: What you are basically saying is that the Government's idea of trying to allocate a proportion of their traditional spending method to small businesses is doomed to fail.

Martin Rice: I think it is doomed to fail. I am more interested in looking at it another way around. You have a little old lady who has to interact because she is vulnerable; she has a certain amount of benefits she gets. Whom does she trust? She trusts her local Post Office. Let them deliver the service; she trusts them and she has got a relationship.

Q178   Chair: Are you seriously suggesting each post office should be allowed to buy its own IT system?

Martin Rice: No. The front-end service systems can be delivered by companies like mine and others. We could put in the systems in the same way as you go and buy a lottery ticket, because somebody has decided to put in a pay­as­you­go service. All the infrastructure is in place, because it is profitable to do so. The delivery of most front-end Government services can be done through the places that the citizen already has a relationship with. Why can't Tesco or ASDA run part of the jobcentre? We shop there, we get banking services there. If Tesco want to invest and produce a front-end to interact and deal with Mrs Miggins and her interaction with Government, why shouldn't they be able to? Tesco don't want to get involved in a £1 billion procurement, but they want more up­sell to Mrs Miggins. So give them the opportunity; tell them what you require and industry will build it.

Q179   Chair: So how does this prevent the creation of a profusion of systems and therefore the interoperability problems that you are trying to get away from?

Martin Rice: That is bringing it back to an IT stance, and that is not the case. I am a complex tax person because I run a business. You trust a trusted party called a chartered accountant to make sure that what I submit to Government in my tax is correct. We trust certain people. So you bring in Open Standards. You say, "To deliver that service, it has to meet x". You audit what they are doing and if it meets that standard, the citizen can go to them. You can bring it in through standards.

Q180   Chair: But we are not talking about the citizen, we are talking about Tesco here and Tesco doing the job search and being the DWP job agency. Are they the customer in that sentence?

Martin Rice: No the citizen is the customer. Because you have defined what you need, if Tesco or ASDA or whoever chooses to build one, anybody can have multiplicity of supply if they meet the standards. A good example of this was the SAP calculations you were doing for home buybacks. The Government published a set of constraints that said, "If you build a system that meets these constraints, you can deliver a SAP service". So companies built systems, they got them accredited to say "Yes we will meet those constraints", and you had a multiplicity of supply. Government did not have to build the system.

Q181   Chair: But how does that apply to NHS records for example?

Martin Rice: NHS records: Google have a service called Google Health in America. They have solved the problem. It is a service that you could use.

Q182   Chair: So the Government doesn't need to build it at all?

Martin Rice: No; it was pig in a poke and I will be careful of the words—is it a cartel? It was a pig in a poke and it was doomed to failure.

Q183   Chair: We will come back to Open Source later but, Mr Clarke, you are sort of nodding.

David Clarke: To some of what Martin says. I have always struggled strongly with: what is the public task? What is the Government role in all of this? My opinion is that it is to set the strategy, the policy and the standards, and to monitor that those standards are being adhered to without necessarily delivering everything.

Q184   Chair: Well when did a large IT company like EDS, Fujitsu or PA say that to the Government?

David Clarke: I don't know; I do not represent those companies. The problem is that the Government could not do that today, because the skills are not there to do it. It is a real mindset change.

Q185   Chair: Yes but the Government are relying on your industry to provide those skills. Here is the industry in front of us explaining that you have those skills and that understanding, why don't you deliver it to the Government?

David Clarke: Because what those companies will do is they will deliver it; they will want to be the delivery mechanism.

Q186   Chair: So self­interest takes over inevitably?

David Clarke: Inevitably.

Q187   Chair: Intellect?

Janet Grossman: I want to talk about—

Q188   Chair: Bringing SMEs and smaller businesses in?

Janet Grossman: Let me give you an example. In DWP there is something called, "Tell us once". If you are unfortunate enough to have a loved one die in this country, you have to notify dozens of people at your time of grief; the local authority, central Government, the Coroner, everyone and their dog—you name it. Exactly as Martin was saying, DWP has partnered with a local point of contact, voluntary sector organisations and everybody else, to make it as simple as possible for those people in their time of need. So there are green shoots out there where this is happening all the time, and they are using us as a means to deliver that service. There are pockets of greatness, but what is missing for me is having an incubation cell where we can explore these things on a reasonable scale and then take them out from there. It is not a choice of big or small, it is a choice of an incubation centre so that we can try things out and see what works and, if we go to Google for health, who is to say that that won't melt down in our country just because it worked somewhere else?

Martin Rice: At least it was delivered and it might work.

Janet Grossman: So I am saying we should try things like Google Health. We should try these things. We should try smaller things.

Martin Rice: I agree with you to a sense. We spend £6 billion and we are always five years off NHS records. It is a pig in a poke. The whole idea of agile development or iterative development is to get something that is good enough out there and then see what the citizen needs; let's do a bit more.

Q189   Chair: We will come to that later.

Martin Rice: That is the idea of Google. It might go into meltdown but at least it is there, it is working and we can make it better. It is adding value for the citizen and it is not just an open cheque book for unbelievable large sums of money for ever, until we cancel it and say "Yes, that is fine."

Q190   Greg Mulholland: Perhaps to take the heat off the industry for a minute, which might be welcome to you, can I ask you to turn to the specific rules around the procurement of IT systems? The average procurement exercise takes 77 weeks, which most people would regard as quite extraordinary as well as clearly unacceptable, and which reflects badly against other nations who have some of the same rules—it is not about the EU rules although they may be part of it. What do you think are the specific problems about the Government procurement process and how would you seek to reform that to make it quicker and better?

Sureyya Cansoy: Procurement is both a complicated and a very simple topic. We have been scratching our heads for the last five or six years, together with our colleagues in Government, trying to understand the causes of that delay in procurement and trying to understand the things that actually make public sector procurement costly and complex in the UK, and then seeking to find answers to those issues. One of the biggest problems in relation to procurement in the UK is, going back to something I have mentioned before, there is not enough preparation on procurement before a Government department or agency goes out to procure and publishes its contract notice. They don't spend enough time understanding the art of the possible, they don't spend enough time thinking about the business outcomes that the project or programme is trying to achieve, whether technology can deliver it, whether they have the budgets, whether they have senior buy­in from the Department etc. So what happens is that you start procurement and then you spend a lot of time during your official, formal procurement process trying to answer some of those questions. That is one problem—one of the biggest problems that exist around procurement in the UK. What really frustrates me is that working together, the technology industry and Government have come up with some really neat ideas and tools to try and address some of the causes of complexity in procurement but we are not really seeing those tools being implemented. You need to ask the question why we are not seeing people in Government departments and agencies using these tools and the answer can be complicated as well.

Janet Grossman: Just to give you a flavour for it—the most basic thing on Earth—you can get a questionnaire with hundreds of questions on it that you have to put a team on to answer. Then you go into competitive dialogue, and I will not bore you with all the steps, but instead of looking at it in terms of how we could do this in the quickest possible time with the least amount of risk, it tends to go in a pattern that is unchangeable. Intellect, in particular, worked with Government over the last four years to come up with some really clever ways to streamline the process, bring in more SMEs, do all the things that I think we want to do, but, honestly, what we lack is the will to change.

David Clarke: The procurement process in the UK at the moment is based on sound contractual arrangements to deliver low risk and value for money though open competition but to get there, we have a very complex, time­consuming, process that is expensive for all parties, which ultimately favours large suppliers. Increasingly, because the cost gets so great, there are fewer and fewer suppliers for each project. We have a system that says that we are going to end up with one or two big suppliers being the only ones who really can bid. The public/private partnership contracts make it very difficult to change suppliers and if we go down that route, we absolutely buy-in to a partnership with one organisation that makes it very difficult to change in the future, with no real flexibility. The UK public sector also outsources far more than any other public sector in Europe; that is for sure. We outsource far more than anyone else. You will find that very few European governments that outsource much at all, but we do. That outsourcing process again passes the knowledge and the expertise to suppliers so we do not have it within Government. All of that limits Government's ability to change the process and to understand what it is being offered. So there are some fundamental issues around procurement.

Martin Rice: It is a very big question. It favours a certain scale of companies because Government is procuring in the big. Over the last couple of years, EURIM Intellect and different groups have looked at what they do in Europe; there has been a belief that the UK adheres to the rules far more strictly than other countries. There could be a lot more flexibility there; it is our interpretation as a country that means it ends up being that long. EURIM sent out a delegation to the Dutch Government to find out what they were doing because they were deemed to be better, and the comments back on the minutes of the Euro meeting were that the Dutch laughed when they realised the scale of the projects we do here. They consider €30 million to be a huge project and I believe there were only three or four projects over €30 million; they still felt that they were not doing it right but they could not believe that anyone would procure such big projects. The other side is that the real innovation tends to come from small organisations. This is not an SME issue, it is that people tend to leave big companies because they have an idea to solve something; they set up a small business and you get elite people working together solving a specific problem. The route to Government is through the systems integrators, and the reality of the situation is that you will be engineered in to the procurement to win the bid because, as part of the procurement, the Government is saying, "Show us the innovation". You are then almost guaranteed to be engineered out once the systems integrators win it, so many SMEs now just don't bother getting involved. Where the innovation that could make a difference could be brought forward, it won't ever make it to the project itself. The simplistic answer is, buy in the small, and don't assume you have to buy in the big; it is not that complex, and it is a con. You have 70 million citizens and there are global companies dealing with hundreds of millions of customers in a very complex way and they do not have anything like the budgets that you spend. Do just close the cheque book, procure in a different way.

Q191   Chair: It is very attractive to Government to just close the cheque book.

Martin Rice: But you can. You really can close the cheque book. The same supply chain won't, but a different supply chain will stand up and you only pay for what you use.

Sureyya Cansoy: On the Dutch reference, what is really interesting is that we have had various Dutch, French and other EU delegations visiting the UK over the last few years because there are certain things about the way that the UK does Government ICT that other countries think they can copy, believe it or not. For example, I mentioned Concept Viability, a service that is offered by Intellect, and the Dutch Government is implementing the concept of Concept Viability—which tests the viability of your project before you start procurement—in their country. If we have these kinds of tools that other countries are copying, why are we not using them?

Q192   Greg Mulholland: It seems fairly clear from what you have said and two of you specifically mentioned it, but do you all agree that the current procurement procedures effectively lock smaller outfits out of the process?

David Clarke: Absolutely

Sureyya Cansoy: Yes. We actually asked a small sample of our members before this Committee Hearing how much it costs to bid for a major project, just to get a feeling of the costs associated with it. It can be as much as £2 million for each supplier to bid for a complex ICT project in Government. That is not an amount that an SME can afford to invest and, actually, it is not necessarily an amount that a larger player can afford to invest in these very tough times, so it is competitive.

Q193   David Heyes: I would like to bring us back to this intelligent customer point which has come out over and over again in the evidence so far. Mr Rice was exhorting Government to behave more as an intelligent customer, and David Clarke said something similar. But Janet Grossman, you used to be part of the Government's intelligence, how can we get you back? Can we afford you? What should we be doing?

Janet Grossman: The happiest time of my life was working for the UK Government, helping citizens, and I live for that. So taking the personal out of it, peppering the UK Government with a few people who have the will, the desire and the expertise will make a huge difference. In the current economic environment, it is not popular to bring in experts on anything, but if you were going to spend a few bob on experts, you would do it in the area of procurement and being an intelligent customer. That is my opinion.

Martin Rice: I agree with everything that you say but the only thing I would ask is, how do you give that person teeth? You need the passion, you need all the things that you said but you have to give that person the ability to drive something through and if they say it is happening, it is my way or the high way, if it is decided that that is what you are doing, and you are going to do it differently, they cannot hit closed doors or bottlenecks.

David Clarke: I totally agree, you cannot afford not to. You have to do that, almost at whatever it costs, because you are spending a huge amount of money down the road on implementation and a relatively small amount of money getting the right skills in Government to manage that properly first. That would save a huge amount of money down the road. You cannot afford not to.

Janet Grossman: The other weakness Government has, which hasn't come up in any of these papers, is programme delivery. Because you outsource your programme delivery work so much, there is no­one on the customer side to watch out for what is going on internally, and that is another area where you have let too much go.

David Clarke: I just remembered an ancillary point on procurement. One of the things that worry me a lot is the lack of career paths now in this profession in Government. So much is outsourced; that work used to be the career paths of people coming up to become those excellent, knowledgeable people at the top of the tree. By outsourcing a lot of the stuff that you do, you don't have that career path within Government and you have no choice but to bring people in from the private sector because at the moment, you cannot bring them through. It makes sense to develop people and over time you will have that career path back but you do not have it right now.

Q194   David Heyes: Let's develop it a bit further, what particular skills are they? What is the skill set that we need to get back into Government and how should the Government go about achieving that? There is the question of costs that will be difficult, there is the question of ideology as well, because of the very powerful ideology that private is good and public is bad. How are you going to effectively get those important people back into the public sector?

David Clarke: Ideology is not the problem. People do not feel that working in the public sector is a bad thing, quite the opposite, but it has to be the right skills, and those people have to be empowered to do the job. That should be strategic planning and there is no such thing as an "IT project", these are all business change projects enabled by IT. So you really need people to understand the business change element and how IT can deliver that. There are lots of those people around but it is really important to have business change managers, programme managers and enough business and technical skills to know how to get the best out of technology, what the right technology is, when to go agile and when not, and what the security risks are of doing one or the other. Those sorts of skills need to be central and they need to be managed, but then the delivery can absolutely be done in the commercial market place. You need people that can monitor what is going on in these projects; people who are knowledgeable enough to monitor what is being delivered. So there is a whole series of skills but you don't need a huge number of them; you don't need thousands of those people, you need a number of them who can control the delivery. I think that will dramatically reduce the cost of delivery.

Martin Rice: I agree with what you are saying about that person being there from the beginning to the end.

David Clarke: Absolutely

Martin Rice: Not leaving half way through and going somewhere else; that is not what you mean by career development, you need continuity. The main thing I would say is, profit is a dirty word but Amazon's IT is really interesting because they don't have an IT, they distribute things. Everybody involved in IT is focused on driving the transaction costs down, otherwise there is no profit and Government does need profit. You take a certain amount of money from the citizen and anything you don't spend on IT goes to frontline services. You need the people who think that way round, looking at how you can reduce the transaction costs, looking at what other people are doing in the industry and how you can learn from that, not understanding IT that much but having that mindset of: if I make that saving, there is more profit to give to frontline services.

Q195   David Heyes: Can I just link this in with the point that virtually all of you have made about involving industry early in the contract specification? Are you saying that is a substitute for the lack of an intelligent customer or is that in fact the intelligent customer role?

David Clarke: It is the intelligent customer role. If you do not have them, then by definition, you outsource it to a supplier and I think that is not the smart thing to do. You should have those in Government and if you do not have them in at the beginning, then that will have a serious impact on the costs of the project, the time it takes; expectations will not be set properly at the beginning if you do not have that expertise in early. People will be guessing at stuff rather than really knowing what it will actually cost, what it should cost and how long it will take. If you get that intelligence built in, you will have successful projects. Part of the lack of success is due to unrealistic expectations set at the beginning.

Janet Grossman: For me the elephant in the room is, under the noble intention of fairness, the Government does not feel able to bring in people to brainstorm and workshop ideas prior to policy coming alive or whatever, because they are afraid that others will complain they were not involved. In the private sector you are free and open to say, "You seem to know something about this, we are thinking about doing this, what about that? You do too", but because of the noble intention of fairness, you either have to invite everybody in or nobody in and you err on the side of nobody in.

Q196   Chair: But we have this Technology and Business Fast Stream in the Civil Service now, and none of you have mentioned that. I am told that there is a zero attrition rate from that which must mean that it must be a satisfactory job because they are not paid as much as people in the private sector. Do you rate this?

David Clarke: Yes

Q197   Chair: And is it producing good people?

David Clarke: It is, but nowhere near enough.

Q198   Chair: Is it a programme that could be called up?

David Clarke: Absolutely.

Q199   Chair: And you would very much welcome that?

David Clarke: Yes.

Chair: Moving on.

Q200   Paul Flynn: Ms Cansoy, you used the expression, "Concept Viability", which interested me. I was reminded what Cecil B DeMille said: "Before we make a new film, someone should read the script." The idea of people coming from Holland to tell us that, before we embark on a concept, we have to make sure it is viable—do we really need to be told that?

Sureyya Cansoy: It is actually a UK initiative; it was something that Intellect launched with the Office of Government Commerce in 2003. What colleagues in the Netherlands are doing is copying, with our permission, the same initiative in their country because they think it is a very simple and neat idea to use the collective knowledge and expertise that exists in the industry to test the viability of an idea before you take it too far.

Q201   Paul Flynn: But is that not stating the bleeding obvious? I don't understand; would you get some special people in who were experts in Concept Viability to find out whether what you are doing makes sense?

Sureyya Cansoy: The way Concept Viability works is that you take a specific project or programme that you are working on, you bring a group of about 50 or 60 technology companies together and you ask them a number of specific questions about the project or programme. So for example, "Is this doable? Has it been done elsewhere? Are there any lessons that we can learn? Can technology deliver this? Are our budgets realistic? What kind of commercial arrangements should we be following?" Intellect then takes that collective industry feedback and presents it as the industry response to the Government department or agency. It looks quite obvious; you would think that people would say, "Actually yes, that is very sensible, we should be doing that", but you do not always get that.

Q202   Paul Flynn: And the comparison with Amazon: we talked in the past about Tesco; these are huge organisations with very rare errors of one in 10,000, whereas comparatively the Health Service has errors of one in 10 and so on. Are there differences there? I apologise for coming late and if this has been covered earlier. There does seem to be a great gulf between what happens—with the enormous waste, the huge amounts of money that are being paid in the public services—and the often dire results that come about.

David Clarke: Can I maybe come back to Connecting for Health? We did a report in 2006—Concept Viability—that said that the basic concept of that would not work. BCS is quite unique in that we have, in a number of sectors, combinations of experts users and, in this case, medical people and the IT implementers in the NHS; we brought all those people together to look at what was being proposed. We did a report there that said that this concept, which is one big, central system, simply cannot work. That was ignored and so a couple of years later we had another go, and that one was ignored. The idea is that you do need to look at concepts and the viability of concepts. This was a Government policy to do things in which it seemed that the only way of implementing that policy was to have that central system. That absolutely was not the case but, even if it were, would that price be worth the policy? And to have one central database of everyone's record, when 95% of health interaction was within 20 miles of where people lived, was just nonsense. We put all of this stuff in the report. So there is a need on some of these projects to stand back and take a look at the concept.

Sureyya Cansoy: If I can just very quickly mention one example of a small pilot project that used Concept Viability about six years ago, at the end of the Concept Viability exercise, that Government department realised that they did not have the budget for that project and it was not doable. They took the sensible decision not to continue with it for a saving to the taxpayer of millions of pounds.

Q203   Paul Flynn: We are just in the middle of a census on a biblical basis of the entire population when we know could get almost as valuable a result if we took 0.1% of the population and examined them, but we seem to carry on doing these things. Ms Cansoy, in Intellect's evidence, it states that technology currently tends to be considered separately from business change. What does this mean? How would you change the way Government behaves as far as that is concerned?

Sureyya Cansoy: The first question that we need to ask, which David and Janet touched on earlier, is: what are we trying to achieve? Take the HMRC Tax Self­Assessment Service. The starting point for that was: "Okay, we need to make the service more convenient for citizens so they can do it 24 hours a day, they can put their details of their finances on this online form and their tax gets calculated automatically". It is a great convenience and it is saving HMRC money at the same time, so you need to start with the business outcomes that you are trying to achieve first and then look at the technologies that are out there to help you do that. That should be the starting point.

David Clarke: I remember talking to a senior civil servant a couple of years ago about this area and they said that, when they are developing the detailed policy behind an idea they have been given, they always ask the Treasury about the financial situation; they always ask the lawyers about what is legal; and they never think of asking the IT people about what is possible. That is always done as an add­on later and it may be that even more would be possible if they asked earlier. It may be that what they are asking to do simply cannot be done. To leave it late means that it almost certainly will not get delivered because IT simply enables the project. It is not the be all and end all within itself; it has to be a subsidiary of the business change.

Q204   Paul Flynn: Are Governments a soft touch because of the ignorance of IT by civil servants and politicians.

David Clarke: I don't know that I would say "soft touch".

Q205   Paul Flynn: Well you made the point about the large amounts of money that have been spent on projects that do not work.

David Clarke: I don't think that the Government have the skills across the board, at the moment, to get value for what they are investing. They are investing a huge amount of money and they should get a lot more value.

Q206   Paul Flynn: If you were down in the Dog and Duck with a Minister, enjoying a pint of lemonade amongst yourselves, what would you say to them? We have a new Government full of reforming zeal. They are Maoist—they want to revolutionise the system and do it yesterday. What is the best advice you could give them in a few short sentences?

David Clarke: Get the right IT skills, and that is not the technology, it is as much about the application of IT to the business. Get somebody who knows your business, knows what is possible, give them the authority that you give to the finance and legal people to incorporate that. Get them in early, listen to them and you will get projects delivered on time and to cost.

Janet Grossman: Being a bit provocative, I would tell the Minister to get involved and to take personal responsibility, understand what business drivers are in the system; if a Minister stands behind a policy, it gets done. If a Minister stands behind a really good innovation, people want to do more of it. It is really hard to be a civil servant these days; nobody tells you that you are doing a good job in the newspapers. So I would tell the Minister to find pockets of innovation, to love it like mad and to be really tough about accountability, but also to be accountable themselves.

Q207   Paul Flynn: The tendency is for Ministers to run a mile when things go wrong and to put their civil servants up to explain, as far as this Committee is concerned—take Tax Credits and so on—but do you think the Ministers, of all Governments, are defective in this in not taking a personal responsibility?

Janet Grossman: It is mixed.

Q208   Paul Flynn: Do they really know what is going on?

Janet Grossman: It depends on the kind of department it is; it depends on the background of the Minister, whether they come from a public service background or not, and it also depends on the nature of the challenge. If you have a really big challenge going on, they get more involved than if it is business as usual.

Martin Rice: Simply put, it is a guardian on detail. You get clever people, and they produce a fully working system in two days that might be completely what you didn't want but you will be wonderfully capable of telling them it is not what you want when you see something that isn't it. Alternatively, we could all sit there and do a nice big report that nobody reads. Just do fast, iterative spikes—it doesn't matter that it is not scalable, and it doesn't matter that it hasn't got security. Is that what the citizen wants? You can do it in days and then you can start saying, "You have totally got the wrong end of the stick," and you can get everybody involved and get the stakeholders in. But actually build, fast, iterative systems and throw them away.

At Erudine, we follow a process of quest; if we are not sure how to solve a problem, we get two or three teams of two people, and we give them half a day to go away and get creative—there is not a wrong way of solving it. They then come back and we will discard one or two of the routes. We might have two weeks of development and then we choose one. We don't predetermine from a big report, we just keep looking and saying, "That is getting closer to what we think we need", and then you are on the right track. So we don't waste millions because we waste a few tens of thousands doing experiments, and then you know that you are on the right track. So in the pub with the Minister, I would be saying, 'Start funding more of things like the geek cells, and get more people who can do these very quick systems who can ask, 'Is that what you meant?' and get everybody involved in analysing your systems and, if they work, use them".

Q209   Chair: Just moving on, we visited Facebook for example, they use Open Source software and they now have 500 million users. Why are Government so resistant to this method of working?

Martin Rice: It comes back to Mr Halfon's point. You go to the systems integrators, they have tier­1 relationships with certain suppliers; they have to push them forward, they have signed contracts to say they will. Facebook is an interesting example; they developed some software called Cassandra. It is a globally scalable database, they decided they did not want to own it so they put it as part of the Apache Open Source stack. It is available to the world and it has a very large group of developers working on the code. It is a brilliant piece of software and it is free. Now it is not free, because you will pay people to use it, you will pay maintenance support, but you are not locked into an adversarial relationship.

Q210   Chair: How is that parallel to a patient record system or a system delivering a tax disk for my car?

Martin Rice: The patient record is a really interesting one. Rather than building this massive database, which is going to cost billions, the way that Facebook works is that they can access any record anywhere in the world, and it comes up like that. The way that Cassandra was developed, as an example of it, was to allow them to do that. So these have been produced with that specific type of access to data in mind.

Q211   Chair: What about the security problems? Facebook is notorious for its lack of security.

Martin Rice: That is beyond the scope of today to go into that, and I believe it is a red herring. If you were to take a copy of Facebook and say, "Let's use it for Government," it would be unsuitable, but that does not mean the underlying technologies are not capable of delivering this. It is a red herring.

Q212   Chair: And is the lack of Open Source development a reflection of the protectionism in the industry?

Martin Rice: I believe so because it is used, pervasively, everywhere else.

Q213   Chair: I only use that because I feel you need to put your side of the story forward.

Janet Grossman: My company, CSC, does not develop anything without the use of quite a lot of partners. It is a myth that we are using less Open Source than we think. We have a lot of Open Source in government; it is surrounded by security measures that obviously we don't want to talk about in a public forum. It is getting to be part of our real world more and more. Dare I say, looking at the age of myself and some of the people in the room, it is a generational thing as well. Those of us who grew up in the IT industry do not really fully appreciate what is out there that could be rapidly deployed, so we have to look ourselves in the mirror as well. But I think you would be surprised to see that more and more of even the big SIs are using Open Source, Open Standards in their cloud applications in particular and we need to be better at communicating that.

Sureyya Cansoy: If I may add to that, Intellect has both Open Source providers and proprietary software providers in its membership and, whenever we talk to members about it, Open Source providers say, "Yes, we would like to have a level playing field when it comes to doing business with Government" and we are strongly behind the message that Open Source should be used where it makes business sense, where it is the best solution for the problem we are trying to solve and where it provides the best value for money. But at the same time, there are solutions where there is a mix of both proprietary and Open Source software, so often Open Source is hidden in a more complicated solution provided to Government; it is not always the case that it is not there.

Q214   Chair: Is there just one more thing Mr Rice?

Martin Rice: I am yet to find Open Source to have a salesman who takes everybody out to lunch. A lot of procurements are sorted in the wine bar, over lunch or on the golf course, so I see that as a problem for Open Source systems with very good software out there.

Q215   Chair: So Open Source is alien to traditional commercial relationships.

Martin Rice: A friend of mine was an MEP a few years and said as a tongue in cheek joke, "Microsoft had a bigger delegation there than most Member States." So the Open Source community did not have a large delegation there and Microsoft keeps the stranglehold. The word that was used down there is key. There is also a big difference between Open Source and Open Standards, and Open Standards are far more important than Open Source because it sets a level playing field. It really does not matter, whether it is closed source or Open Source is actually a misnomer; anybody can interact within that set of rules.

Q216   Chair: And this relates of course to agile development as well, which you have been referring to a lot in your evidence Mr Rice. Why do you think it is not in the interests of large suppliers to promote agile development?

Martin Rice: The Agile Alliance started 10 years ago; the 10th anniversary is in June. We have been an agile development house since it started. The purpose behind agile as we see it is there is no lock­in and no pre­determined outcome. You just have a vision working with a partner and that it is a true partnership. But you are starting on a journey of saying. "We will give you a working system every month which will be better than the last time and then we will all get together and decide what it will be" but at any point, you can stop the contract. Big companies do not like that; they want five or 10 years' lock­in. Agile has been there, and it has been proven to be a very robust, professional methodology.

Q217   Chair: Would agile have the effect of fragmenting the industry?

Martin Rice: Sorry, what do you mean?

Q218   Chair: At the moment contracting processes have militated in favour of the consolidation of the industry because, if you are a big company, you buy another big company and you get more share of the market. But agile development would have the reverse effect wouldn't it?

Martin Rice: Yes but if Government continues to get the big, you will always get the big company. If you start buying in the small, it suits the small, agile development to produce exceptionally good, smaller services. The key thing about agile as I see it though is that the user gets what they want. The key thing there is what Mr Flynn was saying about the business rather than IT. It is not even the business; if you are building an IT system and someone is going to use it in a Jobcentre, talking to a vulnerable person, what does that person want? Not what did the Minister want, but how do I make their life, sitting and interacting with that person, easier? That is the person who agile wants to sit and help with the development; the user, the people who actually interact with it on a day­to­day basis. There is a standard joke in the software development industry that you bring the user in at the stage when we have finished. You very rarely bring them in right at the beginning and say, "How do I make your job easier?" These people, who do the front interaction, probably know more about how to make savings to the business process than anybody back in Whitehall or the big IT company.

Q219   Chair: Okay, so what skills does Whitehall need?

Martin Rice: Learn the principles of agile.

Q220   Chair: You just need people to learn them?

Martin Rice: Learn the principles of agile from agile development companies, not from the current oligarchy, because they are not experts although are starting to read the manuals and say, "We can do agile." I was really interested in the transcript from last week's evidence session when one of the people said that he could do agile within waterfall. That is a fundamental misunderstanding of the point of agile and it is dangerous that that is on record in Government. If you want to know about agile, talk to people who are delivering agile projects, do not talk to people who are not.

Q221   Chair: Can big business deliver agile?

Janet Grossman: Yes. I am amused here, because when I was in DWP, the Pensions Transformation Programme was all bottom up and delivered by the people who were using it and two big companies delivered agile; that was four or five years ago. We are painting too much of a brush one way or the other. We have a long way to go in the big contracts because some of the big contracts are over specified to not be as agile as they could be, but it is happening more than you think and agile is out there. Just to be a little bit provocative, the problem with agile is, if you are not careful and you are not an informed customer, it can be the never ending change project so I will just say that you do have to put some brakes on it at times and I will leave it there.

Martin Rice: If you look at agile from a certain perspective, you are right it can be never ending, but the purpose of agile is to reduce the cost of change. The whole purpose of agile is to make change easy. Lord Erroll has a lovely phrase; he says that it is the job of a systems integrator to extend the problem, not solve the problem. That's a waterfall; it is a throwaway comment. It is the purpose of agile to simplify the problem and reduce the cost of change. So yes, you keep working on it because the world iterates. Facebook did not have a document saying, "Let's build Facebook, job done, go away"; what they do is say, "How can we improve the service?" It is an iterative, constant change and it is cheap.

Q222   Chair: Mr Clarke, you are going to get the last word.

David Clarke: Thank you. It do not think that it fragments the industry. In the end, if the Government can become an intelligent enough customer, the customer is king. If you demand that, then the big suppliers will respond and they will not walk away from that business. It does not fragment it; I don't think they are there now. The smaller companies are more agile but the big ones will not leave that alone if that is how you demand certain projects are developed.

Chair: Is anybody else burning to justify anything that they feel they need to justify? Well I am very grateful to you all, there have been some frank exchanges; one senses an electricity in the air, that is what we wanted. It has been very informative for us, thank you very much indeed.

---

1   Note by witness: Correction - the hack days were run by Rewired State (see Q245). Back

---

Prepared 28 July 2011