The five minute CIO: Niall Barry – Part 1

14 Jun 2013

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

Niall Barry, head of IT at the Irish Government's Department of Social Protection

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

In the first of an extensive two-part interview, the head of IT at the Irish Government’s Department of Social Protection Niall Barry talks about the internationally praised change programme and explains how building in agility helps to respond to demands.

What is the biggest challenge in running the department’s IT – for example, I imagine ensuring welfare payments are sent out every week brings a certain kind of pressure?

The biggest priority has to be the delivery of the income support service to our clients; that is, the 87m payments made under all our schemes every year. This requires systems to be continuously available to support our staff in their dealings with our clients and these systems have to reliably and continuously produce the vital payment instructions to the banks and An Post. I often characterise this as our biggest ‘negative’ priority, though, in that it will only become a talking point if it fails!

The department already had a very wide scope in terms of its ‘product lines’ before the recent transformation initiative under public service reform which involved taking in significant numbers of staff from FÁS and HSE (about 2,000 people and hundreds of locations).

We have done the very basics in terms of providing access for all to departmental systems but there is still a way to go to provide full connectivity and access to all ICT systems from all locations.

The longer-term strategic challenge is in delivering the agility in our systems and approaches to allow us to better respond to customer service demands from clients and the political system.

The number of people claiming welfare payments has increased in recent years. Can you outline the management programme you put in place to handle this?

The challenge here is not really the numbers per se, except in the more recent years. I think that there is a more fundamental issue. In common with a lot of large organisations, there has been frustration at the pace and responsiveness of ICT-supported change over the years and we often hear statements like ‘our ICT systems won’t let us do x’.

When this type of statement is true it can signal either that ‘x’ is so undefined that it can’t readily be done or that the ICT systems are too rigid to respond to business needs – or both. The IS division started several years ago to look at the latter issue and the solution we found is to define, and build in, agility.

So how do you define agility?

We define agility as having three aspects – strategic, operational and technical. Strategic agility is the ability to respond to ‘macro level’ demands – for example, introduce a new scheme quickly. Operational agility means giving staff systems that depart from the traditional transaction processing systems – with their prescribed processing options – so that they can respond better to actual client needs. Technical agility means adopting modern development tools and techniques: Agile programming methods based on an object-oriented approach.

Simply improving ICT service delivery is not sufficient, though. Our investigations led us to ‘Expressive Systems’, now known as ‘Naked Objects’, which was particularly suited to domain-driven design (DDD). DDD is an attempt to collaboratively discover/develop and use a language that is understood by the business and then implement that as directly as possible in software. Our agreed language and set of business concepts is called the Business Object Model and our ICT system is thus the Business Object Model Implementation (BOMi for short).

What effect did this have?

There is an interesting impact on the business as a whole. Once we standardise on our description of the business objects and their behaviour, business processes can be similarly standardised where warranted and we can provide ICT systems throughout the organisation with standardised behaviour and information. The approach thus helps with the ‘requirements definition’ issue (although it can sometimes be difficult to get the full message across and get the right sort of participation).

While there is a way to go, we have achieved sufficiently impressive results in the organisation to continue to pursue this path.

Referring to the increasing numbers aspect, the major impact of the growth in our existing business on the ICT side is to put pressure on infrastructure (while we’d have some capacity for ‘orderly growth’ we would not have excessive spare capacity). As a consequence, we had to devote additional resources to building up an infrastructure that could more readily scale ‘up and out’.

For the organisation as a whole, growth in business means that considerable pressure is put on existing staff and facilities and that means that there is less capacity to respond to change and transformation programmes at a time when they’re most needed. We have had to put most resources to dealing with the operational pressures while trying to ring-fence enough to keep the change programme ticking over.

What were the outcomes of that DDD project?

I assume this is referring to the domain-driven-design (DDD) approach and I’d like to stress that it is not a project, or even really a programme, but a design approach. Results to date are very impressive and spread throughout the business. We have implemented systems in about a third of the time previously needed; implemented budget changes in a fraction of the time; added processing efficiencies to business areas – almost a 30pc increase in workload handled with no extra resources. We also added control measures once and implemented them in all relevant areas and provided much faster customer service. We are well on our way to eliminating backlogs in the areas where the BOMi is fully implemented.

How important was the input of the IT team in developing the programme?

I think the IT team were instrumental initially in evaluating and supporting the approach but its adoption relied on finding people of vision and power in the business who bought into the concepts. It has very much been a collaborative exercise where both ICT and non-ICT employees of the business had to understand the approach, try to implement it and, most importantly, learn from experience and refine the approach as required.

Did you have any external assistance in developing the programme, and what role if any did those service providers play?

Since we started down this path, we’ve had a lot of constructive involvement with various service providers like Accenture, BearingPoint, Fujitsu, HP, IBM, Microsoft and PWC. The original idea came from research we were doing which picked up on the ‘expressive systems’ approach. We invited its originator [Richard Pawson, Naked Objects Group] to come and work with us in implementing his ideas.

We ran a competition in which three industry groups successfully implemented a solution, using different technologies, which confirmed the practicality and generality of the approach to us. We then successfully procured and implemented a major system, Child Benefit, which further proved the benefits of the approach.

We put in place a development framework with three vendors and continued to build our computing platform while implementing various systems. We have now procured a new development framework and are busily working on some of the new systems required by our transformation agenda.

I have to say that all the service providers have worked hard to understand the approach and that each, in turn, has contributed to developing, refining and implementing it. I think particular thanks are due to BearingPoint, who participated on both frameworks and helped enormously in ‘production-ising’ our approach.

What were the major milestones in delivering this programme?

There have been a lot of milestones over the years, as you might imagine from a process that has monthly business releases. I suppose the ones that particularly stand out include the implementation of the Child Benefit System. It was the first full-blown implementation of the approach. It provided significant business benefits and won a number of awards.

Our implementation of the Customer Object was next and it was a painful learning process for us all but it resulted in a system that bridged the old and the new ‘technology stacks’ and provided the basis for most of our systems since.

Our development of systems in the Pensions areas were particularly successful on the business front in terms of productivity gains and again relied on very strong leadership from business management. We are now focusing on the local offices and expect to achieve significant results over the next couple of years.

What does it allow the department to do that you couldn’t do before?

At one level, you could say that it didn’t allow us to do anything dramatically new but that would be to miss the very real productivity and customer service improvements already referred to. I also genuinely believe that we would not have survived the various pressures of the past few years if we had not been pursuing this path. We have achieved at least some of the agility that we were aiming for and have the basis for further innovation in place.

There’s a lot said about the need to “do more with less” – in practical terms, what has that meant for you, and how do you set up your IT team to deal with that: what gets prioritised and what things need to be done differently?

‘Doing more with less’ means two main things to me – replacing existing processes with more efficient ones and investing less in existing work. There’s a limit to how much efficiency you can introduce in a pressurised environment, particularly if investment resources are not available. You pretty quickly get limited to dropping anything that’s not absolutely necessary either temporarily or permanently (that’s a problem in a system which demands absolute accuracy and complete-ness – it leads to interesting discussions with auditors of all types!).

Investing less means that you might not deliver the full scope of a project or even consider the ‘nice to have’ features in a project. While a certain amount of this is healthy, over-doing it results in work that neither the worker nor the recipient are happy with.

I think we all want to do the best job that we can do and I am conscious of the number of people that I work with that are extremely talented and hardworking but that are quite frustrated at not getting the time and opportunity to deliver better solutions.

You can read more about Niall Barry on the Department of Social Protection’s change programme in Part 2 of this interview, next Friday on Siliconrepublic.com.