CIOs seem to have more than their fair share of challenges these days. But spare a thought for Discovery's John Robertson, who has to oversee the IT supporting the particular requirements of a financial services company and a healthcare provider rolled into one. Robertson is Group CIO and has been with the company from the very beginning.
"In 1992, I had a contract with Q-Data and I was an independent contractor to them and others," he explains. "[Current CEO and founder] Adrian Gore had the idea of combining a financial services company with a healthcare company and approached Q-Data to look for someone to take charge of all the technology infrastructure that this new idea would have to offer. So Adrian, Barry Swartzberg and I were there at the beginning and I started designing the basic systems we would need. I joined full-time in April 1993."
Robertson's background is in financial services and technology.
"Before then, I ran the consulting practice for Coopers and Lybrand for a long time. I've always been involved in technology, writing big systems for insurance companies over the years, so I suppose technology and the use of technology to support business was my consulting practice."
Use of IT
Today, the Discovery Group employs hundreds of IT staff.
"On the infrastructure side, there are about 120 people running the international data centres, the support and helpdesks, voice, data and security," says Robertson. "Then there are 135 people running the desktops and then, if I add up all the CIOs and their teams - the business and systems analysts, developers and testers - you're looking at another 500 people."
At the top of the pyramid is Robertson, who has to look after overall IT strategy and execution for the group, as well as the governance and compliance sides.
"Discovery has a number of businesses. Each business has its own CIO and they're responsible for the applications architecture and the systems delivery pieces. They report directly to the CEO of each of those businesses. So they're highly focused on those individual businesses. Each of those CIOs has a dotted line to me and through what we call the IT Strategy Forum. Any new project or initiative - whether it's mobile, data, voice, new applications - are debated in that forum."
He says technology is as standardised as possible at the lower levels.
"We try to get as much standardisation across the board as we can. There are some obvious ones: everyone uses Oracle as a database. To think of changing the database strategy isn't going to fly very far with me. But further up the stack, we try to get as much reuse, standardisation and collaboration between all of the CIOs as we can and try to have at least one centre of excellence rather than each one doing his own thing. Is it perfect? No, but our style is not to mandate stuff but rather to try to demonstrate the value you get from using the corporate applications. For example, everyone uses the group commission system."
The demands on Discovery's IT can be considerable. CEO Adrian Gore requires that each division reinvent itself each year and come up with a new product, sometimes more than one. Robertson says he doesn't find the innovation to support these demands in the plumbing.
"The real differentiators are not in the low levels of the stack. We're very good at running big networks, big databases, big data centres, and those support the business, but they don't allow the business to innovate. It's in the application architecture places that true innovation happens. The problem is to manage the demand for innovation across the group. If you take mobile, for example, there's a massive pent-up demand to roll out mobile apps. But that's a poor description in my view because it's really a channel application. It can be run on the Web, an iPad, a BlackBerry or whatever. But the demand is across the group so you need to create a set of standards and mechanisms that allows people to use to leverage off each other."
Discovery 2.0
To manage this pent-up demand, Robertson chairs two forums, a mobile and channel one where the demand is managed and he can get commitments from the back-end people to deliver what the applications need.
"The other one is the group prioritisation forum that is attended by every COO and every CIO, and the objective is to manage the conflicting demands for innovation. Innovation always needs support from the basic source systems."
Innovation always needs support from the basic source systems.
John Robertson, Discovery
Internal company forums on how to innovate have been held on the actuarial side for a few years already, says Robertson, but on the suggestion of the board, he decided to hold an IT-specific one this year.
"All the Discovery actuaries develop papers just as they would for the actuarial society every year and those papers have to be innovative, support business and provide insight. Each team gets to present their papers and they get questioned and voted on by their peers and the winner gets a prize. But most importantly, the winning ideas must be rolled out. One of the ideas was a predictive modelling system that could interpret events and predict when there would be a crisis. Adrian asked me to come up with a similar conference for the IT side. So I modelled it exactly on the actuarial one and we set up the rules: innovative, business-changing, implementable and with ROI. We got 51 submissions and trawled through them all and eventually came up with nine finalists. They ranged from how to get our SOA journey more leveraged, to an incredibly good paper on how we could formalise and improve our testing strategy. The winners were a team that proposed how we could take what we have and turn it into a Web 2.0 experience, including everything to do with mobile. They will be travelling to Silicon Valley, San Francisco, and will be going to the Apple, Google and Facebook headquarters."
He predicts that the ROI on the conference, held in Magaliesberg earlier this year, will be incredible.
"If we just take two or three of the ideas and roll them out, the benefits in terms of customer experience and efficiencies will be enormous. In the conference, people were finding their way with their submissions because it was the first such one, so we have high hopes that the second one is even better."
First published in the August 2012 issue of ITWeb Brainstorm magazine.
Share