Canada’s CIO talks tech disruption in the public sector
Alex Benay, the Chief Information Officer for the Government of Canada, has big plans to disrupt the way the federal government buys and uses technology. With an eye on the present government's focus on openness, Benay plans to accelerate procurement, make more use of smaller suppliers and cloud computing, and convince government organizations to design processes for users.
Benay is drawing on his diverse career, which has involved running the Canada Science and Technology Museums Corporation - where he made most of its data open by default - and working as the vice-president of enterprise information management at OpenText.
Q. How does your experience working in the private sector inform your new role as Chief Information Officer?
If you're in the public sector, you often hear that the private sector does it this way, and sometimes in the private sector you'll hear that government or public sector does it that way. I don't think either realize each other's realities very well. There is a big difference: in the public sector you're spending the public's money while in the private sector it's about the bottom-line. I'd like to think being a hybrid, in my case, is a good thing, because I've lived both at senior and junior levels.
There is no set way of managing and operating - you can take the best from both worlds. For example, the hierarchy in the public sector is not always conducive to innovation: people tend to follow the most senior person in the room, because that is how hierarchical organizations work. Whereas in the private sector, especially the tech space, you put all your bright minds in a room together, let them argue and debate, and the best idea becomes the product.
Hierarchy in government is required - we're dealing with people's lives, and if we get things wrong, it can be really bad. The right approach is to find the balance between how much innovation you need, and enabling the right environment for that innovation. We’ve tried to flatten the Chief Information Office branch of the Treasury Board, and we’ve got a group called Braintrust where you leave your level and hierarchy at the door. It’s all about the best ideas and trying to scale the concept. It's an idea from the [computer animation] company Pixar.
Q. What benefits do you see from public service employees being `open by default', publishing most information openly as it is created?
`Open by default' isn't for everything - you can't do trade negotiations out in the open, for instance. But if you're doing science or heritage, working in the open means you get to use the power of the crowd, amplify your work, develop new partnerships and probably deliver more for less. We have four or five departments that are starting to produce content open by default to the public.
At the Museum of Science and Tech, my previous employer, we believed we were the world's first public sector institution that was fully open by default. About 80-85 percent of the content was made available to the public within two hours of being created, with areas like human resources, procurement and legal matters making up the rest. People created video game apps from the museum's content - we would never have had that reach as an institution if we had not opened it up.
Q. How do you overcome fears of such openness?
You have a conversation, educating people that the data we are working with is actually the public's information, not our information as public servants. You also build it in as seamlessly as possible into the work. We used an enterprise-wide deployed solution for the federal government, and through business rules automated the process as much as possible. People started to realize it's easier to work this way than trying to decide what to release on a case-by-case basis. The percentage of what should not be released is much lower, so you're flipping the equation on its head.
Q. How will Canada's plans to move to agile procurement help smaller IT suppliers to sell to the government sector, and what are you doing to encourage them?
Traditionally, most countries procure by spending one to three years defining requirements by listening to the big vendors who pitch based on their old products. Then the requirements and RFP (request for proposal) are written, which takes a year or two, and it may take one to five years to deploy - so the current procurement process can take a decade. The private sector is used to replying to our RFPs in a certain way, and we're used to writing them in a certain way, so we perpetuate this situation.
The alternative is that we say `here's a problem, here's what I'd like to do, please fix it'. It may not be a tech company that answers - instead it could be a research institution or another country. There are things you can't break like security policy, laws and privacy, but if I work at defining my problem statement better, as opposed to spending three years writing requirements, I can cut my time to deployment pretty drastically.
We've just done a two-month pilot of this, on the issue of how we go open by default. We put our problem out, had 11 bids from across Canada, carried out a Dragon's Den style competition [a TV program titled Shark Tank in some countries] and ended with four very different solutions to the problem. We picked the best solution and issued a contract on the spot. You can scale that model to bigger challenges across the system.
Q. You've previously said that governments' adoption of cloud systems could be slowed by the legacy of large IT departments. How do you plan to tackle this?
Cloud seems to be a very emotional conversation for us; it represents a change, whether it's public or private cloud. We estimate that the Canadian government has 8,600 applications running across the [Canadian federal] government with multiple versions. That's a lot. You don't move a mess from one environment to the next - you have to do some streamlining and cleaning, and we're looking at that as we speak.
Taking everything we have and moving it to the cloud is not a good strategy, but doing nothing is also not a good strategy. We've reinstated our practice of enterprise architecture, which we had lost for many years. This means we ask about the business strategy, the business architecture, the information and data standards, what applications are running what and then we talk about infrastructure. Until we do this for every single new IT project launched in the Government of Canada, it will be very hard to have a proper cloud strategy.
Q. Will the Canadian Digital Service draw on the experience of similar units in other countries, such as the UK's Government Digital Service and the US Digital Service, and why have you said that the CDS could eventually become irrelevant?
Our CDS is lucky in that it focuses exclusively on user design, user-centric testing and bringing the empathy back to designing solutions, which frankly is something we had lost. Although it differs from other countries' digital services, CDS will use lessons from elsewhere.
My statement saying `hopefully one day they [digital service agencies] will become irrelevant' is because every single department should work this way. This way of working should be embedded in how we do digital government, whether that's in year one or year ten - although I hope it's not more than year five. If CDS only continues as a standalone organization, as with GDS or the US Digital Service, their values will not become part of the DNA of the institutions it serves, and we'll always need a CDS. If that is the case, then frankly, the experiment will have failed.
From 2011 to 2014, Alex was Vice-President of Government Affairs and Business Development at OpenText, an enterprise information management company. He has played a leadership role in Canada's digital industry as well as in promoting the global shift to digital in organizations such as the G20, the Commonwealth Secretariat and the Olympics. Before joining OpenText, he managed various teams and programs at the Canadian International Development Agency, Foreign Affairs and International Trade Canada, Natural Resources Canada, and Library and Archives Canada.
*The views and opinions expressed herein are those of the interviewee and do not necessarily represent the views and opinions of KPMG International or any KPMG member firm.