Transcript - Paul's speech at The Mandarin/IBM event Public Leadership in the Digital Era (September 2015)
Paul Shetler: I started out this thing with like a really obvious question, which is, you know ‘Why are we here? What is this digital thing anyways? Why is the DTO here?’ Because a lot of people have been asking that question. I’ve been hearing that quite a bit recently. And I guess the answer’s like really obvious, it’s because people are online. So all of us are online. Everybody in this room is online. Everybody in this room, I’m assuming, everybody in this room uses the Internet, has been using the Internet for quite a while, uses it to order services, uses it to consume services, because Internet is no longer just about the actual requesting of services, but about the consumption of them, right? Those things have actually kind of dissolved. They’ve dissolved that boundary. And people’s expectations between real life and Internet have merged. And expectations of what you can do are now increasingly set by what you can do over the Internet. And government, although we have an awful lot of things out there, we have, as Taavi pointed out, we have portals and all this kind of stuff, we haven’t really been keeping pace with the outside world. So that’s why DTO is here.
I just have like one simple fact. We had this - we did a study. We found out that, for any given four-week period, you’d have more than one in eight Australians who’d be trying to look at government information, or do something with government, something that only government can do for them, and over half of them had a problem. So over 50% of people had a problem when using online government services. And when people have a problem using online government services, typically what happens is, either they don’t get to use them, they don’t actually complete what they were trying to do, so if there were some kind of policy intent behind that, it becomes obviated because people can’t actually do what you want them to do, or they call a call centre, or perhaps they go to a drop-in centre, or something like that, and all of a sudden the cost of the transaction ratchets up. It goes very, very, very, very high, right? And of course, I think - people, then again, have to look at what happens in their everyday life, and say, ‘Why is this the case? When I order a car via Uber, do I need to print out a document and bring it into some office somewhere? When I order a book on Amazon, do I need to print out a PDF, send it to an Amazon agent, have them type in the data?’ No, right? If that happened, Uber and Amazon would be out of business.
And that’s the context that we operate in. So our simple view is that we have to do a lot better than that, must do a lot better. We can’t have a situation where over 50% of our users have difficulty consuming services which we’re offering. In the private sector, a company that did that would go out of business, just wouldn’t exist, right? Would be crushed by competition. So there’s lots of reasons for this. And one of them has to do with the simple fact of old brownfields - I mentioned earlier, you know, public servants are working really really hard. Everyone who works in the public service works for the public service for a reason. They believe in serving the people. It’s not an easy job. There’s lots of things you have to go through. So to be in the public service, you already have to have a motivation to do the right thing. That’s why we’re all here. But we’re often also hamstrung by antiquated IT - I think Taavi said thirteen-year-old systems, I think we can think about twenty-year-old systems, thirty-year-old systems and plus. Green screens, people reading from one green screen, writing maybe notes, typing that into another green screen, maybe printing out a document, passing that to somebody else, and then retyping that somewhere else. I know that sounds kind of crazy, but this is the kind of story that I’m hearing. And I believe them, because I’ve seen them in other places.
That kind of stuff just leads to just huge amounts of unhappy users, political pressures on the state, people saying, ‘Can’t government do anything right? Not anything? Can’t you guys do anything right? Does everything have to be outsourced to the private sector, because can’t government do anything right?’ How many times have you heard that? It’s in the press all the time. That’s what we’re up against, right? The steady drip-drip-drip of poor service, it destroys public trust. It just does. And that’s not a policy problem, it’s a delivery problem. It’s a problem with the way we actually organise our work, with the tools that we have, and the way we actually present ourselves out to users. And again, we’re not in a competitive marketplace, but we’re judged by the standards of service delivery that people expect in their everyday life. It’s compounded by one other thing, which is that, when people access government services, it’s usually for a relatively stressful moment in their life. It could be starting a business, and that can be something which is really exciting, but it’s also really stressful. It’s really stressful. Getting lasting power of attorney is one that I know really well, because we delivered that in the UK. Visiting somebody in prison. All these kind of - even paying taxes, which is mundane, you do it every year - still, these are not things that people seek out. They’re not like going on a holiday. It’s not like ordering a book, or a toy, or something like that. And at those moments when people are already stressed out, they have services which aren’t that great, where over 50% of the people say that they’re having real problems actually using them. So there’s an injury, and then there’s an additional insult sometimes.
So we must do a lot better than that. Love to get to the point where Taavi is at. We need to do a lot better than where we are right now. Now, our view is our opportunity’s huge. We have a relatively low bar, I guess you could say, to start with. But also, just because of the times we live in, right? So, compute, storage, many of the things which sort of kept people back because they were so expensive, because IT was so incredibly expensive, because it was so incredibly difficult to get things done in the first place because of procurement issues, are now dirt cheap. Cloud did that. We can now invest all of our effort, or much of our effort, a huge part of our effort, in redesigning services instead of paying for tin. That’s really awesome. That’s really amazing.
And I mentioned start-ups earlier. I think that there’s, you know, start-up and government, actually, those things - it might sound sort of oxymoronic, but when we’re trying to talk about digital and digital transformation of services, that is the way we need to start thinking about it. We need to think about radical service redesign, across all tiers of government. So federal, state, local, and across all channels, because digital is just a channel for a service, right? People typically also have to do other things. People find out about things through digital, people might transact through digital, people might also for whatever reason need to go into a shopfront, and it might not be because they can’t do something via digital, it might be because sometimes there are things that people do better in person, and maybe people should do people things, and machines should do machine things. The point is that people want to do things - need to do things - that only government can do. We haven’t reached the point where we’ve outsourced everything. And people also don’t want to carry a map of government around in their head, they don’t want to have to deal with all the complexity, they want things to be simple. Again, think of, you know, think of AirBNB, think of Uber, think of Amazon, think of what we normally use in our everyday lives and the kind of complexity - or simplicity - that they have, and then compare that to every single one of our services.
Often times, what I’ll hear is that, ‘You don’t understand, Paul. It’s really complicated. What we’re doing here is really, really complicated. It’s really - the scale is huge.’ And I think, ‘Have you guys ever worked in financial services? Do you really know what complicated is? Do you really know what scale is? The scale is small. The complication is often because we admire the complexity of the problem. Our job is to do the hard work and make things simple. So our goal is to have a radical redesign of the way that the government presents itself to the people, so that people don’t have to carry a map of government around in their heads, so that people when they want to do something don’t need to know, ‘Is it this particular agency in the Commonwealth, or is it this particular agency in the state, or is it this person here at the locality? How do I talk to them? In what order do I do the things?’ We want that radically simpler. That takes a lot of work.
And if we do that, and if we do do that, we’re going to have some very happy customers, or users. I used the ‘Customer’ word, I shouldn’t. We’re going to have some very happy users. And we’re also going to radically reduce a lot of the costs of delivering service, because many of the costs of delivering services are the costs of doing things poorly It is the cost of maintaing all that legacy, it is the cost of all that terrible integration. Now, I’m making a relatively bold statement here, saying that I think that we can become the best in the world at delivering public services, and I’ll still stand by that, even having heard what Estonia’s done. But it’s a huge job, and it requires all of us to work together. It’s not something that DTO is going to do by itself. This is something that the DTO will do working with agencies, departments, states and localities, start-ups and academic, and a broader ecosystem of service providers. The actual hard work is all going to be done in the agencies. DTO is not the start-up, DTO is the incubator, and we’re incubating talent and teams, in every single department and agency in Australia. That is our goal. So we talk about what that means in terms of becoming best at public services.
We mean really simple, really really simple. I like Taavi’s idea of doing something in less time than it takes to boil an egg. Actually, that’s a really good way of looking at it. But for that to happen, governments need to do, or teams need to do, or service teams need to do the hard work to really radically simplify what happens. That means going through policies, understanding, ‘Why are we asking for this? Why are we asking for this piece of information? Why are we putting somebody through this particular step? Why are - why do things take so long? And why do sometimes admire the complexity of a problem too much?’ So think about what the user’s trying to do from the standpoint of the user, and not from the standpoint of the agency who’s providing the service, because often times an agency is providing a service, is providing something which happens in the course of many interactions, right? The user journey may not be something which is only provided by Industry, or only provided by Tax, or only provided by Finance, but is part of a larger set of needs, perhaps a transition that the user is going through at that time, and think about it from that context. So we can undo the fragmentation and join up information and transactions.
Making things clearer. I had just this issue myself recently, because I’m a recent migrant to Australia. And going through some of the information that was on the immigration pages, for instance, was pretty difficult for me, because I didn’t understand it. And I’m a relatively educated guy, and I’ve been working in digital for quite a while - a long time, and it wasn’t easy for me. Our Head of User Research had similar problems. If you look at Tax, if you look at pretty much any website, any service that’s offered by the Australian Government, plain and simple English probably isn’t going to be featured very highly on that. But people need that to get the basic guidance on what they need to do, so government can help them. Again, you don’t need a lawyer to figure out what to do next on Amazon. It’s not that complex. It’s really, really, really simple. Because they’re goal is to get you to do something, is to allow you to do what it is you want to do. It’s not to expose the complexity of the problem, it’s to radically simplify it.
Faster. I think by simplifying things, making things clearer, we can make things faster. We might not get to the point of being able to do it as fast as an egg, but registering to vote takes all of three minutes now in the UK, and that used to be a very complex process, and it required all kinds of documentation to be sent in, and scanned, and sent in, and after living there for twelve years, I still hadn’t registered to vote, because it was too complicated. It wasn’t worth it! Why would I want to do that? To spend a couple of hours, to figure out this stuff? When it became something that I could do in the space of all of three minutes, and I am quite serious, it was all of three minutes, then I registered, and I voted for the very first time, this year. And it was pretty amazing. I was quite happy that I did. And I think that if you looked at the stats, when we introduced the new digital service, and you saw several things happen. First off, there was an inversion of tendency in terms of who registers to vote and who doesn’t. Normally, it’s older cohorts, 30 and above, who register to vote; typically, younger people don’t. That was at least the case in the UK. What wound up happening was a complete reversion, and a huge bump in the number of younger users. Pretty much the same for older. We had a huge bump of younger users. We were able to embed the service inside people’s Twitter streams, we were able to embed the service inside people’s Facebook streams, and it just reached a huge number of people. And it was brilliant. And that’s the kind of thing that making very simple, beautifully-designed digital services allows you to do if you’re government.
And then it gets back to more humane. So we’ve been saying ‘Simpler, clearer, faster and more humane public services,’ because it’s not all about saving money, it’s not all about efficiency, it’s also about the fact that people, when they use government services, are using government services because something has happened in their life which government needs to help them with. It can be - again, it can be, you know, somebody’s mother who is becoming - who is getting dementia. It can be somebody who is in prison, it can be somebody who just needs to do tax. But, in any case, it’s going to be stressful. Our job as public servants is to make it as simple as possible for them. We don’t have customers. We have users. People can’t choose to use government services, they can’t say, ‘Oh well, you know what? I’m just going to go use Amazon instead.’ That choice doesn’t exist. So we have an ethical obligation to make things as simple, as clear, as fast and as humane as possible, because there is no marketplace for our services. It is an ethical obligation.
So we get there by putting users first, considering their needs, not our needs, user needs, not government needs. We get there by thinking big, having a big vision for where we want to take things, but realising that it’s not going to happen overnight. We can’t do a ‘Rip and replace,’ we can’t take out every single legacy system we have and just throw them in a bin. It’s not going to work. It’s not going to work, guys. Just - we’re brownfield. If we did that, we’d shut down for about a year. We can’t have that.
So how do we do transformation, how do we transform our services, the face of our services, radically, and at the same time actually do it in human timeframes? Well, we do that by looking at the actual service itself. So this is the case of one that I was responsible for in the United Kingdom, this was civil claims. Civil claims are something that - well, they’re basically when somebody has - somebody doesn’t like somebody else, and they want to take somebody to court, whether that’s for getting somebody out of their flat, whether somebody owes them money, or whatever. It’s not a criminal claim, it’s a civil claim. It’s a fairly complicated process. Now we were told by GDS, they said, ‘Hey, guys, we want end-to-end service transformation.’ And we’re a brand new digital team, relatively new. We had about 80 people at the time. Pretty much every single person in the team had come in from the outside. So they were all digital professionals, but they were digital professionals who were used to working for very small companies, and they were people who hadn’t yet sort of experienced the octopus of big government IT, and how it kind of pulls you in. And so we were looking at this thing, and we were hearing GDS saying, ‘Oh, gosh! Fix everything, you know, do the whole thing. And by the way, you have 400 days to do it.’ And we’re looking at it and thinking, ‘Wow, there’s an awful lot of entry points, a lot of exit points, a lot of forms, a lot of processes, a lot of loops,’ and saying, ‘How are we going to do this? We don’t know where to start, and we don’t know when we’ll actually really be done. It’s a really complicated process. And this [gesturing to slides] isn’t all of it, by the way, this is just a few. I think it’s actually just four examples.
And so we realised that that kind of overall ‘Get it all done,’ sort of smacking very much of Waterfall, smacking very much of multi-year transformation programmes where the people who have started them are never the ones who end them, and perhaps they never do end, and perhaps they cost a lot of money. So we said, ‘We can’t do that. If we’re actually going to be transformative, if we’re really going to be transformative, we’re going to bin that whole idea of ripping and replacing everything, and we’re going to look at, ‘What are the actual problems? Where are the actual pain points? Where do people actually have an issue, and can we solve that within a human timeframe? Rather than, again, looking at so much to do in several years, can we do something which we can knock out in, say, 20 weeks, where we said 20 weeks would include a Discovery period, where we would understand the context of the problem, it would include building an Alpha, so we had a working prototype, live code, a Beta that is open to the public, and then a certification from GDS. Three stage gates along the way. Can we do that?’’ We said, ‘Of course we can. It just takes some discipline. It just takes an ability to focus on, ‘Where can we find actual concrete value for our users, and deliver it quickly, and not allow ourselves to be sucked into scope creep, and focus very heavily on that?’
And I’m giving this sort of slide here today, I’ve given it before, but today I’m giving it for a very particular reason. And that’s because we’ve identified a number of partners who we’re working with to start delivering similar kinds of exemplars here in Australia. And for it to work, I think some of the lessons that we learned from the UK have to be applied. And one of them is that. Think big, have a transformative vision, know that you’re going to get there by slices. And be ruthless - absolutely ruthless - about delivering something in a tight timeline. That is how we cut costs, that is how we focus on users, that is how we gain trust, that is how we start transforming the public service more broadly. Because we can talk about Agile, and we can talk about culture change and all this and that. The way you do it is actually through delivery. The people who are responsible for large transformative projects realise that they have more visibility, more control, when we use an Agile and Lean approach, they’ve adopt it, and they become evangelists. And we started seeing that. So one of my favourite stories, because it was one of my favourite experiences in the UK, was working with the Office of the Public Guardian. And the Head of the Office of the Public Guardian was a guy named Alan Eccles, and he’s an old, old-school civil servant. He has been there for years, and he’s absolutely dedicated to his users. He believes very strongly in the mission of his organisation. And he also had been imbued with all of the sort of programme management, PRINCE2, heavy governance, Waterfall-y type stuff that people had to do in the UK, and was comfortable with all that, but he also knew that perhaps we could do things better, perhaps we could try this other approach and see what we could get out of it. And so he took a risk, and he did that. And now, if you go onto YouTube, and you type in ‘Alan Eccles Agile,’ you can see a video he’s done where he starts explaining - and keep in mind who this man is, he is the Public Guardian, he is not an IT nerd - he starts explaining for people how Scrum works, how Agile works, what a Minimum Viable Product is, and why it’s so good to work that way, because now he has a control he never had, he has a visibility he never had. He knows what’s actually going on, and he can make sure that if they’re going in the right direction, they can fix that before it costs a ton of money. Because he cares about his users. You want to deliver a brilliant service to them. He doesn’t want to - he doesn’t want to waste.
I think that gets us to the next point: iterate wildly, right? So we delivered several MVPs, and if we were doing it in sort of the standard IT kind of way, we would’ve said, ‘Oh, well, we’ve delivered your product, time for change controls’. But we didn’t. So the expectation we set from day one was that, ‘Yeah, we’re going to deliver something, and it’s going to be small, it’s not going to be the complete solution, it can’t be the solution, because we don’t even know what the complete solution would be, because that would assume that we have an infinite amount of knowledge, which we don’t have. But what we will do is, we will deliver something really useful, and we will see how people use it, and we’ll continue doing our user research, and we’ll understand the problem better, and as we understand the problem better, we will continue to enhance the service. And we’ll make it better, and better, and we will start adding more and more functionality, until we have met the user needs for that particular thing. And the fact that we were able to do that was why Alan became one of our biggest proponents. When you actually don’t walk away from your products, when you don’t walk away from your projects, when you don’t say, ‘Oops, change control, kaching,’ then you start gaining trust, and that’s really really important. And iterating, learning from the users, understanding that when we start the project we’re not this omniscient deity that knows everything, but we have a lot to learn, is extremely important.
I guess the last thing I’d say is - so we’re part of a worldwide movement. If you think about what’s happening in government and digital right now, it’s a wave. The UK, obviously, with GDS, but not only GDS, I think it’s really important to keep that in mind. The vast majority of people working in digital in the UK are not in GDS. They’re in HMRC, the equivalent of Tax, they’re in Home Office, the equivalent of Immigration, they’re in Ministry of Justice, which would be the equivalent of agencies who are in the states over here, they’re in DWP, which is the equivalent of DHS, they’re in Business, Innovation and Skills, which is the equivalent of Industry. And those Departments have all built massive digital teams - not massive, a couple of hundred people each - who are delivering pretty big portfolios of service transformation, and are doing it rapidly. And GDS is sitting in the middle, and has a view from the middle and can join up, and say, ‘Well, actually, this user here actually uses this thing over here. And maybe, when we’re talking about - maybe we should look at things that go across borders, rather than just people and just trade, for instance.’ I think that’s an example that people in DIBP are looking at just now, and that’s brilliant, that kind of holistic view. That’s what GDS’ role is. The major work, the real work, the hard work, is all being done in the agencies and departments, and that’s a lesson which we want to bring over here. Again, DTO is not the start-up; DTO is the incubator for all Australian teams that are already in departments and agencies, and that will be growing up in departments and agencies as a result of some of the exemplars we’ll be doing with them. There’s also the United States with 18F, the USDS, there’s New Zealand, ‘All services will be digital by default,’ there’s Singapore, there’s Israel, there’s South Korea, and there’s the shining example of Estonia. I think actually Australia has a chance to lead in this. We have a Prime Minister who’s absolutely dedicated to this - not only a Prime Minister, but other Ministers. I think you might have heard about one of the Ministers right now just talking about doing a ‘Policy Hack,’ a ‘Policy Hack Day’. And when I first came in, I recommended it as a thing I had a conversation with David Hazlehurst actually, we thought, ‘Well, actually, that might be a really good idea, let’s talk about doing that, but, oh, it might be a bit radical’. But now we have a Minister talking about that in the space of two months. I think that’s a sign of just how far Australia’s already going. I think it’s just an incredibly optimistic and brilliant time to be here right now, and I’m really glad to be a part of it.
So I guess the last message I want to give out here is, ‘DTO is doing this with you, not to you’. We want to work with every department and agency, we want to work with the start-up community, with SMEs and with other suppliers and vendors, because we know we can’t do it all ourselves. This is going to have to be a team effort. Digital Transformation is definitely a team sport. We’ve all got to be in this together.