Design gurus: turn policy development upside-down to solve tough challenges

By Stephen Easton

June 22, 2017

Sydney, Australia – May 12, 2014: Alternative and creative view of Sydney City

Interest in applying design thinking to public services is leading to a popular belief that traditional top-down models need to be flipped, so that experience and evidence from the delivery end plays a bigger and more ongoing role in policy development.

Some design gurus even believe some federal departments should rethink their role along similar lines, allowing learnings from design-led approaches in the state and local sectors to filter up to give federal policy a stronger evidence base.

“We need to be inverting government,” argued public sector consultant Dominic Campbell at the recent Explore Design conference, an APS initiative hosted in its second year by the Department of Employment.

“We need to be seeing Canberra as the ‘force multiplier’ for communities, state and local governments, not as the place that tells other people what to do.

“You guys need to be at the bottom of the triangle, not the top of the triangle. Because ultimately your role is to codify, and then scale, it isn’t to have a big guess and then tell everyone else what your great guess was and how they should do it.”

The impressive Central Coast Multi-Agency Response Centre developed by the New South Wales Department of Family and Community Services was later presented to demonstrate how design-first approaches can lead to policy improvements, instead of policy coming down the pipeline first, and various speakers described other examples as well.

But the view that design thinking often clashes with traditional linear policy development processes was never far away, and featured prominently in a panel discussion on policy design later in the day.

Martin Stewart-Weeks suggested that “patterns of power, control and authority” might need to change for design thinking to be widely incorporated in public policy, noting that the possibility of an inherent mismatch or tension between the two is discussed in the 2014 book Design for Policy by Danish author Christian Bason (in video below).

He suggested examples that bucked this trend — like another NSW FACS project, Childstory — did so through the willingness of leaders like secretary Michael Coutts-Trotter to allow user-centred design to take the lead.

Another panellist, Peter Overy, commented that public sector agencies often call on design in hopes of simply improving the implementation of a set policy, but in the process other, sometimes unwanted, questions emerge about the deeper assumptions behind the policy.

Upside-down policy development

In the plenary session Nicholas Gruen made similar points to Campbell, beginning his presentation by illustrating the traditional policy process with pictures of trees, river deltas, lungs and the like — to show policy normally flows down a central trunk, out to where delivery takes place in the small branches of the system. The model can be inverted so that “learning goes upward” from the delivery end to policy, he said.

This normal top-down process works well enough in some cases. The first speaker, John Body, explained that decision-making tools like the Cynefin framework can help categorise different issues (Wikipedia uses the known knowns, known unknowns, and unknown unknowns made famous by Donald Rumsfeld to explain this).

Gruen suggested that some policy problems only require a “thin” understanding or explanation of the problem at hand, whereas others require a “thick description”.

Policy solutions to thin issues are simple; they can be developed in an ivory tower and handed down, like the HECS scheme, or happen at the “stroke of a pen” like lots of simple deregulation decisions. But Gruen thinks most problems don’t work like that, they require a “thick” understanding, gained by observing human behaviour in context.

“Ideologies are thin; issues are thick,” he said, adding that politicians repeatedly fall back on the idea that “a thin idea might solve a thick problem” and find their blind faith cannot provide good answers to complex challenges.

“The cult of announceables is part of the problem.”

“The cult of announceables is part of the problem,” Gruen added.

He also noted two related issues that have been recognised by public service leaders like the head of the APS Martin Parkinson and one of his predecessors, Peter Shergold: policy is seen as more important and more prestigious work than delivering results through implementation, but at the same time it is developed from a weak evidence base.

In the design thinking approach, the two can theoretically be connected through iterative processes, with ongoing experimentation and research at the delivery end to find useful “causal” data, and then scaling up what works.

Body also quoted public service gurus Terry Moran and Gary Banks on the increasing failure of a lot of policy developed through the standard process. “The conclusion? The current policy design process is flawed, and a new model is required,” he said.

Campbell suggested federal policy was often something like educated guesswork, “where people think that they know what’s right for the people of Australia” but are, to some extent, crossing their fingers and hoping for the best.

“What we need to be doing is thinking with the system in mind and the larger change, but actually — instead of starting big, and eventually going to implementation and crossing your fingers and hoping it worked, when it’s been designed in isolation in Canberra — actually thinking about, what do ten people need?

“Does it work for ten people? Great. Does it work for a hundred people? Great. Does it work for a whole council area? Or a whole state?”

Out with the old

A design conference wouldn’t be complete if the old ways weren’t castigated with the same revolutionary zeal as the almost mystical power of design is promoted.

Campbell is about as critical of traditional approaches to reform — in the United Kingdom, where he previously worked in government — as he is enthusiastic about design. He claimed working in local government was his dream job but he had been frustrated by efforts to transform human services using “inhuman technology and change processes” that relied too much on inadequate off-the-shelf computer software, or “mega-systems that no human on earth could intuitively use”.

He said all governments regardless of political system faced similar problems and encouraged the public service delegates to aim higher, arguing there was “massive gap” between citizen expectations and the aspirations of government the world over.

“Our aspiration is rubbish, and I think we all need to get off our asses and try a bit harder and aspire a bit higher,” he said. “Because I think even when we’re at our best in government, we’re not doing what citizens want and their aspirations of us, and expectations of us, are ten times higher than we actually have for ourselves, and I think it’s incumbent on all of us to actually raise our game a bit and find ways through all of this.”

Campbell said design thinking could improve overly “technocratic” methods of policy development that rely heavily on the expertise of public servants rather than the input of stakeholders. He thinks the methodology is being helpfully carried within the “Trojan horse” of digital transformation, but feels there is still too much technophobia.

“God only knows how in 2017, people in government think it’s OK to go, ‘Oh I’m not a techie so I couldn’t possibly comment,’ or, ‘That’s not my thing, I’ll leave that to the head of IT.'”

Some of the UK Government Digital Service’s work that is seen as “leading globally” and inspired Australia’s Digital Transformation Agency has done little more than create a shiny new facade, he added.

“A lot of it is really just about spoofing a government that we wish existed,” said Campbell. “It isn’t actually about building a government that we want and that needs to exist.

“It’s about lipstick on pig, or it’s about putting shiny wrapper around the complexity, cost and brokenness of our government, and just hoping it will go away if we kind of sweep everything under the sofa.”

Later, however, when DTA’s Lisa Reichheldt brought up some of the challenges of working within current structural realities in a follow-up question, Campbell said he had recently accepted the pig was better off with the lipstick than without.

“I think my even less realistic self a few years ago would have said it causes more problems than it solves. My more pragmatic self two years on says that it’s definitely better than not doing it. Absolutely.

“The danger that I think I’m alluding to is that we do it and then it just freezes the system underneath it, because people stop [and wonder] why do we need organisational reform?”

It is difficult to redesign services and products without rethinking organisational design at the same time, he contended.

Campbell emphasised that genuine “empathy” is the key goal of user-centred design, which means talking to lots of real people and observing real life. And it needs to be an active process; design thinking needs to lead to a fair bit of design doing.

“It isn’t a thing that just allows you to sit in an office and stroke your chin.”

Campbell promoted multi-disciplinary teams because “you can’t make policy in isolation” and advised that “agile policy is more important than agile tech, so don’t let ‘agile’ as a concept just get stuck in tech-land” — repeating his earlier point that acting like digital technology is a mystery known only to the tech people isn’t good enough.

“Strategy still matters,” he added. Tipping the policy process upside down doesn’t lessen the need for leadership; user-centred design will not reveal eternal truths that light a single path forward.

“It gives you, like, eleven choices,” said Campbell. “You still need leaders and you still need a vision and you still need a strategy. It’s just about strategy as a direction of travel, not strategy as a fixed plan for doing something.”

About the author
1 Comment
Newest Most Voted
Inline Feedbacks
View all comments
5 years ago

You need both, neither in exclusion. But most of all, we need to actually have the people for whom the policy is intended to affect having a voice early on. They do not need to be a representative sample, they do not need to be part of a massive set of focus groups. They just need to be respectfully listened to and be allowed to cut through our privileged arrogance. Because when we are at the top of the tree doing policy we are so removed from reality that we miss a lot of simplicity and promote a lot of absolute tosh, often in ignorance of what the person we are expeding public money for actually needs. If policy outcome was a marketable product, most of Government and most major consultancies would be in receivership by now. Oh, and we need to understand that a solution that will work in an inner suburb may not work in an outer suburb and is almost guaranteed not to work at all in regional Australia outside of major cities and towns…

The essential resource for effective
public sector professionals