A few people scratched their heads last year after hearing the Australian Bureau of Statistics won a public sector innovation award for running the same-sex marriage postal survey; after all, how hard can it be to run a mail-out?
The manager of the project, in a panel discussion this week with other public servants involved in last year’s winning projects, said the question had been put to him directly and ventured to answer it.
Of course the innovation awards are for administration, not policy; it was the successful delivery of the project that earned the ABS a “citizen-centred innovation” award. The judges praised the project for user-centred design as well as efforts to ensure equity of access and sufficient public confidence to get a high response rate.
“This exercise was actually an exercise in public trust,” said Duncan Young, observing “it could easily have been a dismal failure” if too few votes had been cast, damaging the poll’s credibility.
That it went smoothly was a relief after the ill-fated 2016 Census, which he also managed, Young said at the launch of this year’s awards, hosted by the Institute of Public Administration Australia in Canberra.
When tasked with rapidly getting the marriage law survey off the ground, he worried it might suffer from a loss of public trust in the ABS caused by the issues with the Census, which also faced calls for a boycott over privacy concerns.
A few years on, he is able to laugh about it. He opened with a gag about a newspaper headline playing on the decision to run the mail survey—“PM to go postal”—that appeared on August 9.
“Which was sort of a bit ironic because it was exactly 12 months earlier that I stuffed up the Census,” he quipped, noting the line also neatly described Malcolm Turnbull’s reaction: heads would roll, he promised in the heat of the moment. Another headline made sure readers were aware it was the “Census bunglers” running the survey.
Young said the marriage-law survey involved a “genuine co-design” process, in contrast to a Clayton’s consultation process he candidly admitted was common in the Commonwealth bureaucracy.
“I’ve done Canberra co-design before,” he said.
“Canberra co-design, as I rudely put it, is we come up with a solution, then we go out and consult on it, and then convince the people that they’re wrong and go forward with our solution and call it co-design.”
“We didn’t have the liberty, with this, of enough time to come up with our own solution, so I actually called some of the key stakeholder groups on August 10 and said, ‘Hey we’ve got a problem here, we’ve got this survey to run. I don’t know how it’s going to work in nursing homes. Can you help me out with that? What’s the best way to do this?’
“And it really changed that relationship in a way that we actually ended up with a better experience.”
If you drop the ball, pick it up and hold on tighter
Performance measurement involved asking individuals for their views on the process as well as their perceptions of other people’s experiences, and of the government’s ability to handle complaints and solve issues like replacement-form requests. Running the project involved collaboration between 15 agencies.
Young accepts the ABS “dropped the ball with customer service” for the Census, which he also managed, and were only able to answer one million of 3.5 million phone calls.
It did much better with the marriage law survey. Human Services ran the phone line, which had an average speed of answer of 12 seconds by the end. (The average waiting time was over half an hour for some of the other DHS phone lines in 2017-18, according to a recent audit report.)
“One of the things that we focused on in the marriage law survey was how to make sure that there weren’t bottlenecks in our service flow, and so we really focused on how to automate the process of requesting a new survey form … or an online access code if you needed one,” said Young.
“And to get a system in place and up and running within about 30 days out of a 99-day period, was pretty remarkable and what we saw was ultimately, 95% of all of the requests for a replacement form were done automatically, without any manual clerical intervention.”
The innovation award judges were also impressed by the steps taken to allow everyone to easily participate, including people in the most remote parts of the country and overseas at the time, which meant running an online polling system as well as the postal vote.
As well as the rapid time frame, the immense scale of the project was a challenge. This is a common factor in government programs that often doesn’t occur to people outside the public service, as observed later by the Digital Transformation Agency’s Anthony Vlasic, discussing a new channel for IT vendors to pitch innovative solutions.
Young said his go-to guy to source the materials wasn’t sure there would even be enough paper available in the country at first. Put end-to-end, the forms and envelopes would reach halfway around the world, he added.
The short turnaround time required by the PM with the spectre of the Census website problems in the background led the ABS to get creative with procurement, too.
“There are allowances under our procurement guidelines which say, this is an emergency, you can go out and directly source, but coming from the Census experience where we had directly sourced with IBM and been widely, and reasonably, criticised for that, we felt we were between a bit of a rock and hard place there,” the program manager said in response to an audience question.
There were two realistic suppliers for one key capability requirement, according to Young, so the ABS gave both the chance to tender and asked if 24 hours was long enough to get their bids together. Both were thrilled to get a chance at the contract without the usual expense, and the agency knocked up its statement of requirements in about 18 hours.
“We brought together a panel with a number of representatives from different agencies, so again insulating ourselves against [potential criticism of the process], made a call in two days then I got on a plane, went down and did the gritty negotiation stuff and signed them on as a partner,” he added. “The losing party never complained once about the process.”
Anything that can go wrong…
Murphy’s law doesn’t get invoked these days as much as it once did, but Young told fellow public servants they should always expect things to go wrong with big projects. “And your ability to be able to handle those incidents really quickly is critical to maintaining and building that public trust.”
“We had to address a bit of a cultural resistance within the organisation,” he said, referring to the kind of relentless optimism that can lead to inadequate contingency plans and unrealistic expectations, like the idea that sceptics and critics will simply accept a government spokesperson’s assurances that everything is actually fine and they have nothing to worry about.
“So some of the things we did was try and insulate ourselves a little bit. With Census, we ended up in a difficult position where people weren’t 100% trusting the ABS—maybe that’s an understatement—and during that period … our only solution to that problem was to put out an ABS spokesperson, usually me, to say, ‘Well, you can trust us.’
“Well, if your issue is trust with the ABS, then you don’t believe an ABS spokesperson.”
“So one of the innovative things we did was [think about] what are the areas where we might get criticised and run into issues, how do we get some external spokespeople, get them embedded in the team, and get them to make some independent statements.”
That strategy led the team to get Malcolm Crompton, former federal privacy commissioner and principal of Information Integrity Solutions, to audit the privacy aspects of the process and make an independent public statement. In the end, there was high confidence in the integrity of the voluntary process.
How do you sell a new idea to senior execs?
For the ABS team, Young noted, the innovation that earned the award was largely driven by necessity, but that’s not usually the case.
Later, the other panelists—Sarah Pearson, the head of the Department of Foreign Affairs and Trade’s InnovationXchange, and Rajesh Walton, innovation director at AUSTRAC—addressed the question of getting senior executives on board with bold new ideas.
“They’re the people who have to face up to Senate estimates if something goes wrong – how do you convince those people that your crazy idea is worth giving a shot?” asked Nick Ellis, of the Public Sector Innovation Network, part of the Department of Industry, Innovation and Science.
Walton said the key was to prove the idea would produce solid results. “And that doesn’t have to be a multimillion-dollar project, it doesn’t have to be something that you get [new policy proposal] funding for, it doesn’t have to be something that a team’s created to actually do, you just have to pick a problem and find a way to solve it.”
“We hear so much talk about ideas, [idea management] platforms and ‘we need more ideas’ and ‘we need to triage them and make them something’—nah, there’s billions of problems out there that are unsolved.”
Pearson agreed, adding that in her view, “it can be a good idea to do something under the radar” at first, rather than pitch an unproven concept.
“A cheekier answer is find out who influences those people,” she said, relating an anecdote from her time as “innovation champion” with Cadbury, a large multinational that was set in its successful ways after 100 years in operation and not particularly interested in thinking outside the Roses box.
“The chief technology officer was playing golf with his mates from Mars and other competitors and they’re all talking about open innovation, so a little bit of healthy competition can help people to see that perhaps they should be doing it too.”
Images: RLDI / IPAA ACT