Improving schools: Reflections from education leaders in South Australia

| Interview

In the course of 2018, the South Australia Department for Education embarked on a unique school-improvement program in which it leveraged student and school data to develop differentiated improvement strategies, created a robust improvement-planning tool kit for schools, and used best-practice change management for communication and to set high expectations. The program was led by Department for Education chief executive Rick Persse, who was just under two years into his role. Alongside him were Anne Millard, the department’s executive director of partnerships, schools, and preschools, and Ben Temperly, the department’s executive director of system performance. McKinsey’s Marieke D’Cruz, Seckin Ungur, and Bart Woord met with the three executives to discuss the program’s journey and ten-year strategy. The following is an edited version of their conversation.

McKinsey: What was the impetus for embarking on the reform strategy?

Rick Persse: We could point to NAPLAN1 data, which showed that South Australia was losing touch with the national average at a worrying rate. In some ways, the state had almost convinced itself that this was acceptable because of socioeconomic factors. I do believe there is a relationship, but we were convinced that, with a concerted effort and clear strategy, we could do a lot better.

McKinsey: How did you go about setting up for the reform?

Rick Persse: We began with a data-analysis phase, which supported our hypothesis that we had a very patchy system. In addition, systemwide interventions largely hadn’t differentiated enough between schools to bring about effective change. So that triggered a series of conversations with every principal about where schools were in terms of their improvement journeys. These were pivotal for “baselining,” so we could then turn discussions toward how to improve.

Our data analysis showed that we had a lot of schools that were coasting—not going backward but not going forward either. We had a range of system strategies, which were not really differentiated and were not as data driven as they should have been. And we didn’t have the organization to be able to execute effectively.

We were supported by the state government and were able to pitch them our system-improvement story, which lined up with their ambition to have the best education system in Australia. We had a plan, and of course, the government had some priorities, and we were able to weave them into our strategic plan.

McKinsey: It sounds like there wasn’t much of an internal platform for change before you began. How did you generate buy-in from the system?

Rick Persse: The honest conversations we had with each principal were important. In essence, there is no principal who doesn’t want great outcomes for his or her kids. However, some principals are more experienced, have greater scale or more resources, or might be located in an inner-metropolitan area and have strong teams that are more able to deliver than others.

We have 500 schools and preschools in a large geography, and we were able to argue that the way we were supporting these schools and preschools was inequitable.

Ben Temperly: Prior to the school-improvement model, a lot of the conversations around learning outcomes were based on school mean scores. However, that hides the dynamic in terms of how individual students are tracking.

The conversation shifted from mean scores to a set of performance metrics based on both the learning achievement of every child and the change in that learning achievement over time. This moved the conversation from mean scores to learning growth.2

McKinsey: Some of the markers of success in implementing reform are having a defined set of priorities and not trying to fix everything at once—also being laser focused on implementation and communication. How did you think about defining and setting up the reform program?

Rick Persse: Certainly in the early days, it was very center led. Anne, Ben, and myself were key stewards of the design, and we made it clear that this was our number-one priority. We were pretty honest with ourselves and with our stakeholders about what we were going after and how. There was a package of reforms that needed to happen in concert, and there was also a required sequence around resourcing.

One of the obvious improvements was in relation to our education teams and education directors.3 When I came here, we had about 20 education directors, and each was responsible for a portfolio of 40, 50, or 60 schools and preschools. We were asking them to be the mediating layer between the Department for Education and each school and preschool. So we made multimillion-dollar decisions about resourcing the middle layer better.

Ben Temperly: You’ll have your moments with academics, and you might have your moments with unions. But we didn’t back away from those conversations. We were clear in what we were going after.

McKinsey: Speaking of what you were going after, could you describe the key elements of the reform since 2018?

Rick Persse: I think school-improvement planning was a key deliverable. That was a data-driven planning regime that would be consistent across all schools and subsequently rolled out, in a nuanced way, with preschools. It included a clear three-year school-improvement digital-planning template, which had consistent elements but also the ability to make contextual changes.

Often it was a case of schools having too many priorities and lacking focus. The data that Ben’s people brought to the table helped underpin schools’ priorities, and there were a series of interventions aligned to those, including our guidebooks focused on literacy and numeracy.

The establishment of our professional-learning academy was another key element.

Would you like to learn more about our Public & Social Sector Practice?

Ben Temperly: Rick mentioned support for schools through additional local education teams. That was almost the first priority decision we made. We provided a dashboard for every school leader that included a suite of performance metrics, which we intentionally kept limited but well balanced between leading and lagging indicators. It included literacy and numeracy, year-12 completion and achievement, and some contextual information and organizational health indicators. We tried to be focused and consistent but also provide the basis for a shared understanding of where the school was on its improvement journey.

Rick Persse: I neglected to say each school was plotted on a maturity model4; however, we never published the results, because we didn’t want a league table kind of conversation.

In the past, we’d put out a system strategy that was largely designed around the median, which wasn’t helpful. Now our guidebooks and other tools are differentiated according to our maturity model, which means we are much more targeted than in the past.

We also had fairly consistent key leaders, and our senior executive group has been fairly stable. The work we did to identify our ten-year aspiration would have meant the second-fastest education transformation globally, behind Singapore, so we were very aware of the fact that we couldn’t change the goalposts on everything. As such, our communication to the leaders has been pretty consistent, and I think that will be important as we maintain our shape going forward.

McKinsey: Could you talk a little bit about how you managed stakeholders? Which groups were most important to engage? Were there any particular groups that were difficult, and how did you overcome that?

Rick Persse: There are pivotal groups in any jurisdiction. Principals’ associations are one of them, and unions are another, as well as the government, the opposition, and the media.

Our principals’ associations probably weren’t wildly enthusiastic early on, but we developed design principles that addressed some of their pain points. We also spent some quality time discussing things with them.

We activated the middle layer—the education directors. Boosting their numbers by 50 percent meant that the quality and frequency of the conversations were better. We took the opportunity to recruit from within South Australia but also from outside South Australia and Australia, which was terrific.

I’ve also visited 400 schools and preschools. They respond to being listened to—to us “walking a mile in their shoes.”

McKinsey: That’s amazing. It’s a real example of putting the customer at the center in a very practical sort of way.

Rick Persse: South Australia is a good scale. We’re not tiny, but we’re not so huge that we don’t know every leader. We have been in every school and preschool. All of them email me, Anne, and Ben directly. We do little things. We’ve sent some chocolates to every person in a school or a preschool. But on the flip side, we’ve also made them do their reports, despite the COVID-19 pandemic.5 We ask a lot of our leaders, but they always respond.

Anne Millard: We are immersing people in the conversation, so they hear exactly the same things, or pretty close to the same, from a whole bunch of directions—one on one.

A critical pivot we made was to reset the performance conversations we were having with schools and put the improvement model at the center. It was both useful for schools to have an anchor point for how to talk about school improvement, as well as for us to understand how the model was working on the ground.

Rick Persse: We have an annual Leaders’ Day. There are no hidden agendas. If we need to have a debate about something, we have it with the lights on and in the open.

‘But this is the beauty of a world-class aspiration. We look each other in the eye and say, “Is that going to be a world-class appointment? Is that going to be a world-class website? Is that going to be a world-class workforce strategy? Is that going to be a world-class professional-learning academy?”’

Rick Persse

McKinsey: I’d love to talk a little bit about capability and culture. What capability or cultural barriers did you face? How did you address them?

Rick Persse: There were certainly some capability gaps in a couple of areas, including in HR and in technology, so we recruited in those. I felt we had all the key ingredients in the core business—in school operations and curriculum. Rather than add capacity or capability, we lifted key people up. Lifting Ben’s system-performance group into codriving our reforms was important. We also made a change to flatten the executive structure.

But this is the beauty of a world-class aspiration. We look each other in the eye and say, “Is that going to be a world-class appointment? Is that going to be a world-class website? Is that going to be a world-class workforce strategy? Is that going to be a world-class professional-learning academy?”

Ben Temperly: One thing that helped was codesigning the dashboard with principals. We also developed our own data-product capability, which I think was a positive, rather than relying on outsourced options.

But the big capability we developed was implementation at scale—meaning consistency of application across all of our schools.

Anne Millard: I think there are a couple of things. One is that people quickly learned that the people at the center would be responsive and listen. I think the other was trying to become a much more contemporary organization.

McKinsey: How would you describe the success of the education-reform program since 2018? What were some pivotal moments, and how do you plan to sustain this going forward?

Rick Persse: I think a pivotal moment was having a “single source of truth” in a dashboard. Others were the abilities to plot each school and to have conversations about where they were on the maturity model. A pivotal moment was when every school had a school-improvement plan. They might sound a bit “tick box,” but it was a monstrous effort that the team pulled off.

More recently, it has been starting to see some of the fruits of our labors. Before we started, we were going backward in 12 of 16 domains, compared to the Australian average. We have completely reversed that. The numbers are encouraging, and we are only two-and-a-half years into our ten-year journey.

‘Before we started, we were going backward in 12 of 16 domains, compared to the Australian average. We have completely reversed that. The numbers are encouraging, and we are only two-and-a-half years into our ten-year journey.’

Rick Persse

Historically, we decided that the department wasn’t going to have views on what to teach and how to teach it. Now we are going to have those views, but we’re mindful that we’re doing it in an invitational way.

If we can provide high-quality materials that are helpful and curated with integrity, as well as the best evidence to underpin them, people will grab them.

We’re doing a range of other things. We’ve got this balancing act about not moving the goalposts, but then we’re also folding in accelerators. We couldn’t have done the current curriculum-support program without having a school-improvement regime and a planning regime. And we could not have had any of that unless the data supported it. So there’s a natural sequence that is important.

McKinsey: As you said, you’re two-and-a-half years in. Looking back, is there anything you would do differently?

Anne Millard: I think we have been pretty good at staying on course, but we could have put some mechanisms in to make that even more pronounced.

Ben Temperly: I think Anne’s observation about staying on course is right. But it seems like no matter how hard you try, you can pick up some “barnacles on the boat.”

I think the main challenge for the ten-year journey will be, “How do we retain that clarity of vision?” Avoiding overloading schools with good ideas that are incidental to the overall system aspiration is going to be tricky. I’ll also agree that there’s work required to bring everyone along on the journey.

I’d argue that we needed to be fairly tight in that initial period of time. I think one of the decisions that stood us in good stead was that we prioritized the elements of the school-improvement model that we felt were most important. And it was a fairly constrained reform agenda, based around the improvement dashboard, the school-improvement-planning process, literacy and numeracy guides, and building out support for schools from the middle.

The challenge for us now is to hold to accelerators in a way that means they are truly accelerators and not confound the messaging.

Rick Persse: Obviously, there are national reforms that we participate in, and we are always interested in international developments. It is also partly about the kind of people that we want to work with. I want fewer higher-quality partners over the long run, whether that be in technology or strategy or something else.

What we are really excited by at the moment is actually how to capitalise on COVID. We’ve never had parents more engaged in their kids’ learning, and everyone has now experienced the benefits of our investment in high-speed broadband across the country.

There are always things, given the benefit of hindsight, that we could have done better. But all in all, I could not have asked for more from this group or the educators in this state.

Explore a career with us