"Digital transformation will take time and be expensive, but not as expensive as doing nothing"

Johan Magnusson, PhD, is Professor of Information Systems at the University of Gothenburg and Director of the Swedish Center for Digital Innovation (SCDI). His research focuses on the governance of digital transformation, particularly in the public sector, covering topics such as innovation, technical debt, and ambidexterity in technology management.

With a distinguished career in research and consulting for both public and private organizations, Magnusson has published numerous academic articles and contributed to initiatives aimed at promoting efficient and sustainable digitalization. He is also the author of the book Enterprise System Platforms. This year, he will be a speaker at the Rii Forum, organized by the UOC's Faculty of Economics and Business, where he will share his expertise on digital governance and innovation in the public sector.

What are the main challenges that public organizations face in their digital transformation?

The main challenge is associated with governance. Management control is set up in these organizations to safeguard against risk and variation, and digital transformation is about inviting transformative change, i.e., taking risks and fuelling variation, so we can see that governance and control directly counteract change.

How can organizations balance the need for innovation with maintaining traditional systems?

They simply cannot. They need to fork and create parallel infrastructure for the innovation side as a first step. They need to dramatically increase their spending on IT and digital, and they need to understand that the transformation will take time and be expensive, but not as expensive as not acting.

What role should political and administrative leaders play in IT governance?

Leaders need to act as champions, state clear objectives and assure that front-line managers and co-workers have the freedom to operate that they need in order to change. They have to secure resources and psychological safety.

How do you define technical debt?

Technical debt is the restricted manoeuvrability that comes as a consequence of previous technological shortcuts.

What are its most common effects on organizations?

Spiralling development costs and inertia. In one of our studies we calculated that technical debt imposes a 10X cost for development, i.e., massively decreasing the bang for the buck and pace of development.

Have you noticed any patterns in how organizations manage (or ignore) this debt?

Well, most organizations simply don't manage it. We've identified four strategies that organizations employ. They either just don't touch it, they try to migrate (costly), they revamp (risky) or they envelop. The last of these strategies is the least common, but also the most promising.

What strategies do you recommend to prevent technical debt from becoming a major issue?

We have advised most of the organizations we are working with now to employ the envelopment strategy. This involves taking ownership of their legacy systems, creating a real-time data layer and organizing dev-ops teams that work in a low-code environment to develop new functionality. Over time, this transfers the logic of the legacy systems to a new layer, and over time the systems can be decommissioned. We have only seen one organization do this fully, but it is truly promising.

Do you think public sector organizations are keeping up with the private sector in terms of digitalization?

Public sector organizations are innately different and need to be different. Comparing them is largely irrelevant. I think that the pace of digital transformation in the public sector is way too low, given the challenges we face in society. To be able to handle the green transition, we need to build digital capabilities that unfortunately do not currently exist.

What technological trends do you believe will have the most impact on public sector governance in the coming years? Are there any best practices in digital innovation that you've observed in a country or institution?

In terms of trends, generative AI will continue to be at the top of the pyramid. The potential for cheap cognitive automation holds the promise of what we refer to as precision welfare, i.e., timely and individualized welfare in accordance with needs and rights. In Sweden, teachers spend 50% of their time on administration, social workers 70% and nurses 40%. This severely cripples the welfare system, and needs to change. Our work shows that we can automate up to 90% of administrative tasks, which will help us improve welfare. But the macro-effects of generative AI are truly interesting. The prosperity of a nation in essence depends on two things: access to energy and access to labour. AI increases access to labour at theoretical zero margin cost. Renewable energy increases access to energy at theoretical zero margin cost. This will change the parameters for prosperity.

What led you to specialize in IT governance?

I wanted to understand where the true power in an organization lies. Governance is about the settings for management, i.e., it specifies which decisions can be made and by whom. True power hence lies in the design of governance, not in the making of decisions. The same thing goes with IT: the way you design your system is the way the world will work. So, in essence it was about understanding and influencing behaviour.

What advice would you give to students or professionals who want to specialize in digital governance and innovation?

I would suggest that they start to engage with the problems that organizations face, and that they not accept the current status quo as good enough. Innovation is about exploration, and organizations are designed to avoid it. Utilize the latest research findings to educate yourself, and question the way that organizations are currently working. Digital introduces a new logic; understand that logic and use it as a ticket to ride for helping organizations get better.

If you could change one thing about how technology is managed in the public sector, what would it be?

I would shift over from governance designed to keep IT costs as low as possible to governance focused on assuring value. I would put the spotlight on the opportunity cost, i.e., the cost of not utilizing technology. Once this has been done, things will work out just fine!