Unlocking Digital Innovation in Public Organizations: Digital Journey of Asian Development Bank - Part 2

Digital transformation is not just a hype; it plays a crucial role in ensuring that long-term development goals are attained in an efficient manner. MEDICI got the opportunity to connect with Ozzeir Khan, Director, Digital Innovation and Architecture Division, Asian Development Bank (ADB), and discuss the various facets of innovation in public organizations. In the first part of the interview, Mr. Khan shed light on how ADB, as an organization, is working toward digital transformations in its associated public projects. In the final part of the interview, the focus is on how a regional developmental organization such as ADB works to ensure that innovation is sustained over the long term to bring about a positive transformation in the society.

MEDICI: How is ADB working with digital innovations? We have been saying business as usual, but have you made it ‘innovation as usual’ at ADB? What is your approach to deal with innovations within the organization?

Ozzeir: That’s a very interesting question regarding how you ensure the sustainability of innovations, progress, and forward steps. Sustainability has always been a challenge for large organizations because of the nature of enterprises. In the public sector, especially when the demand is high for your current services, how do you engage and ensure that innovations do not affect projects and are not a one-off thing? How do you have systemic innovations within organizations? There are many different aspects to this—there are some foundational pieces that you have to put in place, starting with some framework in place for giving space for innovation to organizations (budgetary resources or support for innovation). There are some organizational elements to enable that innovation. Fundamentally, it’s a cultural adoption issue as well because a lot of this innovation for the past couple of decades has been around digital. How do you bring a digital culture into place? You do it by ensuring freedom and creating some sort of resources systematically within the organization, supporting digital culture formation, giving employees room to breathe, and letting them engage with customers. Once you start doing that in terms of not one department or one group but as an entire organization’s effort, the snowball effect comes into place. The way you keep the momentum going needs to become part and parcel of any discipline in that organization. Look at advanced companies or corporations that have adopted and are successful—it's part of their daily work, and not just about the best practices. It's about how you do things, and that ‘way’ becomes part of every PowerPoint and every e-mail. Digitalization makes its way into their institutional strategic planning, documentation, and efforts. The process takes time in public organizations. ADB started these programs during 2017–18. I can see these programs snowballing and making their way into strategic planning exercises and budgetary exercises—this is how you make a system and make it a part of an organization.

MEDICI: Do you have any theoretical or conceptual model or framework for innovation at ADB? Is there a workflow or process flow to make sure that every action goes through the defined steps, or is it embedded within the process itself (understood by every team member)?

Ozzeir: We do have a process that we follow. Personally, I found that (process) to be successful in the public sector. The process is very similar to how pharmaceutical companies work when they develop a new drug. They have a concept of R&D, R&D testing, and real production testing (testing in real-life scenarios before it’s approved for mass adoption). Generally, you have to follow the procedure as you're testing something new—you have to first test on something dummy. Testing on people and real business transactions, however, is risky. This is where our process allows the sort of pilot test with real transactions. Otherwise, it would never leave the R&D shop. So, yes, we have a process/methodology for pilot testing, which goes from just technology testing to production testing where certain pilot testing audiences are chosen. Subsequently, test results are compiled over a course of six to nine months—depending on the type of innovation we are looking at—so that we can truly test the best practice for the market.

MEDICI: As you said, usually, there is a process for any critical project—testing on a sample audience and moving to a larger audience. However, a key challenge here is time. Do we have a process/methodology or a sprint sort of module that you run to speed up things?

Ozzeir: Yes, because in the innovation game, you never start off with hard requirements. There is no real benchmark in terms of what works and what doesn’t. In your testing, you're finding different dimensions to benefits which you may not have thought through. For example, the pharmaceutical industry is trying to develop a cure for COVID-19; it may find out that the drug cures malaria. That’s kind of what happens in these innovations, especially with AI. 

We're doing a lot of work on AI. As you know, AI requires training; when your first AI is born, it needs to go through many cycles of training before it becomes effective. In our case, we go through the cycles and expedite the process. Depending upon the adoption, we even keep them. Most of our cycles are between three and nine months. Sometimes, if the adoption is low and we see that the long-term benefits to the market are not immediate, we keep it on the back burner and keep it in the pilot phase for a long time to let it brew and let the market sort of adopt, because these are new things and we don't know when is the right time for them to be engaged. I can give you another example. We are aware of how AI can predict or recommend what countries should do for development. The more I feed AI on the history of the world, the more it recommends what we need to do in a certain city, town, country, or region. The first iteration of that was not that effective, but it pointed out on a broad basis where we should focus. As we train it more with the history of the world and current and planned events, it can be much more effective in terms of providing insights into what we should be doing for the future. It’s something that takes time to understand and grow. You can't keep it out there for years either (normally, six to nine months or maybe a year); if you're still not catching, we'll shelve it and go back to it.

MEDICI: You mentioned a very interesting project, Ozzeir, i.e., teaching AI to learn from history and then predict or suggest something for the future what countries should do. This is very ambitious, and I would be interested in seeing this to fruition. My next question is on some of the milestones that you've achieved so far (along the lines of digital innovation within ADB). Do you have any sort of plan or layout (for the next one or two years) for what you want to achieve? What sort of projects do you want to see completed in the next couple of months and, then, years?

Ozzeir: Over the past 1.5–2 years, we focused on laying the foundation in terms of methodology—understanding what works and what doesn't, experimentation, and ways of bringing innovation. We looked at promising emerging technologies. Thus, a lot of foundational work is pretty much done. In the initial couple of years, we did more than a dozen launches of products and services. When you do over a dozen product launches, you get a pretty good sense of what works and what doesn't. The next wave that you're looking at is going to be interesting because we are now done with the basics; we know how to bring innovation and are clear on scenario-specific approaches. We know the audience as well and what type of impact we're trying to make. We'll also have some basics in place in terms of how to manage large volumes of data or manage more data. We have some basics in place, for example, an AI or IoT lab. We have some basic organizational measures in place to support the future. In the coming years, we are going to look at more ambitious projects; we will push the innovation index even higher. 

In terms of collaboration, I think we'd be getting into many collaborations directly with countries. For example, when we see advancements in some areas in AI, we'll work directly in China, or if we see some specifics in terms of FinTech in Singapore, we'll work directly with governments. I think, in public organizations such as ADB, we need to figure out how to connect with the GovTech sector to be GovTech labs in countries. We are trying to develop more consortiums, and I'm looking forward to that. Many of these require consortium buildings (digital infrastructure buildings) in the region. For us, it's like going outside the walls of the bank and pushing the innovation index even higher. Considering the few technologies that we are seeing, I think we'll be focusing more on data-driven insights—more on IoT, more satellites, and new products (especially in climate change, among other areas).

MEDICI: Thanks, Ozzeir. As we are nearing the end of this conversation, is there any message that you would like to share with your innovation partners and collaborators?

Ozzeir: Our interest in digital innovation is about the development of Asia, and we believe that we cannot do this alone. We need partners that have similar interests and similar sustainable development goals. From AI to DLT to digital highways (in terms of financial digital highways), they all require consortiums, partnerships, standards, policies, co-innovation, and co-creation to move forward. Even simple things such as cross-currency payments, cross-country identification, migrant worker issues, and financial inclusion require us to cooperate digitally. We need to share data and work together for success. We are open, we would like to reach out and see how we can partner with the right organizations that have similar goals. We are looking forward to working with not just traditional technology companies but also other entities (let’s say, the CTO of a textile industry). There is something around digital innovation which we can work together for trade and financial inclusion. For example, there's a lot of talk about the waste that is generated by certain industries. We would like to partner with people or companies interested in circular economy issues; we would like to see how we can innovate together to meet some of the challenges that we're all trying to manage.

MEDICI: Thank you for your time, Ozzeir.

Ozzeir: Thank you.


Note: This is the final part of the interview, which was conducted on October 19, 2020, with Ozzeir Khan (Director, Digital Innovation and Architecture Division, ADB).