My path to data started indirectly as a sales trader in an investment bank. Over time, I found myself coding our systems in my spare time. I found this more interesting than trading. I wanted to build things and solve problems, leading me to move to IT full-time. As I progressed through the ranks of IT, data was a regular part of my projects. I saw first-hand how important data was to our business and created new hierarchical data models, data platforms and managed transformational change initiatives which were heavily dependent on data. In 2014, I joined Mizuho as head of risk IT. Data was again an important topic and data-related regulatory programs were on the horizon. In 2015, I was asked to transition to my role as CDO with broad responsibility for creating and implementing our data strategy. This is focused on using data to enhance revenue opportunities, find cost efficiencies and ensure regulatory compliance. Over the last four years, I’ve witnessed a transformation in Mizuho’s cultural awareness about data. I’ve been lucky with a great IT department whom I’ve worked with closely to realise a best-in-class data architecture. This has helped deliver significant operational efficiencies and enable our growing analytics capability.
Seeing the cultural change take hold at Mizuho has been incredibly gratifying, especially as I’ve built our data office from scratch. It’s not often that you get an opportunity to build a new department from the ground up. I’m now seeing real traction deploying the work from our European business across the rest of our global organisation. Seeing how this has been received by our business and the positive impact that it’s made is the real highlight.
There’s a quote attributed to Thomas Jefferson: “In matters of style, swim with the current; in matters of principle, stand like a rock.” I’d refer to this suggesting flexibility and pragmatism wherever possible, except for those things that really matter. As fast as you can, learn what really matters.
I think it has been more in line with my expectations than not, but life has a way of generating surprises. With data science activity generating a lot of questions across the company, we launched a data science community earlier in the year. I was pleasantly surprised at the level of interest and participation across the company. The community is now working through a pipeline of experiments and been a valuable source of input that’s helping to improve our data science capabilities.
I think 2019 will bring a focus on crystalising value from data programmes (both governance and machine learning). As we approach a more challenging global economy in 2019, I think that there will be a growing pressure on economic value and deliverables. This will lead companies to focus on automating data governance and to be more selective in their data science initiatives. Many companies have experimented with data science, but few have actually seen profitable returns. In 2019, this might well change. As budgets are constrained, I think companies will make hard choices about how they invest in analytics.
My focus is on internal staff development and academic partnerships. The data science community that we’ve established is helping to identify internal talent and capabilities. We will continue to grow our internal talent in this way. I believe that one of the key skills that data teams require is domain specific knowledge, so internal development is the key. I want staff with strong banking knowledge to develop greater analytics skills required for future projects. We will continue building academic partnerships with universities. These have proven to be valuable relationships where we can offer students interesting projects that are relevant for their degrees while identifying talent for recruitment. We also want to work with universities to help identify topics in their curriculum that are most valuable when we recruit.
I think the convergence of data sets and analytics across companies (eg, data partnerships) offers new and substantial benefits. There are a number of key hurdles to overcome to enable this properly, including industry-wide data standardisation and the ethical frameworks to supervise these activities.