5 Q’s for Sébastien Deletaille, Founder and Chief Executive Officer of Real Impact Analytics
The Center for Data Innovation spoke with Sébastien Deletaille, founder and CEO of the Belgian company Real Impact Analytics (RIA) which provides organizations with the capability to analyze, and act upon large inflows of data. RIA has operations in Europe, Africa, and South America.
This interview had been edited.
Paul MacDonnell: Your company has primarily developed data analysis tools for the telecommunications industry. Why did you choose to focus on this industry?
Sébastien Deletaille: From the work I was doing at McKinsey before founding RIA, I knew that many telecommunications operators were not getting value from the enormous investment they had made in management information systems (MIS). The structure of the industry has generally led it to work in silos, and its MIS technology tends to follow that structure. What this means is that different divisions of the business, such as network operations, are not talking to the marketing side. This leads to suboptimal decisions, such as investing in new infrastructure based on the number of users in particular location rather than using data showing which users are prepared to spend more on different services carried by that infrastructure. When we looked at what many major companies were doing with their marketing we saw that 7 out of 20 marketing campaigns they were running were destroying value.
So here was an industry that is rich in data but, because much of it was in silos, companies were not realizing its potential. We saw the opportunity to develop transversal applications—applications that would work across the different silos—to extract, interrogate, and present data in a form that is useful for decision-making. So we’re integrating different information silos and providing useful customer and marketing-support data via our platform on smartphone and web-based applications.
MacDonnell: How do you manage working with your customers’ data from across so many different systems?
Deletaille: Our approach is that we don’t want our value to our customers to be dependent on their technology, but we do want it to be dependent on their data. Rather than delivering “custom builds”—that is, writing scripts in our clients technology choices—we develop applications that are agnostic to the client’s legacy architecture. We do this using Docker—we package all our logic in several containers and deploy them within the client’s technology environment. No complex wiring is required. It’s easy to upgrade and extend, while connecting to the client’s data. [Ed. note: Docker is a popular open-source platform used to ensure an application functions consistently regardless of the software platform it is run on.] This allows us to extract and analyse the data we need in any architecture or systems environment without the need for integration.
MacDonnell: How do you customize your applications for different customers?
Deletaille: The customization depends upon the needs of the users. Our applications serve up cross-silo information to staff via our own platform and, based on what our customers want to accomplish, we customize the embedded analytics to present insights that support decision-making and performance and outcome tracking.
MacDonnell: RIA has an initiative called “Data for Good” that produces research and software tools to support development agencies in working towards sustainable development. How does it work, and what is the relationship with telecommunications?
Deletaille: Our work with telecommunications operators has brought us to emerging markets and to the developing world. We were approached by the MasterCard Foundation, the Bill & Melinda Gates Foundation, and the World Bank, which asked us whether data gathered by telecommunications operators could be used to yield insights into poverty and disease levels in developing countries.
Using anonymized data about mobile credit top-ups—equivalent to purchasing more minutes or data for a mobile phone—in Central Africa we found a positive correlation between reduced spending on top-ups and instances of food insecurity that had been identified from household surveys in that region. In essence, less spending on phone credit signalled household financial stress and acted as a leading indicator for hunger. We then developed an app for workers in this region that allows them to anticipate this problem.
MacDonnell: At its core, RIA uses data to help companies make better decisions. How do you think this industry will evolve over time?
Deletaille: Ultimately decision support technology will become embedded into organizations and, like electricity in the modern workplace, it will be invisible. I can see the trend of decision-support technology becoming more sophisticated and eventually playing an essential and direct role in the work of the most senior managers in many organisations. I’m convinced that technology like ours will, over the next 20 to 30 years, transform our economy and bring about an industrial and productivity revolution.