Nicolas de Cordes: “It’s time to scale up Big Data for Social Good”

So Nicolas, how did you end up working in this space and where did the Data for Development Initiative come from?

Well, everything started at the NetMob conference in 2011 at MIT where I was taking part as a member of the scientific committee for these analyses on mobile data. In one of those typical “networking coffee breaks” I ended up discussing the data for social good opportunity with several colleagues and a few months later we formed the Data for Development initiative at Orange. Fortunately, my current role allows me to explore new business opportunities in the telco sector and mobile data for social good is one of the most exciting ones I’m working on at the moment. 

And which project within the initiative are you proudest of?

There are so many incredible projects out there, however, I was particularly fascinated by one analysis which combined mobility data with medical data to fight Malaria in Senegal. We received a request from a local coordination body to provide Data as a Service (DaaS) sending weekly data to doctors on the ground enabling them to optimize their deployment of resources. If we manage to develop this service for example, they could be distributing mosquito nets and vaccines in the most efficient way possible.

Other projects have been looking at poverty index and literacy rates. This was particularly rewarding because we really helped the National Statistic Office to be a lot more precise. Mobile data allows them to measure things more frequently, which is so important in terms of development. Of course there is a lot of bias in our data (e.g. market penetration) but you can still gather great insights and take better decisions as a result. 

What are challenges of working with national statistics organizations being a mobile operator?

There are many, but fortunately there is a common opportunity. They need data more regularly and more granular in particular to measure the new sustainable Development Goals, and we want to contribute. However, there are always some disparities in culture and skills between the private and public sector. By nature, telcos are able to develop Big Data tools with weekly and monthly insights (sometimes even in real-time) from data provided by their mobile networks whilst public organizations don’t always have the resources or infrastructure to extract insights at the same pace and their decisions are taken with different time scales in mind. The key is to work closely together to find a healthy way to collaborate going forwards for both policy making or strategic decision needs on one side and for operational and tactical needs on the other side. 

And how do you think we can achieve that healthy collaboration?

Advancing social good projects is unfortunately quite slow, however, if you take a step back, we can see significant progress over the past 3 to 4 years. Going forwards, data privacy is going to be one of our greatest challenges to address.

“Going forwards, data privacy is going to be one of our greatest challenges to address.”

Everybody is talking about it, whether it’s the data of groups of people or individuals and it’s a legitimate concern which needs to be taken into account. We must work through that issue together to overcome these barriers.

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 16.0px Calibri; -webkit-text-stroke: #000000}
span.s1 {font-kerning: none}

From your experience, do attitudes towards privacy vary much across different geographies?

Absolutely yes, but to answer this one I’ll use an anecdote from a project we participated in in Senegal. The first thing we did there was to visit the Data Protection commission. Coming from Europe, I was expecting to encounter some challenges, however, the head of their legal department stopped me in my tracks and said: “We have a different attitude here, in this country people die because of the lack of information, so we really need your project.” This was the moment when I realized that the mind-set is completely different depending on the stage of development the country is in. 

Many are starting to talk about Data Philanthropy in the world of Big Data for Social Good. How do you think we can set a sustainable model which keeps all parties happy?

Doing this is fundamental so that we grow from doing some exploratory pilots here and there to doing something frequently, as a service. In order to scale up, we will have to make this profitable. Of course, due to the nature of these projects, we can’t make profit the main driver of our actions, the private sector has to take into account the indirect social benefits of making such projects work.

“Of course, due to the nature of these projects, we can’t make profit the main driver of our actions, the private sector has to take into account the indirect social benefits of making such projects work.”

With regard to Data Philanthropy, I prefer to call it Data Access Philanthropy as it’s all about finding ways for communities to access and work with information safely – without completely opening the data in an insecure way. We need a model to enable institutions to explore the use of data, understanding that it can’t always be free – which is precisely what we are trying to do with OPAL (Open Algorithms Project).

It’s 2017 and suddenly we’re hearing a lot more about Artificial Intelligence than Big Data. How is that applicable to your projects? 

As a new complementary tool, next to more classic algorithms developments, AI is really important for the sector, particularly the Machine Learning element. It should enables us to work with massive volumes of anonymous data in a more efficient way, as a lot of the information we extract is not direct, allowing us to infer from correlations, reapplying data sets and achieving valuable insights. In the case of OPAL, when we open code and algorithms, Machine Learning will also be a key piece of our model and we will need to find a way to make it transparent and subject to human oversight by local constituencies.

“Machine Learning will also be a key piece of our model and we will need to find a way to make it transparent and subject to human oversight by local constituencies.”

One of the essential aspect of the OPAL project is to work on the Local Data Governance, to ensure that the outcomes of the project will fulfil the needs of the local communities and partners in term of societal, profitability and environmental aspects. AI is just a new tool to support this more global approach focusing to answer to multiple user needs.

What’s your dream for the Big Data for Social Good space in the next few years?

The biggest priority has to be cooperation, between telcos like ourselves, in line with the 17th Sustainable Development Goal. I cross my fingers for the OPAL project, hoping that the platform will be built within 2 to 3 years with initiatives in both Senegal and Colombia through a co-design between telcos and users focusing on Development sector. I believe that it can become a reference for other industries apart from telecommunications so that we start to see similar platforms, using a “Safe Q&A” mechanism, in different sectors, allowing more diverse projects to access insights, which bring even more value to decision-makers in developing countries.