MarkoVS87 - Fotolia
Margo Ter Bekke is a data strategist and her job is to see what there is in the data: what kind of information an organisation has, what kind of information it can get from outside, and what is lacking. In the process, Ter Bekke needs to know how each dataset can be used in the business, which for her is the local government in South Holland, the Netherlands’ most populous and most industrialised province.
“We are working on climate, waterways, mobility, nature – all kinds of things related to spatial planning,” she told Computer Weekly. “That’s what a province is mainly about.”
According to Ter Bekke, there is a big difference between a data strategist and a data scientist: the former tells a story about the data the organisation has; the latter tells a story with the data the organisation has. A data strategist works with data analysts and people who put data into a data warehouse or a data lake, and people who clean data. A data scientist can then use the data to help people make decisions by presenting maps, graphs and dashboards created with sophisticated techniques, such as machine learning.
South Holland is a good example of an organisation that needs a data strategist. The provincial government has nearly 2,000 full-time employees, who look after a diverse area – with harbours, a coastline, a lot of agricultural land and two big cities – Rotterdam and The Hague. The province has a lot of information that can be combined to create a big, complicated puzzle that helps people make decisions. But to do that, someone needs to join up the dots between all the available data.
Data scientists help to find solutions
Using data to make big decisions is not new to the Netherlands. A controversial topic under discussion right now is the quantity of nitrogen – a pollutant – in the air. Although nitrogen makes the soil richer, it is bad for biodiversity in general.
Using sophisticated data analysis techniques, the nitrogen deposition per nature reserve has been mapped out, as well as the sources of this pollution. These are mainly dairy farms, but nitrogen also comes from industry, roads and housing.
Data scientists use information to tell the story. They take pollution data from satellites, drones, calculation models and aerial measurements and combine it with data from farming permits. Then they use the aggregate information to create maps of the Netherlands that show where the pollution is and why it is there.
“The nitrogen deposition problem is a major social issue for which government and society are seeking a solution,” said Ter Bekke. “The stakes are high, the interests huge, and that places demands on the information presented. This is a good example of a process facilitated by data scientists.”
Another problem that data scientists are looking at is the shortage of housing in Holland. The challenge is to give each community the right share of housebuilding, while at the same time protecting the environment, nature and spatial quality.
There are recent refugees from Ukraine and other countries – and such immigrants need housing too. But there isn’t enough, so new houses need to be built and existing buildings need to be transformed.
The solution is not in the data itself, but the data can facilitate the process of finding a solution. Data scientists obtain data from each region and take several factors into consideration, including political and administrative issues.
“Housing has been a major topic for many years and a lot of institutions have a lot of data on it,” said Ter Bekke. “But the puzzle is getting more complicated, because we have to combine it with climate change and green energy data. Over time, there are more things to combine – more types of data and higher volumes.
“We now have a minster who is really trying to get those houses built. But at the same time, you have the climate, energy, and carbon-neutral restrictions to consider. You have to factor in social housing, and whether solar panels are put on the roofs. You also have to think about heat in the cities and whether there are enough trains to serve the growing population.”
One area of concern is how the models developed by data scientist are interpreted. Too many people assume that their models always work – and that can be disastrous when the stakes are high. Another consideration is that maps and graphs sometimes create the illusion of a clean picture. People can be misled by a nice presentation.
Ter Bekke added: “Whenever you have a model that predicts, you have to remember that it can also predict wrongly.”
Data strategist connects the dots
“My job is to think about how we work within the province, with all our people,” said Ter Bekke. “How can we get a better picture of who is working on which topic and how we can work together with other governments and other parties?
“My main concern is what is happening within our organisation and around our organisation. I am interested in the information that organisations have. What can we know about the different projects? How are groups communicating?
“We depend more and more on dynamic ecosystems. This is true in provincial governments, but it also applies to other organisations.”
Ter Bekke has an engineering background and has worked for the government for more than 20 years, in several different positions and departments, including spells in finance, interspatial planning and waste management.
“The diversity of my background and my technical background help a lot,” she said. “Having worked in different roles has made it possible for me to see how the business of government is run and what kind of data they need to make things happen. To work as a data strategist, you also have to be curious, and you have to be interested in people.
“Business and government are all about people and connections. The informal organisations and dynamic ecosystems have become essential. That is why, over the next five years, the role of data strategist will become even more important.”