This is the first installment of ‘Notes from the CEO,’ a series of reflections from GRID3 CEO Marc Levy on the ideas, trends, and challenges shaping the organisation’s work.

GRID3’s work on settlements grew out of earlier efforts that were part of the drive to eradicate wild polio virus in northern Nigeria in the early 2010s. In order to map settlement extents back then, supercomputers at Oak Ridge National Labs were needed. It was a prohibitively expensive process that GRID3 could not replicate.

Advances came quickly. At roughly the same time that GRID3 was launched, its funder, the Gates Foundation, contracted the commercial firm Ecopia to map all of the extents for sub-Saharan Africa. That investment made high-resolution settlement maps available to all as a public good. But it was still expensive and would have been prohibitively costly to continue indefinitely. A major leap forward occurred when the work that was once being done by supercomputers at a single location started getting done by companies like Google, Microsoft, and Meta, which all had enormous surplus computing capacity and were already acquiring large volumes of high-resolution satellite imagery. The margin cost of turning out detailed settlement data for these companies was so low that they were willing to offer it as a public good without anybody having to raise money for it. That breakthrough enabled GRID3 to begin producing detailed settlement data for all of sub-Saharan Africa and make it available at no charge. 

We are now in the middle of another major breakthrough in settlement data availability. Data aggregation is now done as a public service by organizations like Overture–these companies ingest the free data from Google, Meta, and Microsoft and harmonize them into a single data source. Updates now come out every month and the quality of the data is improving all the time; datasets can provide granular detail such as building height and time-series data, and a user can zoom in on maps and see individual footprints of buildings. These high-volume, high-quality footprint data allow us to see settlements at a much finer level (for example, distinguishing between purely residential areas and mixed residential-commercial areas). 

Thanks to these advances, we are now producing much more detailed and information-rich settlement data that delineate neighborhoods within settlements. GRID3’s extents for Nigeria have just been released, while those for DRC are being finalized. Maps with this level of specificity allow officials to fine-tune campaigns to make them address a community’s specific context, and also allow them to improve accountability by knowing exactly where vaccinators should be going. 

The technology will keep evolving, and so will we. Each breakthrough in how settlement data is produced and shared has been an opportunity to make GRID3’s work more precise, more comprehensive, and more useful. We intend to continue seizing those opportunities.