“A few million dollars is just a drop in the ocean,” said Muchiri Nyaggah, Executive Director of Kenyan think-tank Local Development Research Institute (LDRI), “but if you know exactly where and how it is needed then you can really achieve something.”

Muchiri and I were discussing how access to structured sub-national location data can improve aid effectiveness. For those unaccustomed to the convoluted lexicon of the open data world, the term refers to data which pinpoints exactly where in a country a development project is taking place.

There is a great need for this data.

During various conversations, aid donors said they could better coordinate their activities if they can see what others are doing. They would, for example, be able to identify pre-existing clusters of funding within a country and consequently tailor their intervention to complement other projects in the region. Alternatively, being able to map where aid is being disbursed already can help donors avoid channelling aid into the same areas, and instead encourage them to fill funding gaps.

Development policy-makers also suggested they will be able to use location data to develop bespoke, more impactful, projects. As one example, researchers would be able to map project descriptions, locations and results together to identify when similar development approaches resulted in different outcomes. This can prompt further investigation into why, and help ensure that future projects are more suited to local populations. Others mentioned the importance of being able to see who is operating where, so they might be able to reach out for best practice advice before, or during, the implementation of a project.

However, despite the near universal agreement on the use case for sub-national location data, few are actually publishing it. In fact, it remains one of the least published pieces of data on the International Aid Transparency Initiative (IATI) Registry — a shared data format set up in 2008 to increase the quality and accessibility of aid information.

To understand how available agricultural data meets user needs, the Initiative for Open Ag Funding assessed the transparency of nine donors, who collectively account for almost $5 billion of agricultural investments. We found that just two regularly publish structured data on sub-national locations of agricultural activities. Five do not provide this data at all.

There are reasons for this.

Firstly, different donors have different business models. In some cases, donors are not involved in the implementation of the projects they fund. Instead, some allocate resources to multinationals or “implementing partners”, who fulfil the activity on their behalf. With this, exact details on what is happening where can get lost along the chain, and retrieving this data can be an unwelcome burden. As one donor representative put it, “it’s almost impossible for us to guarantee that [the implementing partners] will put sub-national data into the system. And even if we could, chasing partners for it would likely require time and resources we don’t have.”

Secondly, there is debate on how to define the location of a project. On one hand, it could refer to the physical location of where the aid was spent, for example a specific town or city. On the other, it could refer to the intended area of impact — in other words, the zone or region in which the beneficiaries of the aid reside.

There are benefits and short-comings to both definitions.

If taken in isolation, data on the physical location of the project can misrepresent how donors are disbursing aid. Muchiri offered me an example: “Consider a specialised hospital; this might intend to serve the whole region but if you plot the funding on a map, it would just show that a huge amount has been concentrated in one large city.”

For this reason, some donors prefer to publish data on the area of impact.

However, ascertaining a clear understanding of who the project will benefit is an imprecise science. The risk with this approach is that areas of impact are defined so broadly it loses its value.  “We need accurate [location] data so we can plan and budget. But it becomes somewhat predictable; the [Myanmar] government is told a project is taking place in several locations and in several states or regions but that the impact is nation-wide, but saying nation-wide is less meaningful for those in partner countries trying to figure out what aid goes where,” Leigh Mitchell, a Senior Advisor to the government of Myanmar, explained.

There are no clear answers yet.

If nothing else, the pursuit to improve sub-national location data has revealed how muddy the issue of transparency is. What is clear, however, is that there is an unmet user-need for this information and with this, the need for a continued discussion on what constitutes useful location data and how it can be best presented. Accordingly, the Initiative for Open Ag Funding has deemed this one of our priorities going forward.

The Initiative for Open Ag Funding hopes that by taking a sector specific approach to broader transparency problems we can make the challenges more manageable and therefore easier to overcome.