From Silos to Synergy: The Journey of Open Source in Energy Access

Globally, about 655 million people live without electricity, and about 2 billion rely on polluting fuels and technologies for cooking. Closing the energy access gap starts with knowing where it is. Energy planners need to identify unserved communities, understand their energy needs and determine the most cost-effective way to meet them.

To answer these complex energy planning questions and manage operational challenges, practitioners rely on specialized tools and software. Most of that software is proprietary: It is available only to licensed users, and its code and assumptions are closed to outside review. A growing share is open sourceⁱ, published under licenses that let anyone use, inspect, modify, share and commercialize it.

Years of development have produced a vast but fragmented collection of tools. For energy planners, that abundance creates a new problem: overlapping tools that rarely work together. More individual tools will not close the energy access gap. Fewer tools that work better together could. Building a tool in isolation is faster in the short term, but it adds to the fragmentation. The lasting value comes from a community that holds itself to shared standards over years. WRI and its partners have started one. Sustaining it will take coordinated development, consistent licensing and funding aligned with these standards.

For example, a well-intentioned, donor-funded tool launches, and six months later another arrives to solve nearly the same problem. Because these tools rarely connect, energy planners are left to sort out which to use. Each new tool brings new data requirements, new formats and new training. Planners end up acting as human middleware, copying outputs from electrification analysis tools into demand estimation or cost modeling platforms and spending hours running several tools to answer a single question.

This is not only a complaint from energy planners. At a recent meeting with a national energy ministry, we heard the same message in plainer terms: There are too many tools, and they need to work together.

Connecting the Ecosystem

The issue is not a lack of software. Tools capable of responding to operational and planning challenges in energy access already exist. The issue is a lack of coordination and widespread duplication.

Working within the sector, we also noticed a pattern of teams taking existing code, building their own disconnected spin-offs for niche uses such as cooling or agricultural planning and rarely contributing improvements back to the original project. Other organizations were simply building from scratch because it was faster in the short term, without considering future maintenance and evolution. The underlying problem was the lack of contribution back to the open-source ecosystem.

To address this, EnAccess Foundation and WRI initiated the Open Source in Energy Access Symposium (OSEAS).

Good tools were being abandoned, reinvented and abandoned again. The consequences of this cycle are severe: wasted investment, eroded trust in digital solutions, burned-out development teams and, ultimately, delayed electrification decisions and underuse of tools in operations.

Collaborating doesn't mean folding everything into one tool and asking tool developers to stop building. It means breaking the problem into modules, building on them and feeding improvements back into the ecosystem so everyone benefits from the advances.

Lessons from Abuja

At our first OSEAS symposium session in Abuja, Nigeria, in 2024, WRI asked tool developers and energy planners to map the landscape of energy access planning, design and operation tools. The resulting digital whiteboard (Miro board) showed dozens of widely used platforms (Figure 1), such as OnSSET, RAMP, Energy Access Explorer, OSeMOSYS, OffGridPlanner and Autarky, spanning mini-grid planning, integrated energy planning and demand simulation. But all had multiple near-identical tools competing with them, and some were operating in complete isolation.

A Blueprint, Not a Magic Wand

We spent the next year working out how to approach this and ultimately developed a shared collaboration framework, a practical governance document that defines how tools, data and organizations in this ecosystem should connect, share and hold each other accountable. The framework rests on five principles: a shared vision, inclusive co-creation, interoperability through open standards and modular design, local capacity and ownership, and accountability.

Figure 2 shows the four groups that shape the ecosystem (solution providers, data providers, funders and donors and users) and what the framework asks of them. The sharpest ask is of funders and donors: to consider supporting consortia rather than isolated, parallel workstreams; to fund organizations that take on or extend existing tools rather than starting again; and to encourage collaboration in practice, including through joint KPIs between grantees where feasible. For solution providers, the framework proposes that they check the project registries and literature for existing tools before building new features and publish high-quality user guides and separate developer documentation so others can adopt and extend their work. The framework asks data providers to publish datasets in open, standardized formats with complete metadata. For users, it offers a way to identify different tools, see where they can integrate their outputs and raise questions with tool providers.

This need for institutional alignment is not a new finding. The Open Energy Modelling Initiative wrote up its experience of opening code and data in 2018 and concluded that institutional change matters as much as individual choices, and that a community rarely builds itself without someone to take responsibility. Developers of the OSeMOSYS tool talk about developing a community of practice and acknowledge its limitations: A  systematic review of OSeMOSYS found the governance structure the community established is still not consistently applied, a decade on. Outside the energy modelling ecosystem, the Principles for Digital Development ask funders to understand the existing ecosystem and to share, reuse and improve.

Learning from Past Experience

The framework also calls for a lightweight governance structure: a community manager, regular stakeholder calls and designated coordination champions within participating organizations. These are not bureaucratic additions. Unless someone owns coordination explicitly, it simply will not happen. Rather than creating a completely new governance structure, the framework could build on existing ones, such as Energy Modelling Community or the Open Source Energy Access (OSEA) community.

The framework is a blueprint, not a solution in itself. A document sitting on a server cannot force collaboration, and no single organization can do this alone. It needs organizations to dedicate staff to integration; donors to fund the coordination layer and encourage collaboration between grantees through joint performance indicators; and users to demand interoperability. The OSEA community offers one example: Over the past year, a dedicated community manager has run the community and recently piloted an ambassador program.

Changing the Tone in Nairobi

When we convened for our second session in Nairobi, Kenya, in 2026, the atmosphere had shifted. Abuja was about diagnosing the problem. Nairobi was about asking what teams would do and by when. We wanted participants to make a basic, tangible commitment to show their motivation, anything from a small, shared dataset to a co-authored technical blog.

The practice of open-source collaboration is not new. Energy Access Explorer already incorporates the outputs of OnSSET in Kenya and OnSTOVE in Nepal, alongside demand profiles for healthcare facilities from RAMP in Uganda. These tools build on, rather than duplicate, one another.

We asked developers to record testable commitments on a shared board. They committed to ensuring data exported from one tool feeds directly into another within a specified timeframe. Teams sat with each other to discuss their work with an open mind rather than a defensive mindset.

We left Nairobi with a documented list of commitments, including contributors and indicative timelines.

The Hard Truths

Real progress also uncovers hard truths. Two major hurdles remain unresolved.

The first is licensing. Our framework supports sensible open defaults, such as standard permissive open-source licenses, and a new tool can adopt them on day one. Tools that have run for years under different legacy terms cannot simply switch. These are institutional and legal questions, not technical ones. We can point the way, but we cannot make it simple.

The second is capital. Shared documentation, community managers and joint testing infrastructure cost money and are often the first things cut when budgets tighten because they are hard to tie to a new deliverable. The governance structure we propose has no dedicated funding, and the Nairobi commitments rest on time people find alongside their funded work. This is where we ask funders and donors to come in. Coordination has to be treated as fundable work in its own right, not as goodwill.

Looking Ahead to Kigali

Our next session meets in Kigali, Rwanda, in October 2026, hosted with Carnegie Mellon University Africa. We are going to review the Nairobi commitments one by one: what was met, what stalled and which real-world integrations we can learn from. We are also bringing more end users and data providers into the room, so the people who experience fragmentation daily can help shape what comes next.

The whiteboard from Abuja is still fresh in our minds. Kigali will show how much of it we have managed to untangle.

The collaboration framework is available here: Collaboration framework.pdf

ⁱ Open source software is distributed under terms meeting the Open Source Initiative’s definition, including free redistribution, access to the source code and the right to modify and redistribute derived works. See opensource.org/osd