Developing software in Silicon Valley is a world away from delivering global health solutions in Addis Ababa, but the uncertainty and risk inherent in both sectors mean that both can benefit from agile practices.
Agile’s highly customer-centric approaches, involving nimble teams, iterative solutions and networks, may not initially appear to be a good fit for global development. The sector typically has rigid rules on how to deliver aid: funding has to be pre-approved, and projects are designed overseas and in advance, far from where they’re implemented.
Development agencies, however, are discovering how effective agile methods can be in this context. An agile approach offers the ability to react to evolving circumstances in a complicated environment with multiple stakeholders – and to build on realised successes. So could agile be a game-changer for the aid sector?
Aid delivery has been traditionally dictated by what tech bros might call the ‘waterfall model’ – a project management technique based around the fixed implementation of a pre-designed plan with objectives often set in an HQ far from the context of their intended destination.
The argument for localisation in global development is that direction comes from a place of priority, not the vision of a distant donor with their own strategy. That’s especially true when the international funder isn’t the source of money but instead reports into another layer, which may not necessarily be developmentally minded. This set of realities can lead to bad outcomes, says Tessa Dooms, who runs Jasoro Consulting and has more than a decade of development experience.
She cites an example from South Africa when, in the midst of the coronavirus crisis, a development worker realised that all their projects required face-to-face contact and therefore couldn’t go ahead. They asked their organisation if they could help their implementing partners to assist their own marginalised communities, only to be told that they weren’t a “humanitarian” organisation so couldn’t help feed them – but could buy them laptops instead.
“That kind of stuff, it evades logic,” Dooms says. “If you’re going to actually have partners on the ground, you must see their needs and be able to respond. If not, why are you there?” Any innovation needs to have close ties between the people and the contexts, she adds.
Agile in development allows responsiveness in the changing environment and adaptation when pre-programmed activities are no longer appropriate. Many projects have a complex set of stakeholders on the ground – including communities, implementing partners, employees, governments and suppliers – where unequal power dynamics can emerge. Proponents of agile say the methodology’s focus on end-users help redress those imbalances, with communities rather than governments or donors having more control over inputs and outcomes.
Short-term, flexible planning
Aid agencies from the World Bank to USAID are embedding agile methods – using small, nimble teams on the ground, engaging partners to co-create solutions or using an agile approach in certain areas such as proposal writing – rather than adopting the method wholesale.
Agile doesn’t replace their management principles but allows for more short-term and flexible planning, says Torbjörn Pettersson, director of HR and communications at SIDA, the Swedish International Development Cooperation Agency.
He also stresses how the pandemic illustrated the rapidly and unpredictably changing context for development work; conflict zones and climate change are other examples of how an environment can change during a project.
Covid-19 restrictions, Pettersson recollects, led SIDA to reprogramme around a third of their billion-dollar budget in 90 days.
“I’m quite certain that the fact we had experimented for a couple of years with agile methodology made it easier for our teams to respond very quickly. And the result was quite impressive,” Pettersson says.
The rise of mobile and other tech influencing delivery in the field means there’s more scope for software solutions to problems. Performance metrics are also more easily identified and applied to improve a project.
Water mapping technology is an example of how clean water and sanitation projects can use data to have the most impact. The World Bank and AusAid used the open tool Formhub to create smartphone data collection forms for a rural programme in Vietnam. This collaborative application shows in real-time how much of the population has access to clean water and if changes in delivery need to be made.
The rigorous approach to monitoring, evaluation and collection of metrics is something private businesses could take as inspiration from the aid sector, says Brigit Helms, executive director of the Miller Center for Social Entrepreneurship.
“What is built into these projects is some kind of regular data gathering from the ultimate users,” she says. There must then be the incentive to learn from that data and incorporate it back into the project, programme or product.
There are limits to the change in course, however, with the agility being in how you deliver, not what you deliver. Wider changes to how development works at scale are also unlikely in the near future. One major challenge is that donors tend to issue requests for proposals for multi-year and complex projects that are still outlined in a framework requiring step-by-step detail.
“It’s really hard to plan development at that level of certainty and oftentimes, these designs lock you into a certain course of action that becomes very difficult to adjust,” Helms says.
It poses the question: who is the real customer when the end-user isn’t the stakeholder holding the pursestrings? But that’s another parallel with the tech sector, where the design of websites and apps may be more informed by the demands of advertisers or other funding entities than end-users. Yet more food for thought on the role of agile in the future of global aid.