Ana Carsalade on what happened when BFS surveyed 700 people before designing a waste programme in Rio, why the J-PAL methodology inverts the logic most development projects use, and the lesson she returns to when she thinks about what this decade of work has actually taught her.
This is the ninth in a series we are calling 10 Years. 10 Lessons, where BFS team members share what they actually learned from their hardest projects.
- Ana Carsalade is an Economist and Consultant at BFS. Her background in development economics shapes how she approaches waste problems: she thinks in terms of evidence, community incentives, and systems, not just technical outputs.
- She has been at BFS for five years. Two and a half of those she spent as Technical Assistant to the CEO, working at the centre of the company’s decision-making. She has seen BFS from both ends of the organisation.
- The J-PAL methodology inverts how most development projects are designed. Instead of deciding what the problem is from a distance and designing down, J-PAL collects evidence from the ground and designs upward.
- On the Guanabara Bay in Rio, the project surveyed 700 community members to inform the design of the collection and reward system for a four-year plastic waste project. The evidence from that survey shaped the entire architecture of the programme.
- BFS has evolved from wanting to work on everything to working on specific things very well. Ana sees that shift as one of the most important things that happened in the last five years.
- Her lesson: there is no away. Waste does not simply disappear. Instead, she argues that we must change the logic from waste to resource, designing systems for circularity to eliminate, as much as possible, what eventually becomes waste. Systems designed around this circularity work differently from those built on the illusion of disposal.
|
60% Of all marine debris worldwide is plastic, with rivers and informal settlements near water bodies among the primary pathways Source: UNEP Global Marine Litter Assessment |
1,100+ Randomised evaluations conducted by J-PAL-affiliated researchers in 93 countries, directly shaping policies reaching hundreds of millions of people
Source: J-PAL Global, 2024 |
7.2% The fraction of all materials the global economy currently recycles or recovers. 92.8% of everything extracted is eventually wasted or permanently lost Source: Circularity Gap Report, 2024 |
Ana, you started at BFS as Technical Assistant to the CEO, and now you are a consultant running your own projects. What actually shifted in that transition?
It was not a hard transition. I was already shifting while I was still in the first position, because Kevin was working on projects too, and I was assisting him on those before the formal change happened.
What shifted is the focus. Before, I was looking at the bigger picture, the company as a whole, how do we move forward as a team, what is the environment we are operating in. Now I am completely focused on the work on the ground. When you go into projects, you are really into that specific topic for months, sometimes for years. The project I am working on now, the Social Recycling Shipyard on the Guanabara Bay in Rio, is a three-year project. You dive deep.
I think of it as moving from a wider lens to a more focused one. The skills and instincts from the first role still apply. But the direction of the work is different.
You did the MIT J-PAL MicroMasters just before you joined BFS. What were you looking for when you chose that programme?
The J-PAL methodology inverts the standard logic of how development projects are designed.
Most of the time, people sit somewhere very far from the problem they want to solve. They decide what the money should go for, what the problem is, and what the solution should be. J-PAL does randomised controlled trials on the ground. They get the data from the specific community they are working in. Then they use that data to inform policy. The evidence travels up, not down.
What I find important about that approach is that it takes seriously what is actually happening in that specific context. Not what the theory says should happen. Not what worked somewhere else. What is true here, for these people, in this situation?
I think about waste management the same way. The projects that work are the ones that start by listening. The ones that don’t are the ones that design the system first and try to fit the community to it afterwards.
You are currently running a three-year project on the Guanabara Bay in Rio. Can you describe what the community-first approach looked like in practice?
One of the first things we did was a survey. 700 people from the three communities where we are piloting the project in communities around the Guanabara Bay, because the whole point is to stop plastic waste from entering the bay.
The survey covered who these people are, what their habits are around waste, whether they already sort, what kind of reward would be meaningful to them, and whether they use smartphones. We are building a reward system for plastic waste brought to the collection points, and we needed to understand what would actually motivate people to use it. The answers came from the survey, not from what we assumed before we arrived.
All of that data is public. We built a dashboard with the answers. It guided the design of the entire system, where the collection points go, how the rewards work, and what the training covers. The project now runs training on social-emotional capacity building and entrepreneurship. The goal is that when the project ends, the infrastructure remains and the communities can see themselves as agents of change. Because they are the ones who understand the problem best. They are the ones who should own the solution.
This is also what matters in global development more broadly. Not that you do an intervention, but that the impact remains after you leave. Otherwise, you have built something for them, not with them.
BFS turns ten this year. You were there for half of it, and for a good part of that, you were close to the centre of how decisions got made. What changed most?
One thing that has not changed is the desire for actual impact on the ground. Not reports that will not be used. Not work that exists only on paper. Projects that solve real problems where they are actually happening. That was one of the things that attracted me to BFS when I joined. I am glad it has not shifted.
What has changed is focus. In the beginning, BFS was excited about many things and wanted to be everywhere. Now we are specialising. Going deeper into specific topics rather than wider across all of them. And I think this is important, not just commercially but intellectually. When you focus, expertise actually grows. You develop a depth of understanding that breadth does not produce. I think this is one of the most important things that happened in these five years.
What is the one lesson you want readers to take away?
There is no “away”.
When we say we are throwing something away, we mean we are moving it somewhere we cannot see it anymore. The waste does not stop existing. The contamination does not disappear. The problem relocates.
I studied economics, and I always found it strange that we built an economic model that treats the environment as an externality, something outside the system, an accounting note rather than the condition that makes everything else possible. We operated inside that model for a very long time. We are still mostly operating inside it. The linear model, extract and dispose, is still the default.
What is changing, slowly and not fast enough for the actual pace of the climate emergency, is that legislation is forcing a more circular view. Where legislation changes, behaviour changes. That is the pattern.
But the deeper lesson is not about regulation. It is about worldview. We cannot solve these problems with the same framework that created them. The climate crisis is affecting most severely the communities that were already most vulnerable. That is not a coincidence. It is the same logic, extended. The social problem and the environmental problem are the same problem.
Everything is connected. There is no throwing away. Designing waste systems that take that seriously, as a fact rather than a philosophy, produces different results from designing them as if disposal were a destination rather than a transfer.
Key Takeaways; Ana Carsalade on Evidence-Based Development, Community-First Design & Circular Economy
- Evidence-based design inverts the logic most development projects use. J-PAL’s approach starts with data from the specific community, then uses that data to inform design upward, not the other way around.
- On the Guanabara Bay project in Rio, 700 community members were surveyed before any design decisions were made. The results shaped collection point locations, the reward system, and training content. All data is public.
- The global economy is currently only 7.2% circular. 92.8% of all materials extracted are eventually lost. That is a design and policy gap, not a technology gap.
- The social and environmental problems in waste management are the same. Communities that contribute the least to the climate emergency are disproportionately exposed to its consequences. Programmes that treat them as separate tend to produce solutions that work for auditors but not for the community.
- There is no “away”. Waste does not stop existing when disposed of. It relocates. Waste systems designed around that fact produce structurally different outcomes from those designed as if disposal were a destination.
“There is no away. We cannot solve these problems with the same worldview that created them. We really need to shift.”
Ana Carsalade, Consultant, BlackForest Solutions
About the series: BlackForest Solutions turns 10 in 2026. Instead of a celebration, we asked 10 team members to share the hardest lesson from their hardest project. This is what a decade of doing the work actually looks like.
About Ana Carsalade: Ana Carsalade is a Consultant at BlackForest Solutions GmbH. She holds a Bachelor’s in Economics from PUC Rio de Janeiro and an MSc in Environmental and Energy Planning from the Federal University of Rio de Janeiro (UFRJ). She completed the MIT J-PAL MicroMasters in Data, Economics and Development Policy in 2021. She is a Youth Climate Leaders Fellow. Before becoming a consultant, she spent two and a half years as Technical Assistant to the CEO at BFS. She is currently leading a three-year community-based waste management project on the Guanabara Bay in Rio de Janeiro, funded by Germany’s Federal Ministry for the Environment (BMUV). She has been with BFS for five years.
Key Concepts
J-PAL: The Abdul Latif Jameel Poverty Action Lab, a global research centre at MIT that studies what works in development by running randomised controlled trials on the ground and using results to inform policy upward. The method starts with evidence, not assumptions.
Randomised Controlled Trial (RCT): A research method that tests the effect of an intervention by randomly assigning participants to a programme or a control condition, allowing researchers to measure causal impact rather than correlation. J-PAL uses this approach to understand what actually changes behaviour in specific communities.
Community-centred design: An approach to programme development that starts by listening to the community that will be affected, before any design decisions are made. The Social Recycling Shipyard project on the Guanabara Bay used a 700-person survey as its first step, not its last.
Marine litter: Waste that enters water bodies through rivers, drainage systems, or direct disposal. Plastic is responsible for approximately 60% of all marine debris globally. The Guanabara Bay project is specifically designed to interrupt the flow of plastic waste from surrounding communities into the bay before it reaches the ocean.
Linear economy: The extract-produce-dispose model that has characterised industrial production for most of the twentieth century, treating raw material extraction and waste disposal as external to the economic system. The linear model treats disposal as a destination. It is not.
Circular economy: An economic model designed to keep materials in use for as long as possible, eliminating the concept of waste through design, recovery, and reuse. Requires a change in how production and consumption systems are architected, not just better waste treatment at the end.
Externality: In economics, a cost or benefit that affects a party not directly involved in a transaction. Environmental damage has historically been treated as an externality, meaning it was not priced into the goods and services responsible for generating it. Ana Carsalade’s point is that treating the environment as an externality is not a technical accounting choice. It is a worldview.
FAQs: Evidence-Based Development, Community Design, and Circular Economy Systems
Q1. What is the J-PAL methodology, and how does it apply to waste management?
J-PAL stands for the Abdul Latif Jameel Poverty Action Lab. Its core method is the randomised controlled trial: testing an intervention by randomly assigning participants to a programme or a control condition, so researchers can measure causal impact rather than correlation. The approach inverts the standard logic of development projects, which typically decide what the problem is from a distance and design solutions downward. J-PAL collects evidence from the ground in specific communities, then uses that data to inform policy upward. Applied to waste management, this means designing collection systems, reward structures, and community training based on what surveyed community members say actually matters to them, not on external assumptions about what the community needs.
Q2. What does community-first design look like in a waste management project in practice?
On the Guanabara Bay project in Rio de Janeiro, the first step was a survey of 700 people across three communities before any system design began. The survey covered waste habits, attitudes to sorting, smartphone use, and what rewards would be meaningful enough to change behaviour. The results shaped the location of collection points, the design of the reward system, and the content of the community training programme. The data was made public through a dashboard. None of the design decisions that followed were made independently of that evidence. This is the practical difference between community-first design and consultation: the evidence from the community shapes the architecture, rather than being gathered after the architecture is already fixed.
Q3. Why does systems thinking matter in a circular economy?
Because the alternative is fixing individual components while the overall flow continues to produce the same problem. A circular economy framework is not primarily a recycling technology decision or a regulatory compliance requirement. It is a design question: how does the system as a whole keep materials in use and prevent loss to the environment? Ana Carsalade’s point about “no away” captures this directly. Waste does not stop existing when it is disposed of. It relocates. Systems that take that fact seriously, by designing for recovery and reuse from the outset rather than managing the waste generated by a linear process, produce structurally different outcomes. Technology can improve individual steps. Systems thinking changes the steps themselves.
Q4. What is the Social Recycling Shipyard project on the Guanabara Bay?
The Social Recycling Shipyard is a three-year project run by BlackForest Solutions, funded by Germany’s Federal Ministry for the Environment (BMUV), working with communities around the Guanabara Bay in Rio de Janeiro. The bay receives plastic waste from surrounding informal settlements and urban communities. The project is designed to interrupt that flow before the waste reaches the water by establishing community-level collection points with a reward system, and by running social-emotional capacity building and entrepreneurship training so that communities develop local ownership of the solution. The design was driven by a baseline survey of 700 community members. The objective is that the infrastructure and the community’s capacity to manage it remain after the project formally ends.
Q5. What is the difference between the linear economy and the circular economy?
The linear economy extracts raw materials, produces goods, and disposes of waste at the end of a product’s life. The economic system treats disposal as a destination. The circular economy is designed to keep materials in use for as long as possible through repair, reuse, remanufacturing, and recycling, so that the concept of end-of-life waste is designed out of the system rather than managed after the fact. The Circularity Gap Report (2024) estimates the global economy is currently 7.2% circular, meaning 92.8% of all materials extracted are eventually lost. The gap between where the global economy is and where a circular economy would require it to be is almost entirely a design and policy question, not a technology question.
Q6. Why is the social component inseparable from the environmental component in waste management?
The communities most affected by inadequate waste infrastructure are consistently those that were already economically and socially marginalised before the environmental problem arrived. The Guanabara Bay communities are an example: communities around the bay live with the consequences of plastic waste in a way that wealthier districts with full waste collection infrastructure do not. The climate crisis follows the same pattern: the communities contributing least to the problem are suffering its effects most severely. Ana Carsalade argues that the social and environmental problems are the same problem, not two separate problems that happen to overlap. Programmes that treat them as separate miss the point and tend to produce solutions that work for the auditors and not for the community.
Q7. What should organisations building waste programmes in new markets understand about evidence-based design?
Start by listening. Not consulting, listening. There is a practical difference: consultation happens after the design is already formed, to gather feedback on something that will largely remain as designed. Evidence-based design happens before the architecture is fixed, and the evidence genuinely shapes the outcome. The J-PAL approach provides a rigorous framework for doing this at scale, through randomised trials that can separate what actually changes behaviour from what appears to. But the principle is simpler than the methodology: the people closest to the problem understand it best. Programmes that begin by learning what they do not know about the specific context tend to produce systems that work. Programmes that begin by assuming they already know what the community needs tend to produce systems that communities do not use.