Digital Green: Alesha Miller and Eric Firnhaber on Starting With the Business Model

Lissa Harris: If you could introduce yourselves and tell me about your work and the problem that you’re tackling and what makes you unique in the way that you are approaching it.

Alesha Miller: I’m Alesha Miller. I’m chief strategy officer of Digital Green, an organization that is co-creating a world where farmers use data and technology to build more prosperous communities. We are a nonprofit that spun out of Microsoft Research India about 17 years ago. The original challenge that Microsoft Research was looking at was figuring out how digital technology could improve the lives of small-scale farmers.

There are about 570 million small-scale farmers in the world. Most of them farm on less than two hectares (about 5 acres), live at or below the poverty line, and they have a lot of needs. The original solution that Digital Green landed on spinning out of Microsoft Research [that] was quite elegant and simple, and is not considered very high-tech anymore, was video. We created videos by and for farmers explaining best practices in their local language, in their local setting. Then those videos were screened in community groups where they use the power of peer influence and learning to increase farmer adoption of these better practices.

Over the course of those 17 years, we did not stop asking the question of what digital technology could do for farmers. As digital technology kept evolving, so did we. We kept testing new solutions. We were never satisfied, even though the videos were quite powerful. We’ve now reached almost eight million farmers, more than half of those are women. Video is very powerful. But it can’t solve all things for these farmers. It could only deliver one key practice at a session.

As digital technology kept evolving, so did we.

– Alesha Miller

Over the last couple of years, especially with the advent of generative AI, we’ve been able to incorporate that into a lot of different bodies of work that we already had. [We] see the power that generative AI has to address the needs of farmers, which are hyper-local, dependent on weather, different languages, different challenges and trade-offs farmers are making. Our current work is really approaching smallholder farmers with a solution that is meant to be their pocket coach. It’s an AI solution in their pocket that responds in voice and text with photos and soon [will be] able to use videos and answers in a language that they speak with precise locally relevant information.

Our vision is that we will eventually have a pocket advisor that’s helping the farmer adapt to climate change and adapt to markets. [So that] every farmer in the world is using the power of digital tech. Eric, over to you.

Eric Firnhaber: I’m Eric Firnhaber. I’m our senior director of communications at Digital Green. I think the only other add-on piece … is mobile phone technology. We take it for granted in America and the Western world. I live in the San Francisco Bay Area. I always joke that the big tech companies suck up all of our bandwidth and cell phone service because, on some days, I’m like, “Why is my internet so slow? It should be the fastest of anybody,” but that’s when the pipes get clogged. I would add that mobile phones have allowed us to experiment in a lot of ways over the past several years and [let us] come to this mobile-first solution using generative AI.

I always like to say that the original video that we did, it was pre-YouTube. Rikin Gandhi, (co-founder and CEO of Digital Green) would show videos to people with literally two car batteries hooked up to a projector and a sheet on the wall. The projector was like your old classroom one that looked like it was the size of a record player. Now, farmers share the videos over YouTube, and we don’t even have to be involved in it, which is amazing.

That’s unlocked how we get farmers better information about their farms, access to services, markets, different ways they can grow their crops. …In the old model, even [if it was] a video … [it was] one modality of an extension agent going to them. We realize farmers have a lot of different ways of interacting in the world. It’s not just the mobile phone. It’s other community groups. It’s other farmers. It’s going to market. It’s talking to buyers [and] input agents. 

[Using] mobile [as the] solution allows us to think about what are the other touch points that the farmer has–or a farmer has in a given context–and how we can have our solution or the best information delivered to the farmer where it’s most natural for them as opposed to having to push for one specific avenue that may not be most effective as well. [We don’t have] just a single app. It can be an ecosystem provider depending on who already has a really effective inroad to getting information to farmers. It has a lot of potential, it’s a lot of vision and things to be worked out, but that’s where we see all this going, even beyond AI.

Lissa Harris: Can you talk about the communities that you serve? You mentioned you were in India, but what communities are you in and how do you reach people? 

Alesha Miller: Our key countries are India, where we were founded; Ethiopia, which is our second largest footprint, Kenya, and Nigeria. In Ethiopia, we’re working in four states right now, about six counties in Kenya and two states in Nigeria. 

Eric Firnhaber: [We are in] 10 states across more Northeastern through Eastern and Southeastern India.

Lissa Harris: Where is it working and why?

Alesha Miller: Let me give you maybe a couple snapshots. Then, I’d love for Eric to jump in. He just got back from Ethiopia so he’s going to have some interesting insights. These communities have wildly different access to mobile phone technology and the internet. Also, the gender divide is more or less significant in spaces across these geographies.

For instance, Andhra Pradesh is highly connected with very good mobile phone penetration, good access to cheap internet, and the digital divide, even among women, is not as bad as in a lot of other places. In Andhra Pradesh, a couple years ago when our team was visiting, they were actually showing us the Pico projector that Eric mentioned as if it was part of a museum. It was on a shelf and they said, “Oh, we used to watch [videos] on this and now we all have smartphones and we watch it on our phones on YouTube and we share on WhatsApp.”

In a community like that, you can directly onboard farmers onto Farmer.CHAT. We go through trusted networks that we’ve been working with for a long time. We go through self-help groups like village resource people within these self-help groups, extension agents who are part of the Ministry of Agriculture, Ministry of Horticulture, and they help  onboard leaders in their community first, and then they onboard additional people.

We’ve used things like leaderboards to incentivize competition between people who are onboarding. In a place like Andhra Pradesh, you can do direct onboarding. That said, we noticed in our early days that there’s a pretty big difference in digital confidence levels. It may actually take more time in person with groups of women to onboard them and get them to feel confident to ask a question, to do a follow-up question. It doesn’t necessarily look the same for every kind of farmer.

In a place like Kenya, you also have good mobile phone penetration and almost no gender divide when it comes to access to mobile phone and internet. In Kenya right now, more so than India, we have digital campaigns, not just in-person groups like I’ve talked about in farmer groups and extension workers. Eric’s group did a lot of the work on that, so he can speak to using social media campaigns to raise awareness. You can cheaply onboard a lot of people that way.

But in a place like Ethiopia, there are very low mobile penetration rates and very few spaces where women can actually participate in public as part of an agricultural group. If you’re a widow, you can join a group as your household head, but otherwise, you’re probably not in a farmer group. We’ve had to establish women’s groups where we can distribute content.

In India, Kenya and elsewhere, we’re also facilitating access to Farmer.CHAT. When people are getting together for a video screening, which we still do, or for some other reason like a savings co-op, then somebody who does have a phone and is using [Farmer.CHAT] will answer questions from across the group in that moment, or send content later in a WhatsApp group by screenshotting content. We have a lot of facilitated access through these trusted intermediaries that are already extension agents, lead farmers, and farm organization leads. 

Eric Firnhaber: Those are all good highlights as far as these indicators that we use as a rough guide of what kind of approach will work best and what kind of stage of maturity on a digital spectrum an organization might be or a group or a farmer user might be in. A lot of that, as Alesha mentioned, is mobile phone penetration, connectivity, digital literacy rates.

I think the telling example is from Ethiopia that was very personal for me. I was there about two years ago and we were in a rural area and there was zero cell phone coverage for whatever data network that my phone was trying to connect to. Then when I could connect to WiFi, it was very slow, very spotty. When I was in Ethiopia [recently] the 5G Wi-Fi was great. It was a completely different experience. In a place like Ethiopia, they have one telecom provider, Ethio Telecom, that’s government-owned. That’s it. You can’t even pick between two or three different ones, so you’re at the pace of government intervention, government progress in terms of some of those high-level macro indicators.

The other telling thing about what we’ve seen in Ethiopia is that to those farmer groups, especially in the women’s groups, we had to, in some cases, give them a phone because they didn’t have one because of lack of confidence, lack of resources, or just cultural norms. This time, in a group of 20, half of them had some kind of mobile phone, whether it was just voice only or feature, and half of those people, so about 5 out of 20 had actual feature phones.

Farmer.CHAT is primarily an app, so you need a feature phone that can download an app. It makes us think about, for these other potential users, without a feature phone,  how do we reach them? How do we get this into their hands? Whether it is, like Alesha mentioned, working through [facilitated] groups or other things. [For example], agriculture in Ethiopia is government-controlled and they have a fairly robust government network of call-in centers, an 808 hotline, that people can call in to. It’s basically an extension agent hotline. 

It’s a little bit more effective than on-the-ground extension agents because you’re calling as opposed to waiting for them to come to you or you having to travel to them. They still have the same body of knowledge effectively. If they don’t know an answer to your question right away, they will say, “Call back tomorrow or next week after I do some research on it.” They’re saving the travel time of that individual.

One thing that we’re thinking about in Ethiopia, which we’ve started piloting in Kenya, is finding call centers like that and equipping the call center agents with Farmer.CHAT. Instead of, “Oh, I’m not quite sure if I know the answer. I don’t want to give you the wrong one, so let me go do some research and get back to you,” they say, “Yes, let me check that out.” They can use an AI chatbot trained on agronomic data for those climate conditions and based on the value chains and they can give a faster response to people who call-in to people who don’t have feature phones.

It’s funny because it was two months ago that OpenAI came out with a 1-800-ChatGPT and we were all chuckling, like, “Oh yes, we’ve been trying to figure that out for rural farmers in Kenya for months now.” It’s an interesting way, again, to get to where those farmers are and get the information there.

Lissa Harris: Can you share an example that illustrates the impact of the work that you do? How do you know that what you’re doing is working or not working? What’s that evidence that you’re looking for?

Alesha Miller: Our ultimate impact goal is income. We want to see changes in farmer income. We want to see increases in resilience of the household, and we want to see increased agency or perceived self-efficacy. We want to see someone feel like they actually can take a decision that will have an impact on their life because we believe that leads to change.

[In the shorter term], we’re looking for adoption of an improved practice. In this new world of using a mobile app, we look at active use, are people actually using it? Within the data itself, we are looking for stories that show problem resolution. 

Our chief product officer often uses this example of a real conversation where a farmer’s saying, “My baby chicks’ feathers are falling out. What could it be?” Then they get an answer saying, “It’s likely that it’s this disease, you probably need to go get this vaccination or medication.” Then you see a series of questions that relate to how to use that medication. Then a few days later, it actually switches to broader health of the flock and then thinking about marketing solutions and then thinking about other questions. It shows this arc of problem resolution.

We think that that’s what we need to find on a larger scale, figuring out how to query the data to [find that arc]. The way we do it right now is through focus groups where we ask for those kinds of stories. We have also done third-party case studies. One third-party study that we did last year showed in a 90-day period, frontline workers were saying they had a 57% increase in quality of life, which is a shockingly high amount. I was surprised at that level. They attributed that primarily to time savings and increased confidence and motivation.

Sometimes what constitutes success is a little different than what we think, even though those make sense if you’re a frontline worker serving 250 or 500 farmers. We are looking at adoption at scale, usage at scale. We’re trying to figure out essentially how to get farmers to tell us if their problem was resolved, but that is easier said than done. Just like you and I, people don’t want to answer extra questions in their app so we’re trying to figure out how to gamify that.

There are stories about farmers asking questions about crops that they’ve always been curious about, what would it take for me to plant this, and then actually taking steps to do that kind of thing. Farmers talking about how they were able to improve the health of a cow. We’re looking at micro changes that lead to income resilience and agency.

Eric Firnhaber: One big thing for us is defining success has actually changed for us. We’ve historically been a fairly traditional development project organization where we have a four-year grant that has milestones and some impacts. We do a big 6 to 12-month study at the end of [the grant] to figure out if it did the thing we wanted to do or not.

We still have a lot of donors and projects like that, but as we’re moving into more of a product orientation with the app and there are new impacts that we hadn’t anticipated. Our first thought was, “Is Farmer.CHAT driving income changes?” But that is hard to measure. Along the way, there’s a lot of intermediate and potentially more impactful benefits such as quality of life or feelings of empowerment that are self-reported.

It’s not the 12-month randomized control trial (RCT) study that typifies this type of development work. It’s user focus groups, it’s more product-oriented or tech company-oriented ways of evaluating effectiveness of a product that we’re trying to learn ourselves. We have to figure out what those metrics are, what progress is, and then also how we explain to donors because it’s new to them as well in terms of them asking “Where’s the big RCT?”

We’ve been doing this for six months and we’re still working on getting people to stay on the app for more than two months, figure out how it works, while also trying to roll it out. It’s a whole new landscape in terms of what impact actually means, and we’re finding that it’s much, much wider range than we first thought possible.

Lissa Harris: Sometimes you learn as much from things that don’t work as things that do. Is there an example of something that you tried that really didn’t work out, but that you learned from?

Alesha Miller: I’ll give you an example that sticks out for me that was a big learning [experience]. It’s from a few years ago, but I think it still is relevant. We are concerned and sensitive to data privacy and thinking about data ownership, data stewardship, and who manages farmer data. We know it’s a powerful asset and it’s even more powerful when it’s aggregated with other farmers, such as a farm organization.

A few years ago, we were working hard on how we could set up backend software to ensure the security of data while also enabling good sharing of data across networks to get services. We decided to do a pilot essentially on data rights, trying to communicate in a simple video-based format, our history, what data rights were among farmers, and then how we were using their data within a farm organization that we were working with. We had given the farm organization some software and we recorded a two-and-a-half minute video explaining in a cartoon way what data is, what you’re giving access to, for what reason, how could you revoke access if you didn’t want to give that access anymore.

It was shown to them by someone they already knew and trusted. We were checking with an actual study for comprehension and also going back a few months later to see if they still recalled that experience and knew what it was about. We found that it seemed like farmers were understanding at the moment, but … then I don’t even think 50% of people [from the comprehension study] actually understood what they were seeing, even though we tried to simplify it down to the bare bones.

What was interesting to me is [when we] went back three months later and asked people about it, if they remembered it at all and they had a negative reaction to it. They were scared that something was being done with something valuable, but they weren’t really sure what. Instead of empowering people with a message about all the choices they had to manage their data and benefit from their data, we actually freaked people out a bit and confused them about a concept that really didn’t mean much in their daily life and so we just stopped it.

Honestly, we realized it was not a useful approach. We still have some basic data privacy and consent, but now we’re not going to the same lengths as we were then because we realized it’s just not serving the purpose we want. It’s more important to focus on the people who are leaders within farm organizations who are making choices about collective data assets, who maybe have a bit more training on what I’m doing when I share an Excel spreadsheet with a bank or something like that. 

Eric Firnhaber: I have an example on the acquisition side of things and rolling out of products. In Kenya, for example, we’re thinking about a couple of different channels. Alesha mentioned the digital one. But we also have a number of in-person channels to drive downloads of the digital app. We tried a couple of different farmers groups and extension agent networks. There’s one that distinctly did not work even though we thought it would be really useful or powerful. It was working through input suppliers. We had this hypothesis that every farmer needs seeds, fertilizer, pest control chemicals, things like that, or organic methods. We thought, “Let’s have promoters or influencers and these agri dealers have QR code[s] or talk about how great Farmer.CHAT would be to these farmers in an effort to get them to download it.”

We figured if we can’t get people through other channels, at some point during the harvest season they have to go through this channel and get people before the harvest or before even the cycle would start. We ran that for a couple of weeks and we found that, especially compared to all other channels, it was the lowest performing and actually drove a lot of negative interactions in Farmer.CHAT. People would download it and then immediately have a negative thumbs down moment or give very short negative feedback on it.

We did focus groups several weeks later with users that we knew had come in through that particular channel. The short of it was that they didn’t trust it because it came from the agri dealer. They said, “Well, I assume that this is just like an advertising thing, that they’re going to just promote their products or that their product’s going to be overpriced in this app and they now have my personal information that I didn’t want them to have.”

It wasn’t just a not-good channel, it was actually a negative channel to drive usage of the app. That was quick learning to say, “Cut that off. We’re not doing this channel anymore and put in the other ones.” There’s a lot of quick learning as well, which is very natural to the tech world and we’re trying to figure that out in these very different environments.

Lissa Harris: Switching gears toward the support that you’ve received as an organization. Could you talk about something that was surprising or helpful to getting things to scale. What kind of support have you gotten that’s made a difference for you and how did you know?

Alesha Miller: We have one donor, who as a part of their due diligence process provides an output that is essentially an investment brief on the organization that they’re  funding. They do interviews with lots of questions and they interview different parts of the organization, [including] finance strategy programs. They have their own way of calculating return on investment or value for money of the grant that they’re planning to give, or PRIs as well.

They make the pitch on your behalf because they’re writing it. It’s the output of the due diligence, and it is used by the principal funder of the organization that requires this as part of the foundation. The bigger use, I think, from what I’ve heard from them and from what I’ve observed, is they then circulate that investment brief to other funders that they work with who are in their broad network of trust. Then that often actually leads to funders reaching out to those grantees to say, “Yes, I saw the brief on your work. I’m really interested. Can we talk?”

We’ve received additional funding as a result of their investment brief about us because they advocated on our behalf. Not only that, there’s something quite remarkable and refreshing to having someone else try to tell your story using their own words. I think in a nonprofit, maybe anywhere, you get into your own language and your own ways of talking about things, and you think that that’s the logical way that people will understand until someone else writes about it differently and you go, “Oh, actually that is what we do. That’s a really interesting way to say it.”

That asset that they’ve created has unlocked a lot of opportunities for us. Every year, it gets refreshed with them. It has a section about what we’re doing and our financials. Then it’s the things we’re watching, which might either be cool things that we’re doing,  or things that they’re worried about, like end risks. That same funder–they’re not the only ones other donors have done this too–also have quietly benchmarked us in the sector. I remember many years ago talking to them and they said, “You guys are really underinvested in BD [business development] or fundraising. You don’t have enough BD people. Similar organizations to yours have X number of people working on proposals and they have this many finance people.” 

Because they deal with so many organizations and they do this kind of deep due diligence, they were able to help us understand where we had underinvested in ourselves and they were able to see the effect of that underinvestment. They gave us unrestricted funding, but part of their recommendation was, “If we were you, we’d be investing in another finance officer and make sure we had another fundraiser.” We would have no way of looking around at peers to know what the insides look like, and having any sense of what our structure should be without a funder like that. 

Lissa Harris: Could talk about what role trust plays in your relationship with funders and how can a funder cultivate trust?

Alesha Miller: One of the things about Rippleworks that I think has been really amazing is that they have done a lot of that. They’ve done a lot of trust-building work with us. We’ve had multiple engagements with them and lots of different parts of our organization have had engagements with different teams within Rippleworks.

[Rippleworks has] done a lot of trust-building work with us.

– Alesha Miller

They’ve also done a lot of support work for us. Offering longer-term neutral advice. A sounding-board place to go where they’ve said, “If you need advice about how other organizations are doing this, come talk to us. “ I remember talking to them when we were getting ready to do our strategic plan and I said, “I’m thinking about writing the plan myself instead of going outside and getting external help.” They said, “Oh, we can connect you to an organization that just did that. You want to talk to them?” That connector role where funders demonstrate genuine curiosity in helping you to solve your problem–or just understanding the problem–even if they can’t solve–is a big way to build trust.

I appreciate when funders are transparent about whatever the process is going to look like, [especially] when you’re in a new funder relationship. I had one organization who said, “I always tell every organization I talk to, what the odds are that they’ll get funded.” He actually assigned odds. He said, “About 50%. If you stick with it, maybe 50% odds.” He [said], “If I tell people less than 10% and they still engage, that’s on them.” Organizations that give you clear milestones, such as it’s going to take six months to get through this process and then X will happen. Anytime you can just have transparency into the relationship, that’s really meaningful.

The other big one and one that we are constantly just so thankful to our supporters for is how funders respond to change. We are constantly having to navigate change. We’re now working in the world of AI, which changes every five minutes. Something that we write in a proposal that isn’t approved for six months so by the time we start implementing it, it’s probably out of date. Some funders have a high tolerance for that, as long as you communicate carefully and frequently, and some funders don’t. I think that you get more trust and you probably get more transparency when you demonstrate that you have a genuine curiosity about why things changed. As long as you’re aligned on the outcomes and the values, then it’s fine. Demonstrating willingness to weather change collectively, I would say that’s another one.

Eric Firnhaber: On the trust issue, another question is what can nonprofits do to better build that trust on the other side as well? If somebody gives us money and it’s unrestricted, that’s pretty quick trust, but what can we do as an organization to build that trust with others without lengthy diligence? We have one funder who is very discreet about the timeline and very effective. We have other funders who give unrestricted [funds], but it takes a year of back-and-forth conversations every two or three weeks. By that time, we’ve spent quite a significant amount of staff time to get to that unrestricted amount. It’s about trust on both sides of it.

Lissa Harris: Do you see bold shifts that are needed in the funding landscape to center the voices of those that are closest to the problem?

Alesha Miller: Eric tipped to this earlier, so I’m going to come back to it. I don’t think that all social change is going to be affected by digital technology. I think plenty of social change work is going to be in person, not mediated by tech. However, in the world we’re in, where more and more people are connected, and by 2030 almost everyone will be connected, even in some of these most rural areas.

Thinking about how we [work with] people who are in the product world for the work that organization does is really important and actually requires change management. In their hearts, product people have a deep, deep respect and understanding of user requirements, user needs, user voice. That should align perfectly with people in the social change space who have a real deep interest in human-centered design and voice. But they use two totally different lexicons. They use different toolkits. They use different measurements. We have seen this in our own organization. You can come into conflict over things where you’re not actually in conflict, you just don’t understand each other or you don’t value the metrics of the other because you don’t know what those metrics are supposed to tell you. I remember the first time I heard Net Present Value, I was like, “What does that even mean?” Or Net Promoter Score. Once you hear, “What’s the likelihood that somebody is going to recommend this to a friend?,” now I understand what that means and maybe how that could be useful.

When digital technology is being embedded in social change work, we have to bring in product people and help them understand the social impact space. Social impact people have to understand the role that product can play and how it can contribute to the traditional toolbox that we’ve had. Same thing with metrics. We need to get more comfortable with caring about monthly active use or lifetime value of a customer. 

We have to have more people who are experts in UI-UX (user interface and user experience) inside of social change organizations. One, hopefully, could fund those kinds of people who are often not cheap. Are funders willing to invest in that is a good question to think through. Second, how do funders get interested or learn about some of these alternative metrics and why they should matter?

Sometimes, reasonably so, funders say, “I only care about change in income.” I care about change in income too, but to do a randomized control trial to assess rural income can be a half million dollars. How often do you want to do that half-million-dollar study versus looking at other interim metrics where people are self-reporting different kinds of behavior, self-reporting changes in quality of life, or demonstrating problem resolution in a digital technology? I think we need funders to get really curious about, what does the social impact organization look like in five years?

I bet most of the NGOs I worked with don’t have product teams, and they don’t have any of the skill sets there. How do the people who work within NGOs look different? How do funders give more flexibility? Can they get interested in some of these alternative metrics even if it’s alongside traditional metrics for a while? I think those are some things that I would love to see.

I come from an international development background. I’m a rural development person. I don’t know all those new metrics. I’m learning them. It’s not to cast out the stuff that we’ve been using, such as theory of change. It’s to say, we have new tools and frameworks. Let’s integrate them. Let’s use what’s best from both sides. 

Eric Firnhaber: This is something we’ve been thinking about a lot given the fast learning that we’ve been doing. I know Alesha mentioned the six-month gap between proposals to funding, but in Ethiopia, for example, they’re just wrapping up a five-year project based on a worldview from five years ago. They want to get into some of the newer technologies that we’re developing, building, deploying, but they literally can’t because of the donors and how the funding is structured right now and their orientation and operational build out there. Faster, flexible funding.

Lissa Harris: Talking about Rippleworks support, you mentioned that you had capacity building and tailored support that was not monetary. Did you also have either a low-interest loan or an unrestricted grant?

Alesha Miller: We did have an unrestricted grant and we’ve had multiple engagements. In fact, Eric was part of one with one of their experts where they do a sprint project with you. We were also part of their venture studio, which is the open-ended leadership support. We meet quarterly with Rippleworks staff who [ask questions such as], “What are the three big things right now that you’re worried about and how can I help you with that?” They also have connected us to resources and we’ve had a lot of different connections. [For example], the person who’s leading strategic comms at Rippleworks, we’ve done some stuff with them so it’s been different monetary and non-monetary [engagements].

Lissa Harris: What can you tell me about the freedom of an unrestricted grant and how it aids your work or helps your ability to scale?

Alesha Miller: It’s super critical. Given what I was saying before about how fast-paced things are, a lot of times you need to try something else or you need to pivot, and you may not have the freedom to do that in a restricted grant, or you need to seize on a particular opportunity. Had we not had significant unrestricted resources in the last couple of years–Rippleworks was part of that and we had others that were new to us because before 2020 we had no unrestricted funding and since we’ve had unrestricted growing amounts–we would not have been able to build a product team, a tech team, an AI team. We wouldn’t have been able to expand to new geographies that we’ve done. We just couldn’t have done those things. I think it’s enabled critical [work] both on the outside and on the inside, like upgrading our accounting system. That’s something that no restricted funder cares about. But it’s a huge inefficiency for the organization to have an outdated accounting system.

Lissa Harris: You mentioned the example of the investment brief. Are there other requirements from funders that have been useful in ways you didn’t anticipate? 

Alesha Miller: That particular funder said the first time we talked to them, “For the next two weeks, we need you to clear your calendar. You’re going to be talking to us every day. We need to talk to a bunch of people in the organization. Then after those two weeks, you’re not going to hear from us for a while and then we’re going to fund you.” It was that clear. Having that annual accountability of refreshing the story is actually a helpful thing for us. 

[Other funders create cohorts.] There’s a really interesting project where we’ve been part of a cohort of people who are adopting AI and also working in agriculture. We have requirements to be part of these cohort meetings, in-person and also virtually. That’s led to a ton of networking and partnership ideas. That’s a cool thing. I think donor requirements around that kind of collaborative stuff have been quite useful.

Honestly, we like RCTs. Eric and I have both been a bit critical of them here, but there’s nothing more powerful to other funders than showing rock-solid evidence that this intervention actually changed something. We have some donors that require them. If they want to pay for them, even though they’re a huge lift, we are delighted that they would like to pay for that. 

Eric Firnhaber: My biggest thing is capacity building, such as a six-month expert residence. It’s the pieces that don’t always seem as obvious to impact because it’s not programmatic. You might consider it overhead; overhead is what gives people good benefits at competitive salaries and we can hire the best talent. 

Alesha Miller: [We went through] a process with Rippleworks more than a couple of years ago that ended up being a really challenging project. I bet if we talked to the Rippleworks folks who were part of it they would wonder if the engagement was actually as high value as they hoped. I was challenged by it. It was around product design.

What ended up coming out of that was the realization that we were really underinvested in the product function and in the tech function. A lot of the things that would be fundamental in a tech organization or true product-led organization, we weren’t able to do yet. It probably led to the accelerated timeline of hiring a chief product officer and hiring other product folks. It peeled back some layers of what we didn’t yet know how to do very well by giving us exposure to a real product professional that helped us see that.

When we would have calls quarterly with Rippleworks folks, they said, “Don’t come with a big agenda. Come and tell us what the big things are on your plate, the big rocks you’re worried about over the next three to six months. Then let’s talk about if we know anybody, can recommend anything to help you with those big rocks.” [That’s] saying let me come alongside you and also not judge you for not having answers to these things.

Three years ago, four years ago now [they said] “You have a weak product function.”  That seems meaningful and important to know and I think we did know it, but we didn’t know the depth of it. [We have] conversations where we’re basically saying, “We’re really struggling with this and this and this,” and it’s nice to be able to say that and know that it’s not going to be a ding against us as an organization. We can genuinely show up with authenticity and vulnerability and they meet us with a problem-solving mentality. 

Eric Firnhaber: One thing is the communications operations and the Rippleworks network that they brought to it. I think they call it the expert network. They really took the time to understand the singular problem that they [were going to] look at, and then they said, “Okay, we’re going to go talk to a dozen or so people and figure out who has the right experience to solve the problem, but also too who’s solved the problem before and then the next two or three steps after it.”

I talked to [a person] who worked at a much, much bigger organization than ours, but also [someone who] worked at that organization when it was smaller and in other smaller organizations. They basically had been on the same path before and could say, “Oh, don’t do that because I made that mistake five years ago.” It’s really helpful to get that kind of specific expertise. The experience ended almost a year and a half ago and we still chat. We still have biweekly calls. I assume, it’s out of the goodness of their heart or maybe they find it fun or something like that. It was incredible to be able to tap into the network that Rippleworks specifically had and just the thoughtfulness that was put into it.

If we’re calling out names, Kyle on the Rippleworks team was phenomenal about finding that right talent. I saw him last year at an event and he asked, “Hey, how is so-and-so that we hooked you up with?” He knew exactly who I was and who the expert person was. He really cared about it. It wasn’t just like a diligence checkbox thing. It was a commitment and trust to build the capacity of the organization beyond saying, okay, when we stop funding you, you’re done. It’s like, no, we want to fund you, but also build your capacity so that when we’re done funding you, you’ve actually grown in your ability to operate as an organization. You can move to the next level.

It was a commitment and trust to build the capacity of the organization.

– Eric Firnhaber

Lissa Harris: Are there gaps that you can see in funders deciding what capacity support you’re going to receive? Are there things that funders don’t understand about capacity building that you think would be useful for them to know?

Alesha Miller: Every funder has almost a specialty that they bring. We get something from each funder and so every funder also probably has blind spots. I’ll give one that’s at the top of my mind because of probably recency bias. We’re all experiencing a big shock in the international development space with USAID (United States Agency of International Development) appearing to be shuttered for the time being with $40 billion in aid being cut off. There’s been a long history of people believing international development shouldn’t be around forever in the way that it is. There’s been a history of needing to think about business solutions to how we create sustainable support for people in different ways.

I was a funder. I worked as a funder for eight years, and now I’ve been working with funders in many different ways. Every funder wants to ask the question: what’s the sustainability model for your organization? Because they also don’t want to feel like they have to invest in you forever. However, I don’t think many funders actually invest in getting you to a business model that’s sustainable, which is a business question. What are the creative solutions that people could think of? How would they incorporate private sector investment, or how they would use innovative finance, or how they would set up a social impact bond?  

I don’t see a lot of funders investing in sustainability models. But I don’t know that there’s a lot of funders that actually invest in moving organizations that are part of a traditional donor-funded model towards something else.

That would be a space that would be really interesting for a funder. To be able to help network organizations together. A lot of organizations are just going to go away, probably, in the next little while or consolidate together because there isn’t enough funding. Maybe there are some creative solutions for how we could find new business models. Maybe there are some funders who could actually invest in that. 

Lissa Harris: You’ve received both money and capacity support. How did the combination or sequence of those work or could they have worked better?

Alesha Miller: I thought it was pretty logical or well-sequenced. We did the capacity-building engagement first, then we got an unrestricted grant, then we did another engagement. Then we got the Human Capacity Building Grant. We got two grants, and now we’re part of the Leader Studio.

In the five years I’ve been here, we’ve had all these different opportunities to interact with different support functions within Rippleworks and they made sense. I felt they were seeing how serious we were, honestly, about pursuing a product vision in that first engagement. How do we approach problem solving? We showed them what that looked like. That seemed to unlock their desire to invest in us. That led to even more opportunities.

Then this last part feels like a real honor in a way that they’ve invited us [to be part of the Leader Studio]. They invited Rikin [Gandhi] as part of the pilot for the Leader Studio. Now there’s this really high degree of trust where I can email Matt Di Paola, (head of marketing and communications) who I met last year. He’s going to be at Skoll World Forum and he reached out to say, “Hey, I’m going to be at Skoll. Is anyone going to be there? How can I help? Can I introduce you to people? Can I show you what events we’re doing?” Now we have all these touch points, and we really feel like they’re in our corner, so it’s a really powerful combo. I don’t have a way to recommend an improvement, honestly.

Lissa Harris: What’s been the biggest challenge in the support that you’ve received? 

Alesha Miller: It is hard for funders to keep up with us. It’s hard for us to keep on message. If you don’t have frequent contact with funders, then our message seems rather jolted. If things change radically in six months and we don’t talk to you every month, then figuring out how to convey all of this change and learning in a way that maintains trust has been a challenge on our side. Sometimes that runs into donor inflexibility and like, “Sorry, what you said you were going to do is X, Y, Z. I’m hearing that you’re not doing that anymore.” That, I think, can be a point of misunderstanding. Our model of dynamic constant evolution isn’t a good fit for every funder. Some funders want to just know, “I’m going to invest this, and I’m going to get this very specific fiscal output.” We’re probably not for everybody in that way.

We have to be able to communicate our change for these very different kinds of funders who have very different levels of knowledge of agriculture or tech or our regions. Some are very deep, some not so much. I think navigating the change environment we’ve been in, plus figuring out how to do that in a way that works with the timeline has been hard. 

Lissa Harris: What advice would you give to funders who want to see social ventures be successful?

Eric Firnhaber: I would anchor back to the flexibility and trust piece. I also was on the funder side for a while. The organization I was at did everything from full, unrestricted grants, restricted grants, to no-cost loans, to full-on impact equity investments in companies. They did all that, but in silos. They never looked at cross-pollination or movement of their portfolio investments. They would say, “Here’s the equity investment that we need to get the return on. We’re not going to think about if this equity investment also wants to do social impact that’s not going to give the market rate return, how do we support them in doing that?” On the other side of it, “Here’s an unrestricted grant and they want to try business-oriented things. We’re not going to fund that with the grant side of things.” 

I like that movement analogy. In the venture capital world or the niche VC that says I’m the Series B+ firm.  I know all the people before me in the pipeline and I’m going to get a company from them. I need to develop this to sell it off to the next person and they’ve got a for-profit return pipeline going on. I wonder if there’s a similar approach that funders can take in the nonprofit sector to ask where do you want to go with your impacts and what does that look like in terms of sustainability and the funding landscape out there. How can we help you adjust or move with our funding to get to an exit moment, if you will, to use that terminology that exists [in the VC world]?

Lissa Harris: Thinking about your efforts to scale and sustain, what are the top three things you think you need to unlock for Digital Green?

Eric Firnhaber: We’ve got a roadmap that we’re defining for onboarding farmers to Farmer.CHAT. It’s based on the amount of capacity and resources that we put towards it [in the past]. We can say, “With this resource allocation level, we can get this number of farmers over time onboarded or using Farmer.CHAT, and now let’s double that.” We doubled it and we saw an appropriate level increase in the amount of return on that for our own internal investment purposes.

We have a working model that shows us return on impact scales with the return on investments that we put in internally. So how do we prioritize the right types of funding that can help us do that in the geographies that we’ve identified with that point of leveraging? We’re realizing … our own internal learning and needing the space and ability to do that so that we run the pilots to learn so that we come [to] a funder with knowledgeable proof to say, “Here’s what we have seen works. Here’s what we know the levels can be. Will you trust us and join us to be able to do that?”

As opposed to “just trust us to do this well.” We’ve been building proof points for the past two years about what works and what doesn’t and the openness and willingness to have those conversations about, “Here’s the incremental stage-gating that we can do, and we’ve seen work, and here’s what we can do with the next level.” 

Lissa Harris: This is an interesting answer. How do you know the money is working? What are the proof points? What are the steps that you need to take internally along the way? 

Eric Firnhaber: Yes we’ve done that for the past two years. We have internal maturity models for the product in a given community. We can say, “.. if we can reach 100 people and they use it X number of times, we consider it to be an early-pilot stage. Then when it crosses that threshold, we know that it’s going to require this amount of internal resources to sustain it and then if we want to grow it to the next level, it requires additional resources.” It helps us create internal prioritization.

Even if budgets are not increasing by certain amounts, we can make trade-offs. “This community here or this work, we’re going to sustain it. We know it takes that much. This one over here, we want to grow. Instead of growing both, we have to make a choice so we can make a better choice.”

Lissa Harris: Thank you. It’s been really eye-opening talking with you today.

Lissa Harris is a freelance reporter, science writer (MIT ’08), and former local news entrepreneur based in upstate New York. She is currently working as a consultant on capacity-building and local solutions-oriented community projects in the rural Catskills.

* This interview has been edited and condensed.

View More Interviews