The 3 key principles to create social impact at scale

There has been an incredible pace of progress over the last two decades in Silicon Valley, resulting in amazing innovation even for the most mundane needs. Yet in tackling some of the big challenges we face around the world, we have not been able to keep up.

We haven’t been able to innovate, drive change and track progress nearly as fast.

Problems are more dynamic and scaling more quickly than solutions. This means that if entrepreneurs don’t take an innovative approach to creating social impact, we’re likely to fall behind.


When it comes to social impact, innovation has been fairly misunderstood. Often more focused on vanity metrics than on sustainable impact at scale. We need to be clear about what we mean by innovation.

To put that in context consider the following quote:

“Genius is 1% inspiration and 99% perspiration.

Thomas Edison

Within that framework, innovation is 1%. It’s the big idea, the invention. But the core of innovation is 99% perspiration. It’s taking that good idea through the entire process that goes into achieving massive scale. 

Testing, iterating, improving, building the infrastructure and building the business model.

Companies like Google and Facebook are considered among the most innovative on the planet. But Google didn’t invent search and Facebook didn’t invent social networking. They were not the first. We don’t refer to them as innovators because they came up with an idea that no one had ever thought of before. 

They’re innovators because they took those preexisting ideas and they continually improved on them. By modifying the user interface, the feature set of algorithms; and then grew far above their competitors. 

That’s what we need to think more about when it comes to innovation. 


The book Lean Impact is inspired by the Lean Startup. Similarly, Lean Startup is a book inspired by the best practices of how innovation happens in Silicon Valley. It’s the entrepreneur version of the scientific method.

The idea is that when you’re working under conditions of great uncertainty you need to take a different approach. These conditions exist, for instance, when you are starting a tech startup company. Trying to build a product that no one has ever built before. Or in the social impact space, where we are trying to solve long-standing intractable problems that we don’t have sufficient solutions for yet. 

So if we don’t know what is going to work, we need to run tests to find out what a viable solution is. Our goal should not be to deploy as much as possible, but rather to learn as quickly as possible.

There’s a big difference in both approaches.


In the social sector in general, we tend to think based on constraints. For example, nonprofit specialists would often find themselves discussing how much budget is available, the number of volunteers supporting a programme, a time-limited framework. And they’d ask themselves “What can we do with that?”

One of the important mindsets shifts that need to happen is moving from a constraint-based planning approach to a need-based planning approach.

We can start by discussing instead; what can we do that would potentially move the needle on the problem? How can we scale to make a difference into the depth of impact that will make an even greater difference?

The Sustainable Development Goals

Setting the Sustainable Development Goals (SDGs) was an important move. As a global community, we’ve come together and have said: “This is the world that we want to live in, we are going to create 17 different goals towards making that our reality”. 

But as we all know we’re 3 years in and we’re not on track. We’re falling short on many of the SDGs.

Specialists talk about the 2 to 3 trillion dollars annual gap in funding that’s needed to work toward the SDGs and certainly, we should try to get more funding. But it’s highly unlikely that we’re going to have access to that financial funding before 2030.

And yet an element we don’t focus on enough is that we need simply better solutions that can give us better results for the investments made.

We’re not going to close the gap with money, but with innovation.

Over the last 30 years, we’ve built slow but steady initiatives on some of the 17 goals including access to clean water, electricity and sanitation. We assume that it’s hard to do better than that because we’re working with people that are very poor, live in remote areas and therefore it’s hard to access them. However, during the same time, the adoption of mobile phones skyrocketed among the people living in the same rural areas.

So what can we learn from this example? 

First, mobile phones were something people wanted. People would choose mobile phones before they’d choose water, electricity and sanitation in many cases. It’s a real need for people, they understood why it was helpful to them and they demanded it.

Second, there was a business model behind it. People would pay for the phone service in general. This enabled companies to continue investing both in developing better and cheaper products and in distributing the products extensively. 


Consider the case of social enterprise VisionSpring. They wanted to focus on a 700-year-old innovation that has been proven to improve productivity and learning potential. 2.5 billion people that don’t have it could benefit from it: eyeglasses.

It’s not some new fancy innovation. We know how to make eyeglasses and yet 2.5 billion people still don’t have it. 

VisionSpring did what a typical startup would do. They decided to start in a few geographies: El Salvador and India and they set up ‘vision entrepreneurs’ to go out to rural villages and offer eyeglasses to people. They decided to sell them for a small fee to ensure that it was something people wanted, and to have them participate in the process. 

Eventually, they started seeing the results of their work. They had several success stories of people’s whose everyday lives had improved because they were now able to see better.

Many organisations choose to stop there. To focus on the success stories of their intervention and set minimum target goals per year, and potentially continue expanding into different geographies. But that’s not a scalable way of delivering eyeglasses to 2.5 billion people. 

VisionSpring recognised that they were losing money on every pair of glasses they sold and that they were not going to grow fast enough, because they’d be constantly relying on grant funding.

Their first pivot was to shift to a hub-and-spoke model. They set up shops in urban areas to serve more affluent communities with higher-end products. With the profits from sales in urban areas, they scaled their outreach to more rural areas. Through this model, VisionSpring reached financial sustainability. A big step forward that few social impact organisations accomplish.

Their second pivot came after recognising that, even though they were now able to grow at a faster pace, they were still not on track to provide eyeglasses to 2.5 billion people. VisionSpring decided to focus on leveraging existing networks and working through partners such as Bracket Bangladesh, which has community healthcare workers all over the country, to deliver vision health services.

Their latest pivot was creating the Eyeliance. A public-private partnership that brings in eyeglass manufacturers, governments and nonprofits together to look at how to address the market and policy failure. For example: how companies are not incentivised to produce low-cost enough eyeglasses and to distribute them to rural areas. 

This journey is emblematic of what it looks like to relentlessly pursue impact by focusing on the big challenge gap which, in VisionSpring’s case was 2.5 billion people.


Even though it may seem counterintuitive, It’s important both to think big but also to start small. 

Too often in the social impact, we do the opposite. We think too small and we start too big. For example, nonprofits receive a grant, get a programme started and they’re held accountable to start delivering results. Consequently, the pressure of delivering doesn’t give the organisation much time to experiment and figure out the best solutions. 

A key benefit of starting small is that you can quickly experiment, take more risks and learn more quickly and cheaply.

When planning for impact, we’d most likely end up doing a great deal of planning, meetings, and research. Over that period, we’d be building up significant risk and spend considerate amounts of time and, potentially, money. Instead, from a lean approach, we could look at how to overcome risks in advance. Without having spent so much time planning in detail, we would get into the field and test different variations of our strategies for each intervention.

We learn by experimenting

The core of the lean startup is to build, measure, learn and get feedback.

Again this is completely based on the scientific method.

You have a solution that you believe will work for a problem. You form a hypothesis about what you expect to happen, and then rather than just designing that at great length you go out and build a Minimum Viable Product. You then test that hypothesis and measure the results by gathering data and analysing it.

In lean terms, the faster you can go around this cycle, the faster you’re going to innovate. Therefore the focus should be on the speed of iteration. Not on doing things perfectly.

A necessary shift is to go from aiming at perfection to how fast can we go through this process and learn.


When it comes to social innovation, the things we want to test fall along 3 different dimensions: Value, Growth and Impact.

Many organisations get one or two of these right, but rarely all three. 

Value is created by a solution that people want and will demand, tell their friends about and come back for. Making an impact is asking: Does this solution actually work? Does it have a social benefit?. If people want it and it works, Growth is asking: Can this solution be delivered at scale?.

If you miss one of these dimensions you’re unlikely to move the needle on a problem. Let’s look into these dimensions in detail.


Is there a real demand for your solution?

First, it’s not enough for people to accept that you give them something for free. One of the challenges in the social sector is to create solutions that users genuinely see value in. It’s the only way to ensure they’re going to take advantage of it. 

Second, proximity is key. We need to engage users at every stage of the process. From the early design through testing and iterating, to getting it in people’s hands and learning from them; and whenever possible, having them help de us design the solutions. The people that are experiencing these problems understand their nuances best.

Third, it’s important to look at behaviour rather than just focusing on hypotheticals. For example,  NextLeaf develops wireless sensors to measure the usage of clean cookstoves, which is one of the methods we use in International Development to avoid the health consequences of indoor air pollution. 

Usage of Improved Cookstoves (ICS) and Traditional Cookstoves (TCS)
NextLeaf Impact Report 2018

However, evidence showed that even though cookstoves had been distributed in rural India, only 10% of people were using them. The reason for this is that there’s no correlation between what people said that they were doing, i.e.: how often they were using the cookstoves, and what they did. 

This important insight shows that when you ask people what they do and what they want, there are all sorts of reasons they might tell you something completely different. 

With the sensors, NextLeaf was able to find out what was happening and through that, they asked targeted questions that helped them elicit some of the design issues with the cookstove, which they could then address accurately. As a result, NextLeaf eventually got close to 100% usage of the cookstoves.

The best way to approach experimentation in the social impact space is by developing a Minimum Viable Product (MVP). For example, Kenyan social enterprise Copier Global is trying to tackle the issue of getting a much wider variety of consumer goods to people living in remote rural areas.

One way to do it is to could set up warehouses, build catalogues, create a distribution or agent network. What they did instead was to test out their hypothesis where people buy and sell from a catalogue to find out what kind of things people would buy and sell.

They created an MVP. Within one week, the CEO went to the local market, took pictures of the products he believed people would buy, put them into 5 catalogues, gave them to some potential agents in the village and sat back to see what would happen. When someone placed an order he went to buy the product and hand-carried it to the village. This is not a scalable solution but something that could get going fast

The first hypothesis they proved is that people would order from the catalogue. Second, they got insights into the product niche they should focus on. Third, their hypothesis of agents was disproven. They learned agents were more motivated to sell from what they had in their inventory rather than from the catalogue. They found that the best agents to work with were people who ran complementary businesses, such as a hair salon where people could find the catalogue.

By running this MVP, within one week, the organisation learned valuable lessons that helped them better direct their efforts as they started building their solutions.


Does the solution actually work?

We often focus on this question but unfortunately, we don’t always focus on what it entails: a solution that works as well as we know it could.

One of the biggest challenges with taking a lean approach to global development challenges is that a lot of things we’re trying to achieve take a long time, they are harder to measure. The tool for addressing this challenge is the Theory of Change. 

Something social impact practitioners create for their programmes but too often, these elaborate Theories of Change sit in a bookshelf. The necessary shift in this regard would be to move from a static approach to testing the linkages in the Theory of Change, from the very earlier stage.  

For example, consider a programme that aims to reduce the incidence of malaria by distributing mosquito nets. It may take you a year before seeing if the incidence of malaria has reduced in a region. However, you could find out tomorrow if people hung up the mosquito nets, and if they’re sleeping under them. Testing those early precursors of impact doesn’t guarantee you’ll achieve the result. But if you don’t test the early precursors, the optimal impact is unlikely to happen.

So how do we test impact in the early stages?

By identifying, maximising and optimising the precursors of impact before we go too far.

Drawing from the previous example, If 60 % of people are hanging up their mosquito nets, how do we get to 90%?. Can we give instruction manuals? can we deliver lessons in the village? can we go hang up the mosquito nets for them?

By optimising the linkages of your Theory of Change, you’ll be building the business equivalent to a sales funnel. If you optimise these linkages in the early stages, you will have less loss each step of the way. Therefore you’ll be far more likely to get a positive outcome in the end. 


How do we get our intervention at the scale of the number of people who could benefit from it?

Foreign aid is diminishing over time. It’s relative to private sector flows, which are 10 times as much and domestic resource mobilisation which is 3 times even greater than that. As the proportion of foreign aid becomes less and less, we need to think about ways of tackling the world’s most pressing challenges differently. 

A key mindset shift in the foreign aid sector would be to think to take the risks that local governments and the private sector cannot. The foreign aid sector can find better solutions to social problems through innovation, and scale them through the private sector and local governments.

What does this look like in practice?

Certainly, the most straightforward model to build impact-driven solutions at scale is to create market-driven business models where people pay for products. For example, companies doing off-grid solar systems that allow people to use mobile money to pay for their solar system over time.

Alternatively, in cases where people can’t pay enough for a particular product or service, a subsidy would be a viable model. I.e.: Aravind Eye Hospitals.

Replication is another viable way of building impact at scale. For example, VisionSpring replicates impact by working with several different partners. 

How to leverage government spending for scalable impact?

Consider the Partnership Schools of Liberia experiment, somewhat controversial but fascinating. The Liberian government humbly recognised that the school system was failing and brought in nonprofit and for-profit providers to take over schools. Through an evaluation ran alongside the project it was evidenced that most of the providers did significantly better than the government. Over time the goal is that more governments can adopt better educational paradigms.


In the philanthropy space, we’re generally happy if we deliver some impact. Largely because most funds come from donations.

We need to raise the bar.

Just as companies are required by regulation to maximise shareholder value and profit; working for good should be required to maximise the social impact that we’re able to deliver for the resources that we have.

However, there’s no shortage of factors that push against that. I.e.: we fall in love with the problem and we get too attached to a particular solution that may or may not work at scale. These things can often divert us from coming up with the best solutions, or partnering up with those that have these solutions.

Innovation starts with an audacious goal. Scalable Impact is the foundation of innovation.

When you have a goal that you cannot achieve with business as usual, it forces you to innovate. It forces you to take some risks and do things differently.

Nowadays, most organisations include innovation in their narrative. Yet when we look at their goals for the year, we can see that they don’t think big enough and aim at, for instance, 20% improvements. There’s no motivation to take risks and innovate to deliver an improvement of 80% because the goal is just 20% better. 


Culture change doesn’t happen until we shift our idea of what success looks like. Success should be something that you cannot do with a business-as-usual approach. If you find out that business-as-usual is insufficient to solve the problems at hand, you should have a goal with high potential for impact.

We also need to shift the way we measure our progress. Too often nonprofits, foundations, social enterprises and governments focus on numbers. For instance. A common claim among impact-driven organisations is “we touched n million lives”.

But do they have engines for growth that could scale solutions to the size of the problems at hand?

The recommended shift, in this case, would be to focus instead on innovation metrics: unit metrics. To illustrate, at a unit level for each person or every hundred people an organisation reaches, What percentage are seeing value in it and showing signs of that?

Testing your hypothesis around ways to reduce your costs and ways to increase your success rate are the kind of metrics that matter for driving impact. The overall goal of taking a lean approach to social impact is to shift from knowing that we’re doing some good to knowing that we’re maximising the good that we can do, both from a depth and a breadth perspective.