Neela is the Executive Director of Y-RISE (the Yale-Research Initiative on Innovation and Scale), which aims to develop research around the science of scaling policy interventions. Neela holds an MBA and a PHD in marketing and previously worked in the private sector before her most recent experiences and applications in policy and poverty alleviation. She founded the CSBC (Centre of Social and Behaviour Change) which is based at Ashoka University in India.
From private sector to public sector – learning with Neela
We’re wrapping up our series, BI lessons from the global south with a 2-part blog post informed by our conversation with Neela Saldanha, a well-known figure in the BI world. As someone who has experience in the private and public sectors, Neela’s insights are simply unique, and experiences thought-provoking. We spoke to Neela about stand-out projects in low-resource contexts and how this differs to a more well-resourced context, lessons from the private sector, and how these can be applied to her more recent experiences in the policy space and poverty alleviation.
A project from the global South: Anaemia in India
“The challenge before us was that a large number of people, and particularly women, and…women of child-bearing age are anaemic in India.” Neela is explaining the origins of a behaviourally informed project she led with CSBC a few years ago. “If well-nourished this may not be problematic,” however there are serious consequences for a mother and a child particularly if poor and ill-nourished; and if anaemia remains untreated. Neela and her team knew that there were many potential pathways that they could take to solve this widespread problem.
At first, the solution seemed obvious – Neela and her team needed to use communication to encourage those diagnosed with anaemia to take their daily, free-of-cost iron-folic acid tablet (supplied by the government). Afterall, the main issue was compliance and consistency. But sharing the risks of anaemia, and the benefits of the medication, was not the answer.
Solution space: Test, ideate, test again
Working with Busara, Neela and team brainstormed over 50 potential pathways based on the multiple barriers women faced in taking the medication consistently – from the side-effects caused by the medication, not knowing when to take the medication, to simply forgetting to take it.
Out of these 50 ideas, the team prototyped approximately 20, turning 5 into actual experiments. Out of these 5, the 2 experiments that worked really well, happened to be the simple ones: explaining the medicine’s side effects and Goal Progress, the behavioural mechanism that motivates people when they are reminded about progress towards a goal
The first experiment tackled the barriers of side effects and timing.
“People don’t actually know when to take the medication and it’s just that tiny bit of friction that prevents them from taking something – should I take it in the morning, afternoon, is it best done here or there?”
A community health care worker would use a visual card to explain the side effects and suggest the tablet be taken in the evenings since it can cause nausea – which had seemed to be the reason many women would avoid taking it.
The second experiment used a calendar which allowed people to peel off a sticker every time they took their tablet, revealing a picture of a baby, encouraging a sense of reward. This experiment is informed by the motivating benefits of seeing the progress one is making towards a goal - Goal Progress.
Key lessons from India
The project revealed how problems can be multi-dimensional which will influence the success of a solution. The team learnt that very specific information about side effects and timing was required (as opposed to simply sharing the broad benefits of compliance) and that by highlighting progress, the element of motivation was created and sustained.
Another lesson was around realistic timeframes and resources required to run experiments. “I thought it would take 2-3 weeks, it took us eight months”. Neela explains that while the experiment and data collection took only a few weeks, the preparations took months. The team needed to translate the information into local language appropriately for the context, find sample size, gain permission from village heads, and retrofit a bus with laptops creating a mobile lab.
This speaks to the flexibility, and creativity other practitioners interviewed in this blog series have recalled in running BI experiments in the global South.
BI across scales
What’s different from the global North?
BI-related work is well-established across the global North, particularly across educational institutions, which have set up “tested stimuli you can borrow, ready-made scales you can use to measure something…” when a new project takes shape. However, you are often testing both stimulus and measures for the first time in low-resourced contexts and parachuting measures such as “behavioural games” will not have the same impact and may even be misunderstood. In contexts that have less experience with behaviourally informed projects, the key to running experiments is the execution of an idea (ensuring the stimulus is culturally appropriate, testing measures to be understood, ethical consideration in running experiments, getting all the right permissions, working with existing infrastructure etc.) as opposed to the idea itself.
“It’s less about whether there are good ideas – there are always great ideas”. Setting up the infrastructure and local talent is key when executing an experiment. Ensuring high-quality data is also essential.
Neela gives two examples where context played a major role in execution, highlighting the difference between operating in a context that is not as well-resourced or established when it comes to BI-related work.
In the case of the visual card used in the project described above, the recipients couldn’t understand what the underlying message was with the initial illustration used. For example, they felt that the woman who featured on the card looked too old to be pregnant. Even though the team thought they had ironed out all the issues, there was no substitute for testing with actual people in context.
In a second example, Neela refers to the decision-makers who suggested that instead of running a behavioural experiment, they should run a big advertising campaign should be released. This misunderstanding is often the case in places where behavioural science is new, and unfamiliar to many: people believe that information is the key thing that is lacking; and communication therefore will solve all issues. While Neela acknowledges the need to gain the confidence of decision-makers, she also suggests that the only way to really do this is by leading with action. In this case, while they advised the decision makers about the advertising campaigns (which frames and appeals for example could be tested), they also went ahead with behavioural experiments to close the intent-behaviour gap. As the decision-makers saw the results of the experiments and the evidence, they were more convinced in a way that they would not have been if the team had kept explaining behavioural science to them.
In contrast, some may be eager to adopt a behavioural science approach, when the issue does not lend itself to such, and it’s up to the practitioner to acknowledge this. Neela says: “I think we have to be responsible as well to say that behaviour science may not be the solution in this case”.
Neela expresses how she feels the BI community’s “energy”, and “eagerness” to sell behaviour science, and the desire “to say it can solve everything,” but emphasises the need to be “thoughtful about what it can solve and what it’s good to do” and when to note that “it's not the most important thing right now”. This Neela adds, gives the field additional credibility.
Expanding across the global South: Building a BI community
“It’s a very exciting time to be in behavioural science in the global South,” says Neela, as she imagines the potential of BI in countries where 85% of the world’s population live. Although there is great opportunity here, expanding BI-related work can be tricky to do.
Behavioural science can be a lonely and isolated field in the global South in contrast to the global North. In the US, most universities provide spaces for behavioural science research and projects. “You can meet 100 behaviour scientists anywhere you go,” the “communities are unlimited”.
A frequent suggestion for the BI champions working in these quieter parts is to join online communities. This isn’t always ideal when you consider time zones, cultural differences and accepting that online meetings are just not the same as the interactions you have in-person.
Despite the challenges building communities in the global South can present, Neela believes that this is what BI practitioners and projects need to succeed. She refers to her experience in India when working with Busara:
“If we hadn’t had the Busara team who had done this before and said, well yeah, this is hard I think we would have thought that there was something wrong with us”.
Practitioners in the global North make it looks so easy, and “clean,” frequently publishing their research and running experiments, “and here we are struggling to run even just one small, tiny experiment”. But once you talk to others who have experienced the trials and tribulations of working in these contexts, you realise “it’s just really hard to do something”.
The role of the double agent
Neela considers herself a ‘double agent’ having worked in both India and the US. (She credits this idea to Dr. Madhu Pai from this article in global health). For her, double agents are key in activating BI across the global South.
“It’s our responsibility to sort of advance the field, not by parachuting in and doing the work and leaving, but really thinking through how we can build local capacity quickly enough”.
Building local capacity is a critical part of every project. It’s too easy to receive a grant, go to a place and direct everyone as we wish. Enabling BI in the global South means thinking through how to train and work with people and educational institutions that may already be doing some of this work.
BI’s role in developing countries
Neela believes BI can play a huge role in developing countries where big problems exist. She highlights two broad categories of behavioural science that can have an impact: pure behavioural stuff – for example, making sure people exercise or eat healthy and a second area where you can inform service delivery – for example, ensuring access to electricity. While this second category doesn’t seem like a behaviour change challenge, there is a realisation of the importance of and then the need to take up the services being offered.
Take electricity. While electricity certainly provides consumption outcomes (e.g., using it to study or to cool down in the heat) and enhances welfare, policymakers need to think about how to harness it for growth: how entrepreneurs, for example, can take up electricity and what complementary services and technologies need to be provided to help them make the best use of electrical access. And behaviour scientists can then help to understand how to increase this uptake of services.
“So, I think that our challenge would be, you know, to get involved in these big sectors, like electricity generation, like livelihoods, where there are policies to begin with and sort of say, what's our role here to make sure those investments are actually being used?“
Furthermore, Neela recommends building more cross-discipline partnerships between economists who are already dominant in the developing world, and those like Neela with a psychology or marketing background.
Working with government
Neela suggests thinking “like a behavioural scientist even when we are persuading … government officials” to apply the behavioural approach. The same concepts we apply to a behavioural challenge (Neela mentions the Behavioural Insights Team (BIT) framework) can be applied to the government official. For example, instead of trying to persuade them to do something, give them social proof of others doing it. Make it easy for them to do a project. Think about the incentives for them to ensure it is attractive. And make sure you are reaching them at the right time!
“How do we let them know that this is not something new, especially for the government, I think there's a lot of risk with doing something new. So how do we tell them someone else is doing it? And they're often overburdened. How do we make sure that we sort of pitch it at the right moment?”
Neela points out that the “intent to behaviour gap”, exists not only at project level but rather to consider in every part of what a behaviour science team can do.
The future of BI
Neela is ecstatic that behavioural science has entered mainstream. An ideal future would be where we don’t use the words “behaviour science” “Everyone's just doing it and we actually don’t even use the word behaviour science because everyone’s thinking that way”. Second, in this future, there would be numerous new behavioural science organisations and policy groups, as well as universities offering courses in behaviour science, particularly for those in Asia, Africa, and Latin America, enabling them to set up their own versions of behavioural science – who knows the next Nudge could be published from these parts of the world with new insights for the world!
In addition, Neela shares how excited she gets whenever she sees something that's published out of the global South. “It doesn’t matter if it's an academic paper or just something that’s published even on Medium or Open Access”.
Neela reflects on her days in consulting, and when someone advised her that an “idea doesn’t exist unless it's on paper”. In return, she advises behavioural scientists to do just this - putting thoughts and experiences onto paper and sharing this with the wider world. She adds how many are hampered by the thought of sharing work when it’s not a published academic paper, but that this may make it seem that no work is being done in the global South at all. Neela recommends that with good quality data and insights, the behavioural scientist should worry less about whether it's a new academically relevant article, and rather to simply share these to connect and learn from each other.
This blog post forms Part 1 out of a 2-part series. Part 2 coming soon.