September 10, 2021

After almost ten years leading policy and external relations at IPA, I am transitioning to lead the MacMillan Center at Yale this week. During my last official policy community of practice conversation, which brings together all 35+ IPA policy staff from at least 12 countries, the Peru team shared lessons on how to weather government transitions with the Uganda team who was in turn sharing with the Kenya team insights on their engagement with the technical staff at the ministry of education. And I couldn’t help but reflect on how much I have learned from my colleagues in the almost ten years I have been with IPA. 

I could go on for hours about how far the organization has come in this time, but the most striking thing to me about IPA is, and always has been, the brilliant, caring community of people involved - from the local staff in our 22 country offices to the world renowned researchers, deeply engaged donors, and our influential alumni group.


It has been truly an honor to work together over this last decade, and I am excited to have the opportunity to continue collaborating in my new role and honored to share that I will also continue in an official capacity with IPA as the NORM scaling advisor to help get millions of people wearing masks. 

As I move to this exciting new leadership challenge, I bring with me so many lessons from this formative group of IPA-initiated relationships. Here are ten such lessons I hope might be widely applicable to other leaders across various organizations. 

  1. Trust is the currency of success. Whether it’s to convince a government to scale an effective program, to share sensitive results, or to raise major resources for the organization, building trust has been the foundation of all success I have seen in these ten years. Where we invested in building trust in the rigor of our results, or in modeling humility about what we know and what we don’t know with donors, or where we invested in team building that fostered trusted relationships across borders, departments, or institutions, we were successful. But equally when we (I especially) didn’t take the time to have that extra conversation to explain the details in advance of sharing something publicly, or to get input and advice from the larger team on a new strategy, we often failed. This need for trust-building runs throughout all the other lessons.

  2. There is always more than enough credit to go around. When I first started at IPA, we were very concerned with getting credit for our work. This was understandable given how many people thought we were one and the same as our sister organization, J-PAL. But in hindsight we spent a lot of time trying to differentiate our voice and show our own contributions that could have simply been spent by showing how our differentiator was actually as a convener of people, ideas, and organizations - an evidence mediator as we now say. Fast forward a few years, and we launched the RECOVR initiative in the midst of the pandemic - we co-hosted forums, studies, and communications with dozens of partners. At the country level, we jointly led research with implementers and governments which they felt enough ownership of to use and expand. We carried this lesson into the coverage we're receiving even this week on the NORM masking results, and are acutely aware that big studies only happen through sincere partnerships.

  3. I’m often wrong - but I have to listen well to realize it. As IPA grew over the years - and I grew with it - I started to believe that my ideas must be really good because everyone was agreeing with me much more. I finally snapped out of that delusion the day a country director (who I perceived as senior to me at the time) introduced me to a policy partner as his boss and later over coffee said to me: “yes, we tell everyone at global they’re great because frankly we need their support to get anything done - oh but of course I don’t mean you.” I am FAR from perfect at listening (just ask my team), but being cognizant of power dynamics has made me less likely to speak first, more likely to directly ask others to feel free to disagree with me when I do speak, and to actually back them up when they do. This of course goes back to trust - but I have learned that when it comes to Twitter or media, I might have an idea, but Jeff Mosenkis always knows best, and when it comes to engaging with a scale up in Zambia, I should keep my mouth shut and let Tamara Billima-Mulenga tell me what she needs me to do.

  4. Step aside and let south-south learning happen. Perhaps my favorite IPA story goes like this: after forty hours of travel and zero sleep, the head of the Peruvian Ministry of Education’s internal policy lab, MINEDULab, arrived in Ghana for the Education Evidence Summit, which was to be for the first time the first day of the annual sector review. She said to a group of policymakers from Ghana, Zambia, and Cote d’Ivoire something like “the only way evidence use works within governments is if you (government leaders) own and drive the process.” Fast forward a couple of months, and the head of evaluation in Zambia pitches a lab to her leadership as critical for growing cost-effective programs, like in Peru, because “what I learned is that we need to drive that process ourselves.” A few years later, and the Zambia lab (not to mention the lab in Ghana and the emerging one in Cote d’Ivoire) is not only doing some studies to inform teacher allocation, but they have also launched their own Evidence Day to kick off their annual sector review. 

  5. Evaluation work is political - but trying to remain as unbiased as possible as data-informed advisors can help. As we discovered by releasing contentious education results in Liberia, no matter how hard a team might try to be unbiased (and despite the critics, the whole evaluation team tried very hard), evaluation politics are real. They are not always so visible, but sometimes behind closed doors there is pressure to change wording, to support a particular stance, or to simply let something go so that it doesn’t impact financing. But my colleagues always pushed me back towards the most accurate - yet clear - thing we could say. Time and again, even when political institutions or leadership changed, this stance towards being a helpful, unbiased advisor to institutions or governments helped us weather numerous leadership transitions. 

  6. Failing well requires eating your hat and then learning and coming back stronger. Whether it was messing up a presentation of results (so sorry Karlan et al), advising a slight overstating of some impact only to receive a pile on of critics to reply to, completely missing a grant deadline (thanks funders for your grace), or realizing that we had made a major leadership miss, I have stuck my foot in it countless times. Early on, I watched one of my bosses stand in front of a key funder and simply apologize: we messed up. Then they went on to explain what we had learned and how we would change. And then we made a plan to achieve that change.

  7. The road to large scale evidence-informed impact can be incredibly long - prioritize and invest in high impact opportunities. We have written about this framework extensively in IPA’s strategic ambition, and more recently in this SSIR piece, but I think a mistake that many leaders make is trying to do too much (again, especially me - see #8). There are tradeoffs in where we can put our resources, attention, time, and talents, and the IPA team continues to be thoughtful about using transparent frameworks and processes for prioritizing with four key criteria. Sometimes, we learned, we have to prioritize what might be most promising for impact, even at the risk of making other projects feel excluded.     

  8. Change is the only constant - always maintain the ability to pivot based on new information. This one will make my team laugh/roll their eyes. We would do annual and quarterly planning exercises together, and every time, we would say we would need to prioritize even better and say no to some things. And every time by the end of the day, we had added more things to the list, and then a few weeks later the priorities had continued to evolve. But the ability of the team to roll with changes never ceased to amaze me - like a last minute high profile release on a hot button policy topic, or a surprise Tweet from a partner that committed us to something, or an invitation to apply for extra end of fiscal year funding, due tomorrow, or a global pandemic that required us to completely change our operating model overnight (and again at least a few times since). Each adaptation made us stronger, more trusting of each other, and simply more effective. We didn’t stick to rigid job descriptions (though we do have clear roles and comparative advantages that we articulate!) - instead, we stepped in when we were needed to fill gaps for each other and asked for help from the best people to help as work evolved. 

  9. Evidence use requires many building blocks - what internally we have started to call an “all of IPA” approach. Whether IPA leads all of the building blocks or not, evidence use is not something that is done well in isolation. It requires the whole ecosystem of partners that create and use evidence being involved at the right time. Whether it is building that ecosystem over the course of a decade by institutionalizing evidence in decision-making processes, as in the scale up of differentiated teaching in Ghana, or in building it overnight in a crisis, as the NORM team called on dozens of trusted leaders to come together and do, evidence use and on-going learning need to be central goals. 

  10. Mutually caring for your colleagues as whole humans truly matters. IPA taught me that even our most brilliant, hard-working, successful colleagues are whole people with human needs. I will never forget the way my most senior colleagues modeled this in a way that allowed me to be me from day one. I finished my graduate work and had two babies in the past ten years, and I was among the first parents among my colleagues. They let me blaze my own trail, and, despite the organization lagging behind in formal policies (that are now much improved), my colleagues not only put up with but encouraged me to be my whole self, as a parent, a passionate advocate for global change, and someone with human needs for rest and distance from work. That meant taking turns for the necessary late night/early morning calls, fiercely defending each other’s leave time, and for me it even meant bringing my kids to various work-related events. This care extended beyond just my colleagues - I can think of dozens of partners, researchers, donors, even government leaders who expressed this mutual care for me as a whole human, and I in turn tried to foster that in my own engagement and among my teams. This mutual care positioned us extremely well for the immediate transition amidst the pandemic - of course it was hard for everyone, and no organization handled this perfectly. We are no exception, but there was never a lack of understanding for the kid who burst into the call (and the parent’s immediate departure to help with the remote learning problem) or the need to sign off early to support an elderly parent. And this mutual support that we all took for granted is a lesson for many organizations as we all navigate new normals. 

Wherever I am, whatever I am doing, I will be a forever advocate for IPA and our mission to measurably improve lives at scale, and I hope that wherever you are, we will continue to learn from each other and be able to continue inching towards achieving that mission together.