Ten Lessons for Multiplying the Graduation Model

Now what?

With half of the 10 CGAP-Ford Graduation Program pilots now completed and a significant collection of evidence showing the effectiveness of this approach, how do we begin to scale up? How do we proceed on the journey from 300 participants per pilot to scale-up magnitudes of 3,000 or 30,000 or even 300,000? How do we begin to dream of reaching the tens or hundreds of millions of extremely poor people who could benefit from the graduation model’s inputs of training, seed capital and savings delivered in a carefully sequenced, intensively coached method?

At the recent meeting of the CGAP-Ford Graduation Program’s practitioner, funder and research partners, I had the opportunity to lead a panel on the challenges of scale with my colleague Janet Heisey, one of the architects of Trickle Up’s India program; Carine Roenen of Fonkoze in Haiti; and Piyali Bhattacharjee of Bandhan in India. Each one has been involved with the Graduation Program since its earliest days, and each organization is now expanding from its original pilot of about 150-300 participants: Bandhan to 10,000 currently with a goal of 55,000 by 2015; Fonkoze from 1,500 graduates to date with another 1,200 currently in process and a goal of 5,000 by 2015; and Trickle Up to about 4,500 now and another 5,000 by 2015.

Here are ten lessons that emerged from our panel, with more to come as these three organizations continue their expansion and other graduation implementers do likewise.

1. The questions are as important as the answers. One of the maxims of developing new programs is to begin to think about the scale from the outset. Even as pilots are being designed, it’s never too soon to start anticipating the operational issues that will emerge as you go from a relatively small number of participants in one or a very few sites to larger numbers and more complexity.
2. Context matters and context changes. While the graduation program is based on a common set of principles and inputs, how it is applied can differ dramatically from location to location, even within a single country, and from time to time. Fonkoze, for example, put more emphasis on health training after cholera became a major issue in Haiti.
3. Program design is continuous. Based on what we all learned from experience and research, each organization keeps fine-tuning its program elements and timing. As it expanded, Bandhan offered participants a wider range of livelihood options and found that forming village committees, including local leaders, helped with buy-in from communities and participants. Trickle Up accelerated the start of its savings component, even before participants were seeing increased income from their businesses; adjusted the frequency of “handholding” visits by field workers; and put more emphasis on helping women access available public services, such as access to pre-natal care for pregnant women.
4. Scale does not guarantee economies of scale. While everyone would like to find greater efficiency and lower per-participant costs as they scale, dramatic reductions in cost are not axiomatic. The graduation model’s effectiveness is based on seed capital grants and frequent coaching by program staff, costs that don’t necessarily decline through larger volumes. Bandhan did not find cost reductions through greater numbers. Fonkoze found a per-participant cost reduction of about 20% as it expanded. “The infrastructure becomes a little more efficient, but it’s not a huge number,” Carine explained. Ideas that could be explored to reduce costs are using mobile phones for “virtual handholding” and using program graduates for some coaching, both of which could lower labor costs.
5. Finding, training and retaining field staff will be one of the biggest challenges of scaling up. “These are intensive jobs, mentally and physically,” Carine noted. “It’s not a job you can do 15 years, maybe 3-4 years at the most.” What are the qualities to look for in the next generation of field staff? Empathy is a quality that both Janet and Carine identified, and Piyali said Bandhan looks for people from the communities where they will be working and who are highly motivated. “If the staff is not motivated,” she said, “it’s hard to see how the beneficiaries will be motivated.” To help train future field staff, Trickle Up is developing videos and training modules.
6. The second generation of management will be different from the first. Said Carine: “We had to accept that we couldn’t clone Gauthier (Dieudonne),” the visionary and charismatic manager of Fonkoze’s Chemin Lavi Miyo graduation program. Fonkoze is developing career tracks so that it can fill its management ranks with former field staff as the program grows. And going from pilot to scale is more than just “more of the same.” It’s no longer “a close-knit family” running the program, Carine noted. “You need more checks and balances.”
7. Strong systems are essential. Building basic operational systems, such as information systems, becomes especially important as numbers multiply. Keeping track of monitoring and evaluation data for 300 people can be improvised with manual systems or simple spreadsheets, but that essential function will need sophisticated software when program size is 3,000 or 10,000 or more. Both Trickle Up and Bandhan looked without success for easily customizable client tracking software and now have to develop their own.
8. Data is important, but so is old-fashioned conversation. Along with formal monitoring systems, all three organizations rely on dialogue among program staff at all levels to help ensure continuous learning and program improvements. Discussions and meetings have to be intentional and planned, rather than spontaneous.
9. Funding matters, of course, but so does having the right kind of funding. “Funding is important, but the regularity of the flows is just as important,” Carine said. “It’s critical to managing staff expectations during the ups and downs.” Bandhan has found multiyear support from Axis Bank. Trickle Up from Ford Foundation India and Fonkoze from The MasterCard Foundation. It was suggested that long-term tracking of participants in the original pilots and subsequent groups is important, even if only a sampling basis, to continue building the evidence that graduation can be sustainable.
10. Stay flexible but also know your non-negotiables. “When scaling up, we decided we would never compromise on beneficiary selection and staff selection,” Piyali said.

As I look ahead to our future at Trickle Up and the ones our colleagues are building, I think back to the day about six years ago when I first met Dr. Syed Hashemi of BRAC Development Institute, and he outlined his plans for an ambitious effort to test and scale the graduation approach. It’s amazing how much we have accomplished since then and positively thrilling to project forward and imagine where we will be in six years.