SoCap Recap: Taking It to the Next Level…Designing to Scale

The following is a panel summary from one of the sessions in the International Development track at SoCap 2010, the social capital markets conference. It is provided here as a both a record of the information disseminated during the session and an opportunity to continue the conversation.

SoCap10:  International Development Track

Fort Mason, San Francisco
October 4 – 6, 2010
Panel Topic: Taking It to the Next Level: Designing to Scale
Purpose: In designing programs, it is important to think about the initial pilot as well as how to take it to scale after its success has been proven. The panelists have developed and run prototypes and scaled programs to new geographies.
Design/methodology/approach: Reports and expert opinions from panelists
David Lehr, Kenan-Flagler Business School, University of North Carolina
Jocelyn Wyatt, IDEO
Joe Speicher, Living Goods

Summary: Scale can mean replication, or the reproduction of a program in another place, and by other people. Or, it can mean a broadening of services and products to the same community. The funder must determine both its mission and the importance of scaling to fulfilling that mission. Bigger is not always better.

If the approach is market-based, as in the case of microfinance, a market is needed.

If the program is mission-based, volunteers and “evangelists” will be necessary, as in the case of Habitat for Humanity.  If it is policy or mandate based, such as Mothers Against Drunk Driving, or Teach for America, a large base is necessary to change behavior and support advertising

Organizations must determine what is being scaled: the same organization, or the same technology or tool? What process is required to accomplish the goal? The scaling should begin at the community level before reaching the regional level. Sometimes, a model may not expand well from a city to a rural environment.

One panelist spoke of a water-related project which reached only twenty per cent of the population, and would not be sustainable. After a campaign was launched to educate the community about the danger of drinking unsafe water, however, a much broader percentage of the community was reached.

If the mission is a civic involvement such as voter engagement, methods may be scaled to other communities, with adaptations made for each locality. A “tool kit” of ideas or designs may be transported to other communities, as in the case of a program promoting health practices. A program may influence others through consultation.

Most developing countries have health systems, but low funding and high turnover are frequent problems. If a program is begun at scale, or on a large basis, it may not be successful, because it is not human-centered or adapted to the local environment. Programs which are too “top-down” have low adoption rates, low impact and low sustainability. Programs should be willing to share their experiences, and failures. Funders should understand the context of a program, and have a tolerance for its anticipated term. If progress is very slow, funders may lose interest, but if it is too fast, there may be mistakes, and problems compounded. Too many proposals project scale at the beginning, but not set out a clear path to make it happen.

For a program involving the production of goods, the rate of scale will be determined by production, margin, number of agents, and fixed costs, factors that will evolve and change over time. A program design should reflect a community’s perspectives, its means of communication and thoughts about its government. Posters, postcards, or ads may be used to launch a pilot. The focus should be on who’s being served.

Metrics to measure impact may not be rigorous, and are secondary to other markers, like outputs and intermediate steps. If a baseline event is mortality, measures of mortality rates every several years may be the best measure of impact. Job creation is easier to measure than health markers. A randomized trial is the gold standard of measurement, and can be expensive to implement, with the cost amortized over the course of study. Funding should be sought to learn and implement a program incrementally, and tests performed at intermediate points. Once a program has it “right,” funding should be sought to bring it to scale.

People in the field deliver information more effectively than a system will. A business plan may have to change, based on unforeseen events. A good plan will have more than one path to scale, with an explanation for each as to what is necessary to progress.

Markets should be segmented. Will a program reach a relatively small number of people, or a large number? Can local culture be respected as a program scales? People who are committed, thoughtful, and process-driven are essential to maintain a personal connection. A nurse/patient experience is dependent on individuals, and not efficiency.

A franchise model, with a “turnkey business” and models and prices that have been determined, allows for customization. Even McDonald’s adjusts its menu to reflect local custom, serving rice in Asia. Those with less entrepreneurial skill can license models and still purchase successful businesses. A microbusiness carries less risk for the employer, employee and funder, while still allowing for a larger group of people to own a proven business.

At the micro level, the reward structure and branding and distribution channels should be established. The talent necessary to start up the program and encourage innovation may be different than that necessary to systematize it.  The program must ask what it is trying to accomplish , and how can it get there? The objectives may not be the same as the models for other programs. The particular tools and per-unit economics must be proven.

Where the mission is the benefit of the entire public, government must step in to mandate policy. A model must be demonstrated successfully in the community before being embraced on a large scale by government. Where government is good at conveying a message and the private sector is not, the private sector should get out of the way.

Anne Peskoe, Attorney

3p Contributor

TriplePundit's guest author program has had over 1000 contributors over time. If you'd like to be a guest author, please get in touch!