• Design for scale
  There are projects where the benefits of scale are such that implementation on a pilot scale makes no sense. In such cases, the project is designed for scale and may be dependent on economies of scale and a market of a specific size.
  
  For example, the Macedonia Connects project achieved national scale by creatively looking at demand beyond the initial goal of providing school connectivity and by realizing the potential for profitably serving a wide range of customers in underserved areas.
  
  Designing for scale would seem most appropriate – for USAID -- where 1) the available budget -- when all possible cost-sharing is accounted for – is proportional to the scale of the country; and 2) proven technology and business models are being deployed – or at least one of the two.
  
  
• Pilot, then scale – “scalable approaches”
  Many ICT projects are implemented as “pilots”, with an expectations that if they are successful, they will be “scaled up”, yet we still know very little about how to successfully transition from a pilot to a scaled-up initiative.
  
  In many Last Mile Initiative projects, the emphasis is on piloting new technologies and new business and organizational models. At the same time, these projects are under pressure to be designed with scalability in mind. There is little point in demonstrating the benefits of a technology or business model if these benefits cannot be extended beyond the confines of the initial pilot site.
  
  A pilot of such a nature should include careful analysis of the factors that may hinder or facilitate scaling up. Here, the factors that may have helped the project decide on the location of the pilot site – usually selected because it is thought to be a “favorable” location – need to be revisited. The requirements attached to a successful pilot may simply not be found in many locations around the country. Alternatively, the model may need some tweaking to address local conditions – which may be quite different – in distinct regions of the country.
  
  These two models respond to two different needs, different sets of circumstances and have different risks and potential benefits attached to them.
  
  What Kind of Scale-up is Anticipated?
  
• Private sector led scale-up
  Assuming a pilot project has been successful in demonstrating a new technology, the private sector may be willing and able to pick up that technological innovation and scale it up – at least where it anticipates that it will be profitable. This may expand the reach of telecommunication services beyond what the private sector might have done on its own.
  
  
• Subsidized scale-up
  To complement a private sector led scale up; well-targeted subsidies may be required to serve areas where the private sector cannot profitably scale up operations without a subsidy.
  
  Case in Point 1: Unexpected Results in Romania
  The RITI-Access project in Romania piloted a number of local eGovernment applications. Some -- such as the digital libraries -- flourished and expanded beyond expectations. Others – such as the e-taxation and computerization of social benefits – were successful at the local level but did not expand beyond the localities in which they were piloted. Why? One could speculate that there is something about the nature of an application that makes it more or less “scalable” or “replicable”.
  
  Case in Point 2: Scaling up the Use of PDAs for Voter Registration in Rwanda
  This pilot project tested the use of handheld computers (PDAs) to collect and validate voter registration information, to develop a protocol for transferring this data to the National Electoral Commission of Rwanda’s central database, and to train NEC staff in all related processes.
  
  In this case, the pilot did a lot of the work that would have been necessary for a scale-up since it developed all the necessary processes, training manuals and PDA-based forms. Pilot testing within the capital of Kampala was successful and yielded useful lessons of particular relevance for a potential scale up activity.
  
  Specific issues related to a potential scale up included the following:
  
  
• Need to assess the use of the PDAs in rural settings (with power constraints);
  
• Assessing the different potential uses of the PDAs
  
• Scale up of training on basic use of the PDAs
  
• Need to develop a sustainability plan to handle repairs
  
• Increasing local capacity to support the system
  
  It is often difficult for project implementers to keep track of how the projects they piloted evolve, scale up, or get institutionalized. Follow up evaluations, undertaken beyond the project’s end dates, would be valuable in better understanding long-term impacts.
  
  Case in Point 3: Network Power in Mali
  Telecenter networks bringing together previously isolated telecenters often established under separate initiatives increases their sustainability.
  
  The initial phase of the USAID-funded Mali CLIC project, implemented by AED and its local partner, established a network of 13 telecenters spread throughout the country. In the second phase, the local partner has continued to provide support to the 13 telecenters, and with additional support from telecentre.org, has worked to create a national telecenter network support organization covering other telecenters support by USAID, UNESCO and other development agencies.
  
  Given that the 13 telecenters were not all located in areas with a sufficient market for their services, they were not all expected to be sustainable by the end of the project. As a network, however, they would be able to take advantage of some economies of scale (buying supplies for example), and share the financial burden of the common support unit or “secretariat”.
  
  One area where there are clear economies of scale and advantages to working within a network is in organizing training for telecenter managers. It is more cost effective for a network of telecenters to develop a common set of training materials to be used across telecenters than for each individual initiative to develop its own set of materials. An additional benefit of bringing together telecenter managers from around the country is the informal type of peer-learning that can be facilitated through networks -- including virtual networks.
  
  Recommendations
  
• Designers of pilot projects should state explicitly whether the pilots are intended as pure experiments, whether they are expected to scale and how they are expected to scale.
  
  
• Where there is a clear expectation of scale-up after the end of the pilot phase of the project, the project designers should consider integrating “scale up” preparation activities within the context of the pilot itself.
  
  
• Project designers should specify their expectations in terms of the scale required to achieve specific results.
  
  
• Evaluators of pilot projects should include a “scalability assessment”, which looks at the potential for impacts to be scaled up and the requirements for a successful scale-up.
  
  
• Questions related to scale should be addressed in all evaluations and evaluation designs must be careful to understand the “scale” elements of the logic model underlying a project – whether a pilot project or a large-scale project. – and the corresponding expectations in terms of anticipated impacts.
  
  
• Post-project evaluation is critical to document whether pilot initiatives are able to scale on their own or not, and why?
  
  Related dot-ORG resources
  
  
• “From Voter Registration to Health Monitoring: Handheld Computers for Development” ( DOT-COMments eNewsletter)
  
  
• “Broadband Connectivity in Macedonian Schools by September 2005” (DOT-COMments eNewsletter).
  
  
• Mali CLICs – Final Report.
  
  Related external resources
  
  
• (Upcoming) “Making the Connection: Scaling Telecenters for Development”. AED/Microsoft/telecentre.org, 2007.
  
  
• Kusakabe, M. (2003). Knowledge sharing and capacity building: Scaling up information kiosks. In A. Badshah, S. Khan, and M. Garrido (Eds.), Connected for development: Information kiosks and sustainability, New York: United Nations Information and Communication Technologies Task Force.
  
  
• Torero, M., Chowdhury, S. & Bedi, A. “Telecommunications Infrastructure and Economic Growth: A Cross-Country Analysis,” in Torero, M., & Von Braun, J. (eds.), Information and Communication Technologies for Development and Poverty Reduction: The Potential of Telecommunications, 2005,
  
  
• Uvin, P., and Miller, D. (1996). Path to scaling up: Alternative strategies for local non- governmental organizations. Human Organization, 55, 344–353.
  
  
• World Bank, “Scaling-up The Impact of Good Practices in Rural Development.” 2003.
  
  
  
  
  
  
  

top