About one-fourth to one-half of all infant deaths in developing countries occur in the first week of life. Immediate breastfeeding within the first hour, followed by early exclusive breastfeeding, improves the health and survival status of newborns. The aim of this study was to demonstrate that breastfeeding practices, crucial to infant health, can be improved at scale in developing countries. During 1999-2003, the LINKAGES Project, funded by the United States Agency for International Development, implemented its community-based model to bring about rapid change in individual behaviours and community norms regarding early and exclusive breastfeeding, at a scale [LINKAGES' definition of 'scale' was adapted from a CORE Group background paper on 'Scaling-up' maternal, newborn, and child health services, 11 July 2005] that could achieve significant public-health impact. 'Scale' was defined as bringing improved infant-feeding practices to more people over a wider geographic area, more quickly, more equitably, and with sustainability as a goal. During this time, country-specific programmes were designed and implemented in Bolivia and Madagascar, with catchment populations of one million and six million respectively. These country programmes were implemented with multiple local government, private voluntary organizations, and partners of non-governmental organizations (NGOs) through existing health and nutrition activities. Breastfeeding was an entry point to work at all levels of the healthcare system and, within communities, using policy/advocacy and training for healthcare workers, with a particular emphasis on front-line health workers and community members. Harmonized messages and materials, including mass media, were developed and used by partners. Timely initiation of breastfeeding was one indicator measured. Data collected through rapid assessment surveys showed statistically significant increases (p<0.001) in timely initiation of breastfeeding in both the countries. In Bolivia, timely initiation of breastfeeding went from 56% in 2000 to 69% in 2001 and reached 74% by the end of 2003. In Madagascar, the initiation rate went from 34% at baseline in 2000 to 69% in 2001, 76% in 2002, and rose to 78% in 2004. Exclusive breastfeeding during the first month of life was also measured. At baseline in Bolivia, the rate of exclusive breastfeeding for the first month of life was 81% (2000), decreased slightly in 2001, and then increased to 88% by the end of the Project in 2003. In Madagascar, it started high at 86% in 2000, increased during the implementation of the programme, and by 2004, was 91%. These results were achieved quickly and sustained over the course of the intervention.