Purpose: This research, conducted in Ethiopia, involved select stakeholders in the variety evaluation process early: to identify a greater number of acceptable varieties and to shorten a lengthy research and release process. Design/methodology/approach: A Participatory Plant Breeding (PPB) approach was used in both on-station and community-based trial evaluations. Farmers, traders and processors evaluated bean varieties for promotion, considering field performance, culinary characteristics, market value, and canning acceptability. Findings: Within four years, two varieties were identified and recommended for release which were high performing, had excellent consumption and cooking characteristics for local foods, fetched high local market prices and met international canning quality standards. Practical implications: (1) Ethiopian men and women have distinct criteria for evaluating white pea beans, the former putting emphasis on marketability, with women focusing on food security issues. Both groups might be involved in the selection process. (2) Key characteristics cited by farmers, for instance, use for forage, may not be taken into account in the formal breeding process. Expanding the criteria used to determine which varieties should be released would make economic sense. (3) Farmers screened for local product acceptability before researchers put forward lines for testing by the export and canning industry. Local acceptability need not necessarily conflict with market chain acceptability--if the right processes are followed. Originality/value: The context was unusual: a) the region was drought-prone; and b) the crop had to meet rigorous canning requirements, plus local preferences. The involvement of traders and processors was novel, as was the use of a "participatory framework" even when the product had to meet stringent market requirements.