Objectives:  To pilot two intensive hand hygiene promotion interventions, one using soap and one using a waterless hand sanitizer, in low‐income housing compounds in Dhaka, Bangladesh and assess subsequent changes in handwashing behaviour and hand microbiology.

Methods:  Fieldworkers randomized 30 housing compounds: 10 received handwashing promotion with free soap, 10 received handwashing promotion with free waterless hand sanitizer and 10 were non‐intervention controls. Fieldworkers assessed handwashing behaviour by structured observation and collected hand rinse specimens.

Results:  At baseline, compound residents washed their hands with soap 26% of the time after defecation and 30% after cleaning a child’s anus but <1% at other times. Compared with baseline, residents of soap intervention compounds were much more likely to wash their hands with soap after faecal contact (85–91%), before preparing food (26%) and before eating (26%). Compounds that received waterless hand sanitizer cleansed their hands more commonly than control compounds that used soap (10.4%vs. 2.3%), but less commonly than soap intervention compounds used soap (25%). Post‐intervention hand rinse samples from soap and sanitizer compounds had lower concentrations of faecal indicator bacteria compared with baseline and control compounds.

Conclusions:  Waterless hand sanitizer was readily adopted by this low‐income community and reduced hand contamination but did not improve the frequency of handwashing compared with soap. Future deployments of waterless hand sanitizers may improve hand hygiene more effectively by targeting settings where soap and water is unavailable.