ObjectiveGlobally, early and optimal feeding practices and strategies for small and vulnerable infants are limited. We aim to share the challenges faced and implementation lessons learned from a complex, mixed methods research study on infant feeding.DesignA formative, multi‐site, observational cohort study using convergent parallel, mixed‐methods design.SettingTwelve tertiary/secondary, public/private hospitals in India, Malawi and Tanzania.Population or SampleModerately low birthweight infants (MLBW; 1.50–2.49 kg).MethodsWe assessed infant feeding and care practices through: (1) assessment of in‐facility documentation of 603 MLBW patient charts; (2) intensive observation of 148 MLBW infants during facility admission; and (3) prospective 1‐year follow‐up of 1114 MLBW infants. Focus group discussions and in‐depth interviews gathered perspectives on infant feeding among clinicians, families, and key stakeholders.Main Outcome MeasuresThe outcomes of the primary study were: (1) To understand the current practices and standard of care for feeding LBW infants; (2) To define and document the key outcomes (including growth, morbidity, and lack of success on mother's own milk) for LBW infants under current practices; (3) To assess the acceptability and feasibility of a system‐level Infant and Young Child Feeding (IYCF) intervention and the proposed infant feeding options for LBW infants.ResultsHospital‐level guidelines and provision of care for MLBW infants varied across and within countries. In all, 89% of charts had missing data on time to first feed and 56% lacked discharge weights. Among 148 infants observed in‐facility, 18.5% were discharged prior to meeting stated weight goals. Despite challenges during COVID, 90% of the prospective cohort was followed until 12 months of age.ConclusionsEnrolment and follow‐up of this vulnerable population required additional effort from researchers and the community. Using a mixed‐methods exploratory study allowed for a comprehensive understanding of MLBW health and evidence‐based planning of targeted large‐scale interventions. Multi‐site partnerships in global health research, which require active and equal engagement, are instrumental in avoiding duplication and building a stronger, generalisable evidence base.