This article presents findings of cognitive and performance assessment among recipient and non‐recipient school girls in the catchment schools of the Tawana Pakistan Project using standardised matrices. This is the first such evaluation of the relationship of a school feeding programme with developmental outcomes in rural Pakistan. There was evidence of benefits of the school feeding programme on a range of outcomes with significant improvement in the literacy and numeracy scores. The paired analysis showed a significant improvement (p<0.001) in Raven's Coloured Progressive Matrices and Draw‐a‐Person test while no association was established in grade performance assessment (GPA). The correlation analyses in terms of the effects of intervention on child cognitive development confirmed the positive correlation in all aspects. DID estimates also revealed an improvement in cognitive outcomes due to the programme. Findings support such potential programmes for improving educational and cognitive outcomes of primary school girls in impoverished districts of Pakistan.