“…While sample sizes are often small, building models upon recent enumeration data, rather than linear projections from census baselines many decades ago in settings where massive changes have occurred provides more confidence in outputs ( Wardrop et al, 2018 ). A growing amount of anecdotal and quantitative feedback from field teams and national statistical offices on the accuracy of estimates adds to statistical evidence from model cross-validation, as well as assessments on the use of data in deriving metrics or in health delivery campaigns adds to this ( Nilsen et al, 2021 , Leasure et al, 2020 , GRID3, 2021b , Boo et al, 2022 , GRID3, 2020 , Thomson et al, 2021 , Ali et al, 2020 ). Moreover, the explicit measurement and communication of uncertainty in predicted population estimates provides users with quantitative insights on where confidence in predictions is higher or lower, taking small area population estimates a step forward beyond the opacity of many top-down model outputs ( Leasure et al, 2020 ).…”