Stratified random sampling techniques are often employed to obtain more precise estimates of population characteristics, but efficiently allocating samples to strata is difficult because the optimal design relies on the specification of unknown parameters. Adaptive, multi-wave designs are particularly useful in these cases because estimates for the necessary parameters are obtained iteratively as the expensive variables are collected. We motivate and illustrate the multi-wave sampling design. Unlike simpler sampling schemes, executing multi-wave designs requires careful management of many moving parts over repetitive steps, which can be cumbersome and error-prone. Using real-life epidemiological study examples, we demonstrate an efficient workflow for the design and implementation of multi-wave surveys in R. This workflow is facilitated by the ‘optimall’ package, which offers functions for defining strata, optimum allocation, selecting samples, and organizing the various pieces of a multi-wave survey. Although tailored towards multi-wave sampling under two- or three-phase designs, the R package ‘optimall’ may be useful for any sampling survey.