Black soldier fly larvae (BSFL; Hermetia illucens) are one of the most promising species in the field of insects as food and feed. They are rich in proteins and lipids, can feed on a wide range of organic substrates, and have a high feed conversion efficiency. However, the mechanisms underlying feed conversion in this species remain poorly understood. BSFL live in their feeding substrate where non-ingested feed and faeces mix and form a material called 'frass.' Thus, digestibility studies are particularly hindered as ingesta can not be quantified, and it is impossible to collect pure faeces. The present study presents and evaluates two approaches to tackle these issues. Our first approach relies on estimated digestibility (ED), calculated on distributed feed and frass weight. This assumes that, for the same feeding time and amount of feed, working with increasing larval densities would ultimately result in an asymptote of ED, indicating the maximal digestible fraction of the tested feeding substrate. The second approach, approximate digestibility (AD), is based on a usual method to study digestibility in insects or in conventional livestock species and requires the use of an indigestible marker (chromic oxide) incorporated into the tested feeding substrate. Results of these two approaches are compared for three feeding substrates (chicken feed, discarded potatoes, and corn gluten feed). Strengths and weaknesses of both methods are also discussed. This work provides valuable insight into the feed efficiency of BSFL and lays the groundwork for diet formulation based on digestible instead of crude nutrient contents for this species.