Reliable estimates of release or discard mortality (DM) rates for recreational and commercial fisheries are necessary for robust assessment of the effects of fishing on populations and for establishing effective regulatory measures concerning the release of fish. In situ experiments involving the tagging of released fish are most likely to produce accurate estimates given that experimental subjects are captured and released under representative fishery conditions. Data from electronic tags can be used to infer movement or behavioral patterns of released fish, from which mortality times can be derived to then estimate natural mortality and DM rates. However, resulting DM rate estimates are often of low precision (high variance) due to small sample sizes associated with the elevated cost of electronic tags. Alternatively, returns of traditional tags from small scale experiments constitute a relatively inexpensive means of estimating the relative DM rates of two or more groups (e.g., vitality or injury classes, capture methods) of released fish. Combining the two approaches when there are two or more groups of discarded individuals can be beneficial. First, results from the two methods can be used to assess whether experimental effects such as tag-related mortality could bias estimates of DM rates. Second, we propose a modelling approach that integrates data from these two types of tagging methods to produce absolute DM rate estimates that can be more precise than those obtained from electronic tagging alone. We show using simulations under which conditions precision is improved and how different allocations of effort to electronic and traditional tagging can be used to plan DM experiments with respect to the precision of estimates and cost. An example based on a recent study is presented to illustrate the application of the proposed approach.