For each task, consider all software available and try to select the most appropriate tool(s). This can only be approached with sufficient information about available software, and to this end we include a list of all software used for each task within Table S1, and Table S3 includes links to documentation and publications. The selection of the most appropriate tool is not always straightforward, but we suggest considering (i) the extent to which the tool was designed for the intended barcode region, purpose or dataset, (ii) the detail of available documentation and explanation to ensure a tool performs as expected, (iii) the availability and flexibility of options to appropriately apply the tool, (iv) the frequency of use of a tool in other studies with similar research aims, and (v) all else being equal, the simplest approach. Ideally, where multiple approaches exist, reasonable comparison between key methods should take place to fully understand the potential variation in conclusions that might arise from different bioinformatic choices, and the results of these comparisons should be reported. This is particularly the case when considering alternative, conceptually distinct algorithms for more bioinformatically complex tasks, such as denoising and OTU delimitation. The development of software packages and open access platforms integrating a catalogue of common bioinformatic tools, such mBRAVE (http://www.mbrave.net/), may play a fundamental role towards a proper selection and harmonisation of the software used. However, software choices should be made on the basis of appropriateness and usefulness, rather than simply ease of availability and implementation due to inclusion in these packages/platforms.