Abstract Deconvolution algorithms for measurand reconstruction are considered. Their metrological and numerical properties are briefly characterized. Six algorithms most frequently used for instrumental applications are selected for closer analysis. Their comparative study is based on the use of spectrometric‐type synthetic data, calorimetric‐type synthetic data and spectrometric real‐world data. Conclusions concerning computational complexity and accuracy of the compared algorithms as well as their metrological applicability are drawn.