A chemometric challenge was proposed during the "Chimiométrie" congress 2016, held in Namur, Belgium, on 17–20 January. The aim of this contest was to challenge the ability of congress participants to build indirect Quantitative Structure-Retention Relationship models (QSRR) using the linear solvent strength (LSS) theory of reversed-phase liquid chromatography. QSRR is a very helpful method for the identification of unknown analytes, including the prediction of chromatographic retention time. Because of the potential presence of various isomeric compounds, accurate retention time prediction is particularly important in the context of steroid identification. In addition, the indirect prediction of retention time using the linear solvent strength (LSS) parameters S and log kW provides a great advantage for use in any gradient conditions. In the proposed dataset, the experimental values of S and log kW were estimated using Ultra High Pressure Liquid Chromatography separation with two linear gradients (5–95% ACN+0.1% FA) of 15 and 60 min, respectively. The aim of the challenge was the accurate estimation of retention time for a 45 min gradient by applying the LSS theory based on the predicted S and log kW values. Molecular descriptors were calculated from a series of reference steroid compounds using the VolSurf+ software. By these means, a collection of 128 variables related to molecular shape, volume, polarisability, polar surface area, hydrophobic surface area, lipophilicity, molecular diffusion, and solubility was generated automatically. The dataset (n=95) included 76 steroid compounds for calibration and 19 for validation. Experimental log kW, S and retention time values were provided for the calibration set only. The results were evaluated according to the smallest RMSEC obtained for the retention time predictions of the validation set with the 45 min gradient using the LSS parameters. Moreover, each individual relative error should not exceed 5% of the experimental retention time for both the calibration and validation sets. This paper summarises the approaches proposed by the best three participants and the challenge organiser.