The current work was taken up to predict the yield of the concentrate with 2 mass% target alumina from the available iron ore slime material. It was found that the conventional method was not suitable for characterisation of the slime. As an alternative method, a third generation of automated mineral analysis machine named “QEMSCAN” is used to find out the liberation characteristics. However, data generated through this routine method was also inadequate for the assigned job. The routine method predicted a low yield for the concentrate. This was due to skewed distribution of mineral phases and wide variation in chemical composition of the major mineral phases. The database of this machine was then modified so that it recognises three broad subtypes of the major mineral phase. The new set of data was used to establish the yield–grade relation for the concentrate from the slime. This vital information on liberation characteristics depicted the beneficiation prospect of the slime.