<p>Split-lot designs are used in industry, agriculture, and other scientific fields for multistage experiments with randomization restrictions of the experimental runs in each stage. This paper considers split-lot designs with qualitative factors that are based on general symmetric fractional factorial designs, and addresses the issue of their classification into equivalence classes. Two seemingly different split-lot designs can have essentially the same statistical properties for estimation of factorial effects and for model fitting, and thus are equivalent. On the other hand, non-equivalent designs may have the same statistical properties under one particular model, but different properties under a different model. The main results show how, using the concept of coloring, we can treat the problem of equivalence of split-lot designs as a combinatorial equivalence problem of colored matrices. Based on these results, some new necessary and sufficient conditions for detecting equivalence of split-lot designs with qualitative factors are derived. Results can be extended in a straightforward manner to split-lot designs with qualitative factors based on general asymmetric fractional factorial designs.</p>