CS theory is indeed closer to mathematics. But other areas--database systems, computer architecture, networking, user interface design etc., is in fact evaluated via experiments, which is what makes it "science".

For example, if you propose some new technique to make databases faster (e.g. "store tuples column-wise instead of row-wise"), you'll implement it and run various workloads with and without the technique enabled. That gives you a quantitative measurement of the merit of the technique.

I believe that falls under what the OP called engineering, which is fair as in engineering one often has to run experiments to determine if what one thinks will work in the particular case actually works.