If they get unexpected results they could note what they could've done wrong or what they could change in a next trail of experiments. Or look back and see what different happened to they could've hypothesized. There could be many different courses on what to do next.
