The ``Assume all Tests are Flaky'' approach to handling test flakiness establishes it as a risk. Flakiness scores serve as the first frontier for handling that risk. In our previous work, we refined an existing flakiness score and demonstrated, using two artificial datasets, how this refined score could support 3 strategies for dealing with test flakiness; Fix, Rerun and Monitor. The question remains, how will this score, and flakiness scores in general, be applied in actual industrial settings? In this study, we, alongside software developers from two separate industrial companies, report our experience when transferring in-vitro flakiness scoring formulae to in-vivo industrial settings.