Data-informed testing can be used by learning engineering teams to check theoretical assumptions, e.g. using learning curve analysis to check if a collection of performance tasks measure the same competencies or A/B testing to determine which design alternative is most effective.