Internal crowdsourcing offers new opportunities for ideas to gain legitimacy in idea development as not only managers but also experts are able to provide feedback to improve and subsequently accept ideas for further realisation. Given this far limited knowledge about how ideas gain legitimacy in idea development, this paper focuses on the influence of feedback, a main source of idea legitimacy in internal crowdsourcing. More specifically, this study aims to explore how feedback from different feedback providers (i.e., managers and experts) influences idea selection in internal crowdsourcing idea development. Based on text mining methods including sentiment, topic-model, and expert identification analyses, results reveal that feedback sentiment, feedback diversity and feedback amount play key roles in idea legitimisation. Specifically, these aspects of feedback influence idea acceptance in idea selection through the mediating role of idea revision in idea development and that this relationship is moderated by feedback providers. This study extends previous knowledge about the legitimisation process from the perspective of feedback in idea development by offering both a more comprehensive view, including different feedback aspects, and a more granular measurement of these, through the use of text analysis. Based on insights from the study, practical implications are presented for how to gain legitimacy from feedback for idea selection in internal crowdsourcing.