Recently, many companies have established online open innovation communities and employ moderators to obtain valuable ideas from regular product users. However, users’ idea framing as a source of bias has been ignored in the literature. To address this research gap, we theorize and then investigate how users’ idea framing might affect moderators’ endorsements, thereby contributing to the rich stream of research on idea framing and idea endorsement. Based on dual-processing theory, we identify both systematic processing and heuristic processing mechanisms through which users’ idea framing—specifically, the cognitive, emotional, and behavioral framing of users’ ideas—can affect moderators’ endorsements in open innovation community contexts. This work involves three studies. First, using a sample of 1,077,077 real-world user ideas collected from the Xiaomi Company’s online community, Study 1 shows that user ideas with higher levels of idea framing are more likely to be endorsed by moderators, and that this relationship is mediated by peer recognition and peer evaluations. Study 2 is an experiment with 498 participants used to examine the contingent role played by the perceived sender’s authenticity in the relationship between users’ idea framing and moderators’ endorsements. Study 2 reveals that higher levels of perceived sender authenticity decrease the effectiveness of users’ idea framing. Study 3 is an experiment with 500 participants used to examine the contingent role played by perceived message trustworthiness in the relationship between users’ idea framing and moderators’ endorsements. Study 3 establishes that higher levels of perceived message trustworthiness increase the effectiveness of users’ idea framing. Taken overall, our findings uncover how users’ idea framing can affect moderators’ endorsements in open innovation community settings, offering new implications for open innovation community managers.