What information might you need to be able to make sense of your research in 6 months… 3 years…10 years? What might another researcher need to know to reproduce your findings? Issues of reproducibility are a recurring theme in research and data management. There are many different reasons for failures of reproducibility, and not all area simple to address. Insufficient metadata is an area that is relatively straightforward to identify and improve once researchers are aware of the need. We have developed an interactive game for 4 – 24 players using Lego to help researchers explore the metadata they might need to record to aid reproducibility. Our game addresses issues including planning for metadata, formats of metadata recording, standards and automation. Our game also draws multiple parallels between recording and communicating the research process and documenting and the creation of a Lego model. The process of playing the game draws researchers into a discussion of how metadata is captured, recorded and disseminated, which in turn provides an opportunity for signposting to further resources in this area. Feedback from researchers has been positive and we have created a game pack that other research support services could use to run the game with their own researchers. The pack provides facilitators with templates for activities, a list of materials needed and game set-up instructions. There is also a suggested short introduction to get things going and discussion points for afterwards. In this talk, we will give an overview of the game and how it is played with a brief analysis of feedback from researchers who have played it.