[Context & motivation] Many requirements documents contain graphical and textual representations of requirements side-byside. These representations may be complementary but oftentimes they are strongly related or even express the same content. [Question/problem] Since both representation may be used on their own, we want to nd out why and how a combination of them is used in practice. In consequence, we want to know what advantages such an approach provides and whether challenges arise from the coexistence. [Principal ideas/results] To get more insights into how graphical and textual representations are used in requirements documents, we conducted eight interviews with stakeholders at Daimler. These stakeholders work on a system that is speci ed by tabular textual descriptions and UML activity diagrams. The results indicate that the di erent representations are associated with di erent activities. [Contribution] Our study provides insights into a possible implementation of a speci cation approach using mixed representations of requirements. We use these insights to make suggestions on how to apply the approach in a way that pro ts from its advantages and mitigates potential weaknesses. While we draw our conclusions from a single use case, some aspects might be applicable in general.