Business processes are multimillion dollar assets in large firms. Explicit representations of these processes in the form of process models are used to inform strategic decision making by various stakeholders and as blueprints for automation. However, the large number of business processes within an organization, their frequent changes and the variety of process stakeholders and related purposes, lead to significant challenges in keeping these models aligned with corporate reality. In the worst case, entire process model collections become out-of-date and are slowly deprecated, as new models are created from scratch. The result: headache for your analysts and little value for your managers. This talk explores new research directions aiming at solving these important challenges by combining techniques from process mining with those from the management of large process model collections, to empower process models with the ability to self-adapt to organisational changes.