To date, it is commonly agreed that physical practice, as well as mental types of practice, have the potential to bring about improvements in motor performance and to induce motor learning. The perceptual-cognitive representational background of these changes, however, is still being debated. In this experiment, we investigated the influence of observational practice on the performance and the representation of the golf putt. With this we aimed at adding to the ongoing debate on the particular contribution of observational practice to motor learning. Novices were assigned to one of two groups: observational and combined observational and physical practice. Motor performance and mental representation were measured prior to and after practice and after a three-day retention interval. Performance improved in both practice groups from pre- to retention-test. Together with performance improvements, mental representation structures developed functionally and became more elaborate over the course of the experiment. Interestingly, however, the pattern of changes over the course of the experiment and across the two practice types was different. Combined practice led to improvements in motor performance from pre- to post-test with representations developing alongside these improvements. Observational practice alone did not lead to performance improvement until after task execution, as shown by improvements in motor performance from post- to retention-test, even though mental representations changed from pre- to post-test. From this, observational practice seems to promote the development of representational frameworks of complex action, and thus action-related order formation in long-term memory.