Growing evidence supports the notion that proteasome-mediated destruction of transcriptional activators can be intimately coupled to their function. Recently, Nalley et al. challenged this view by reporting that the prototypical yeast activator Gal4 does not dynamically associate with chromatin, but rather 'locks in' to stable promoter complexes that are resistant to competition. Here we present evidence that the assay used to reach this conclusion is unsuitable, and that promoter-bound, active Gal4 is indeed susceptible to competition in vivo. Our data challenge the key evidence that Nalley et al. used to reach their conclusion, and indicate that Gal4 functions in vivo within the context of dynamic promoter complexes.