Open a big hole in the results section. Put your discovery in another poster. Then, use the poster as a carousel that underlies the hole. Now you can rotate it to show different parts of your story! (Saw this idea somewhere but couldn't find the OP.)

I wanted my Algorithms students to program NP hardness reductions so we developed Karp. A domain specific language for writing Karp reductions. Our students are quite good with a debugger, so reducing learning Theory to debugging seemed like a win. docs.racket-lang.org/karp/index.h...

The statement reminds me of Lemma 2.1 of the KMeans++ paper (theory.stanford.edu/~sergei/pape...), which supports their ensuing proof that the initial centroids will not be too bad from the optimal clustering. It became the default KMeans seeding option in Scikit-learn.