In some ways it also reminded me of something I'd read a while back about an engineering team that needed a solution and they realized their best bet was to speed up how quickly they could TEST a zillion options. So not working so much head-on but by making things easy to see, and cheap to be wrong.