Yes. Accepting the uncertainty and publishing more than few.

Often famous/more cited studies are not replicable. But if you want to work on similar research problem and publish null/non exciting results, you're up for a fight. Journals want new, fun, exciting results but unfortunately the world doesn't work that way