When you try to close a post as a duplicate, you are given the opportunity to search for the dupe target.
Still, searching for that dupe target you know exists is hard and more often than not I have to wade through a large number of irrelevant posts to find the one I want. You rarely remember the exact keywords from that one post, there may be many other related questions that hit on the same keywords.
I feel searching for dupes could be made much more efficient if the incoming dupe link count for posts were taken into account. If you added the incoming dupe link count for a given post to the elasticsearch engine, then give results a ranking boost based on that number, then often-used duplicate targets are at least going to be ranked higher and should be easier to find.
Take How do I test one variable against multiple values? for example. It is an often encountered Python problem where beginners try to apply English grammar to boolean logic. Yet searching for test variable against multiple values places it 20th in the list of results. It currently has 155 incoming dupe links, using that number to boost the post rank would easily place it above the scrolling fold. The current top 4 search hits have one incoming link between them.