Whew….lotta different opinions.  What do we do?  For example, there’s a wreck on St Clair that we fish pretty regularly.  Our first time there years ago with side/down imaging we motored over the wreck placing MULTIPLE waypoints following the hull of the wreck.  When we return to fish it, we know exactly where it is, how it is positioned, and where to set the boat up.  I’ve done the same with smaller targets like brush piles and rocks but it makes it super easy with multiple waypoints.  

@Scherbacj- you are correct in your understanding of how it works.  The issue your having will be come combination of the returns from the structure, the error in the placement of the waypoint, and (since you just mentioned it) using downscan.  If you are scanning for structure, I recommend side scan.  With downscan, you are condensign a really wide area of the bottom into a small number of pixels and you lose the 'depth of field'.  Consider the example from HBird below.  While the tip of the tree that is shallowest might be a single point, consider that the entire tree in the right example is probably 50' from port to starboard as the boat goes over it.  On the screen the center of the tree is a single plane.  So to pick a pixel in the dead center of the tree it might be 50' either side of the center line of the boat and you can't tell.  And it entirely depends on which pixel you chose.  If you've run over the center of the tree in the first place then the waypoint should still be close.  But if you've only skimmed the edge with the edge of the beam, the center of that tree could be 75' away.  If you are using side imaging, you don't get the detail of the structure as clearly, but distance from the centerline is much easier to manage.  Do that twice from two angles and you should have two waypoints within a few feet of each other.  Then you go back over it with down imaging to find the details inside.