There are many things that he wouldn't know because he gets presented a filtered view that makes his direct reports look better 1/ how processes actually work

There are many things that he wouldn't know because he gets presented a filtered view that makes his direct reports look better 1/ how processes actually work (theory versus reality) 2/ inefficiencies (wasted time/wasted money) 3/ opportunities (to make more money/save money) 4/ If people are actually happy or just saying they are 5/ If his middle management is any good or not 6/ People at risk of leaving 7/ Training issues (people asking how things work who have worked there a long time) 8/ Attitude issues (laziness, turf wars, "it's not my job", shoving problems to somebody else) 9/ Surfacing ideas from people dealing directly with clients or the actual work (skip-level reporting) I think that your time commitment is too low to really get into it but given that's the time you have I would start at the bottom to gain trust and neutrality and work the way up. You need a credible "role" that would make people open up to you but where it would make sense you talk to everybody (e.g. MBA student from a business school doing an internship and associated dissertation). I wouldn't return value to the CEO until you stop the experiment because the moment you communicate it, things will change and you lose neutrality and trust.