I've actually asked if we could houserule Know The Soul's Price to be pre-errata Thoughtful Gift Technique My Eclipse strongly does not want the Servitude effect, but would actually quite like to be able to tell what people would most like as a gift. It has been suggested she could just use existing KtSP and not invoke the Servitude effect, but... the way the Charm is written, that means if you find out what people want you must then refuse to give it to them, because fulfilling their wishes incidentally involves enslaving them to your will. Which seems awkward.
(Also, every time I think of Thoughtful Gift Technique, it reminds me of the scene in the Vorkosigan books where the master spy was reminiscing about the time when he had to bribe an ambassador and it turned out what he really wanted most in the world was an elephant. This is in a space opera series, note. There are probably no elephants on the planet in question.
Master spy to ambitious subordinate: "I need you to go collect a bribe for the ambassador."
Ambitious subordinate: "Yes sir. And what is the desired bribe?"
Master spy: "An elephant."
Ambitious subordinate: "... and how big does the elephant have to be, sir?")
TGT is a Charm I really liked, in 2e, precisely because it's interesting.
Sure, there's a place for dice-adders: one of my fondest memories of 2e is having my Eclipse persuade Chejop Kejak to read her proposal for the future of the Immaculate Order, using Irresistible Salesman Spirit. It wasn't just because of ISS, of course. He still could've stopped the social attack easily if he'd wanted to, via use of perfect defenses / Resplendencies / Willpower, and I suspect part of the reason he didn't was the other RP elements of that scene (the Eclipse started it with an apology for the Usurpation, and her manifesto was her best good-faith attempt at an answer to a question that Chejop's apprentice had asked her). But at least with ISS I could be pretty sure it wouldn't just bounce off MDV. Sometimes it is fun to just roll a lot of dice and say "Fifty-two successes". But I would probably resent ISS if it was actually five separate Charms, each of which increased your expected successes/die by 20%, and I had to buy all of them before I could learn You Can Be More. (I mean, I didn't like the prerequisite of You Can Be More in 2e, but at least it was only one Charm.)
I like the 3e Charms that say "instantly figure out cultural norms", or "make people see you as a good target for their schemes", or that kind of thing - I can see how to use those in a way that's interesting to the story. The ones I don't like are the multiple iterations of "Due to having a good poker face, you are harder to read, and can add dice to your Guile". Or "Due to your magnetic personality, you can add dice to your Presence roll. In [circumstances], you can add extra dice." We consolidated the Excellencies. These do not need to be separate Charms.
As far as I can tell, the reason they're separate Charms is because Holden et al like the idea of finely granular power/skill levels in various specialties, and advancement being in significant part a matter of getting slowly better at things you're already very good at, so there's a meaningful distinction in terms of just straight-up numbers between The Strongest Guy In The World and Starting Solar With An Athletics Excellency And A Maxed Pool. I find this a very boring advancement model. That doesn't mean they're wrong, just that I'm not in their target audience on this one. But I feel like I'm not alone in that.
(A while back, I went through the Socialize tree with a weed-whacker. I actually really like the tree if you cut it from ~50 Charms down to ~35 Charms, and disentangle some of the prerequisites.)
Regarding Swift Gambler's Eye and other speedbump Charms, I noticed this set of proposed houserules on rpgnet:
list of houserules
The poster did a very careful and detailed read-through of Ex3 on the Onyx Path forums, which I found very useful, and I like some of the ideas here. Might do something like this if I run a 3e game, although if we get lucky some of the codification of mechanics might be in the final version.