My hypothesis is that Larian actually got way more feedback than they ever expected - they sold way more than expected in the EA (something like 1 million copies in the 1st weeks) - and thus were never properly set-up to parse through OR action feedback adequately.
As someone who has done market research as a part of my job, just analyzing the shit a 20 person focus group takes quite a bit of man hours. The amount of input that Larian must have received through their in-game tracking data, the feedback system, and the essays people wrote online was probably absurd. Also, the latter two (forums and feedback forms) are completely free-form and not set-up in any way to easily let them extract focused information. This is probably why in the first couple of updates, all Larian talked about in terms of mechanical feedback were information gleamed from their in-game tracking tools - because that was the easy stuff. Bug fixes are also relatively easy (no analysis or debate really needed), which is why we see that implemented way faster.
Not excusing Larian here of course - the feedback systems were ultimately set-up by them, and this might show their lack of experience dealing this and the troubles of scaling to an AAA company and trying to keep their previous feedback approach.
In terms of actually implementing feedback, Larian's rapid expansion probably hurt it there too. Bureaucracy is inescapable in large companies, especially one that is expanding so rapidly. There's probably a department for everything now, and implementing any major mechanical feedback probably involves multiple meetings with multiple departments, with the feedback team needing to convince a bunch of people with fancy titles why their idea sucks and need to change. I.e., most mechanical complaints like high-ground advantage is probably pretty easy to code in, but the difficulty part is selling that to the mechanics team, and likely also the encounter-design team as they will probably have to adjust a whole bunch of stuff they've already done. Knowing how most meetings go, they probably often talk around in circles for hours without coming to any productive resolutions.
The lack of communication is a bit baffling to me - seeing as even more frequent, non-committal updates like "hey guys, we're working on X and Y this month", would appease the more vocal parts of the community. I must assume they probably just ran the math and said "fuck it".
1) Anyone who's desperate for an update have already purchased the game
2) It's a minority - of the 1 mill + EA players, most simply played and moved on. If you look at the various forums and reddit, there's maybe 1000-ish people actively, constantly discussing the game? Even adding a factor of 10 to be safe, these 10,000 customers is just 1% of the people who have already purchased the game
3) Within this group of "update seekers", most of them are already firmly in one camp of another of "Larian can do no wrong" or "Larian BAADDD"
There's a reason why most the EA process is far less common with AAA games. I get why Larian went for it - as a standalone studio it's very nice to get an early cash injection during development cycles, and it can be both a good marketing and market research tool. But in this case, I suspect they were woefully equipped to handle the process simply due to the unexpected scale.
