Some observations from a game as FanEga+Xenophile Humans, Meritocracy + Beacon of Liberty (+ Shared Burdens), just after winning the game in 2425 (crisis didn't show up). I tried playing tall-ish but still ended up with 52 systems at the end of the game due to spawning in a region with great geography:
Some observations:
Unity is overvalued, Influence is undervalued
Currently, after the initial land grab there is little to do with Influence. Habitats are problematic currently, ringworlds are even more so, so the only things left are claims and GC stuff. I can't comment on how this affects wide empires as I haven't played one yet, but it turns the GC even more into a player tool. With all that spare influence I can call in plenty of favors in order to shoot down or force through whatever resolution I want.
Likewise, Unity is in too high demand. Unity Ambitions are especially egregious, currently a single one would cost me 3638 unity/month. Combined with the fact that many Unity Ambitions are of dubious value (Hearts and Minds, Will to Power, Fortress Proclamation and Desperate Measures are outright bad) there isn't really a reward at the end of the unity tree. An Unity-focused empire might do better, but they'll also not get Ascension Theory as early as a tech-focused empire. Like, maybe turn Unity Ambitions into something that requires Influence instead.
Empire Sprawl works backwards
Make tall play more viable, preferring to balance tall vs. wide play in favor of distinctiveness, and emphasizing differences between hives, machines, megacorps, and normal empires. (This does not necessarily mean that tall Unity focused empires will be the equal of wide Research focused ones, but they should have some things that they are good at and be more competitive in general than they are now.)
Congratulations, you failed. As it is currently Empire Sprawl massively favors wide empires. Every owned system increases empire sprawl by 1, but every colony increases empire sprawl by 10. A wide empire does not build up their space, so they have maybe 1 planet per 5 systems ( = 15 sprawl), whereas a "tall" (I still think the term does not work for Stellaris, "dense" is a much better description) spams habitats in the midgame and beyond. So they end up with say 10 planets per 5 systems (=105 sprawl). And those planets also carry much more pops combined pops than the single planet of the wide empire, so the effective sprawl is even greater.
Now of course the space of a "tall" empire produces quite a bit more resources than an equal amount of space of a wide empire, but since wide empires can largely ignore unity as they rush down the galaxy (which largely requires tech and alloys).
Now while it would be easy to simply flip the numbers, I think what would be really interesting is to tie empire sprawl into sectors. Perhaps the core sector should be completely free sprawl-wise (at least in the late game), but empires could perhaps chose centralization levels. Highly centralized empires would take full sprawl from the sectors but also efficiently extract their resources, while decentralized empires take less sprawl from sector systems while also getting resource penalties. It should of course be tied to the sectors themselves to allow proper management - an empire might not care about the paltry resources in the Evarian Voidwaste but pay great attention to extract the most from conquered fallen empire worlds.
The Tech-Tree needs de-cluttering
Please, please, please get rid of some of the most offensively bad techs like Offworld Trading Company, Void Clinics, Clone Armies and Gravitational Analysis.
The AI is much better, but that still doesn't save it lategame
As long as the AI does not understand how to properly manage fleets of decent ship designs, no measure of (sensible) economic improvement will help them. In my lategame wars, I still traded single-digit battleships for complete destruction of the hostile armadas.
Also please teach the AI how to build gate networks, the biggest pain of fighting against a crisis or the varius pseudo-crises is dealing with the horrible galactic infrastructure.