this subject has been in the back of my mind since early in 2016 when GW2 started to lose key team members to some secret project in WA. State.  a few months later i found out about Amazons latest effort to jump into the game-maker market with Amazon Game Studio & a brand new game engine, Lumberyard (which turns out to be CryEngine 4.0 on the back end.) it got me thinking about Corporations that actually focus on a core engine & if done right could be what separates pack leaders from the followers in the years to come. here is a video, showcasing some of the markets current top game engines: Frostbyte Engine is developed by Electronic Arts Games for studios under their Corporate umbrella.
Source Engine is developed by Valve Corporation the Company behind STEAM.
Unity Engine (simplest to learn/use), Unreal Engine (middle range to learn/use) & CryEngine (most complicated to learn/use) all have some form of terms to use/license. anyway to get to the point in my thinking...  Warner Brothers Interactive Entertainment has at least 7 development houses under its Corporate Umbrella.  each studio has it’s own proprietary game engine, or licenses use of an existing 3rd party engine.  & to me it seems like such a fragmented strategy...  so much unnecessary redundancy. i think the future leaders in the corporate games industry will own & develop their own proprietary engine & share it across all their studios.  & the ones who do it best will be the "creme de la creme." so what key factors will being the best take?

-  Adaptable Architecture:
    this one is perhaps the most difficult to address as a layperson...  all data structures will grow,  new systems will evolve, resource demands will fluctuate...  future proofing to scale accordingly...  you avoid reinventing the wheel, or rebuilding your code from scratch.  architecture that is friendly to modularity.
    success in this area will be at the hands of some truly talented coders able to think far FAR ahead.
    additionally if this area is addresses in the design, it will help with increasing assembly line output as the product evolves and grows.   - Top Notch Software Documentation:
    one thing of note with regard to Turbine, this is an area they are absolutely abysmal at.  i have heard nightmare tales of clutter & unclear developer notations all over the source code.  bad habits & bad practice, but symptomatic of a lack of competent oversight.  & it is a problem that just grows over time with code filtering through multiple teams, or transitioning employees.
    this could also be an advantage to having a central team focused on a proprietary engine.  not only to enforce and maintain standardized documentation practices, but also to provide support to development teams in periphery studios. - Cross Platform support (PC (Mac-Linux) - Console (Generational Platforms) & Mobile):
    again with regard to Turbine, watching them switch to mobile games development & the ridiculous number of hirings they had to do for people to work on back end stuff that the Turbine engine did not support for mobile platforms.  additionally i seem to recall Alywens tale of the huge waste of time & resources developing a console port for LOTRO.  money down the drain.
    this could be yet another huge advantage to having a centralized team developing a proprietary engine.  not only could the engine support cross-platform development, but the engine could have emulation tools to test the source code for multiple platforms.  & having a centralized “source (engine) code” would go miles towards making cross-platform emulators work.  additionally that engine development team could assist in porting games, eliminating the need for 3rd party contracts. - Top Notch Current & Updated Net Code: (Game Client):
    when did Turbine lose the original game client optimizers?  was the first cull shortly before, or after Warner took ownership in 2010? or even further back in 2004 when Turbine lost Microsoft’s assistance when they took full control of Ashrons Call?  regardless, with each additional system added... game performance always took a hit.
    optimization of game engine & client should be a perpetual process & having a centralized team constantly tuning would go miles to improving players overall experience & impression.  issues with lag must be aggressively tackled, in perpetuum. - Graphic Toolsets & Asset (visual & sound) Libraries:
    3D model sculpting & 3D pixel painters with animation emulators.  imagine having graphic toolsets something like Zbrush to create models with & test out animations...  hairstyles, emotes, wardrobe pieces, weapons, even particle effects.  multiple studios producing assets (textures, sounds, animations, etc.) all on the same source code & having those libraries grow over time as new material is created.
    extend this to Architecture (buildings) & world construction (landscape.)  tree & foliage textures & animations.  water body varieties. wind effects.  now add sound.  all created on centralized game engine code across multiple studios.
    an asset library that would grow over time, one that could be modified based on theme & so on.  the time this would save in generating product...  and a common source code that is adaptable - modular;  production assembly time frames could drastically reduce over time.

so the above would be the essentials, but i think a corporation that is planning on being a big league (Triple AAA) contender in the coming market will have to set up a proprietary engine for their studios & multitude of game titles.  the initial investment would be significant, but the long-term gains of having something for your “umbrella studio” properties that would not require licensing fees & with multiple content contributors, a constant growing library of audio & visual assets.  all modular & if done right, assembly production at accelerated rates. also if i had to guess, a company like Blizzard probably has most of these things in place, or is close too. it remains to be seen just how successful Amazon will be with the games developed on Lumberyard, but there are some starts that look to hold some promise.  will be interesting to see how the “Twitch TV: support plays out...  or “cloud” support.  >.> all the above said...  any game studio without this level of structural support, is always going to be lacking & failing to keep up with those that do.