There's a lot of detailing what the author did to try to investigate the issues. That's interesting, but if one just wants what he found, one can skip to the end, which summarizes the results:

So why is Cities: Skylines 2 so incredibly heavy on the GPU? The short answer is that the game is throwing so much unnecessary geometry at the graphics card that the game manages to be largely limited by the available rasterization performance. The cause for unnecessary geometry is both the lack of simplified LOD variants for many of the game’s meshes, as well as the simplistic and seemingly untuned culling implementation. And the reason why the game has its own culling implementation instead of using Unity’s built in solution (which should at least in theory be much more advanced) is because Colossal Order had to implement quite a lot of the graphics side themselves because Unity’s integration between DOTS and HDRP is still very much a work in progress and arguably unsuitable for most actual games. Similarly Unity’s virtual texturing solution remains eternally in beta, so CO had to implement their own solution for that too, which still has some teething issues.

Here’s what I think that happened (a.k.a this is speculation): Colossal Order took a gamble on Unity’s new and shiny tech, and in some ways it paid off massively and in others it caused them a lot of headache. This is not a rare situation in software development and is something I’ve experienced myself as well in my dayjob as a web-leaning developer. They chose DOTS as the architecture to fix the CPU bottlenecks their previous game suffered from and to increase the scale & depth of the simulation, and largely succeeded on that front. CO started the game when DOTS was still experimental, and it probably came as a surprise how much they had to implement themselves even when DOTS was officially considered production ready. I wouldn’t be surprised if they started the game with Entities Graphics but then had to pivot to custom solutions for culling, skeletal animation, texture streaming and so on when they realized Unity’s official solution was not going to cut it. Ultimately the game had to be released too early when these systems were still unpolished, likely due to financial and / or publisher pressure. None of these technical issues were news for the developers on release day, and I don’t believe their claim that the game was intended to target 30 FPS from the beginning — no purebred PC game has done that since the early 2000s, and the graphical fidelity doesn’t justify it.

In even shorter form:

• Highly-detailed models coupled with a lack of lower-level-of-detail models for viewing at a distance.

• Limited culling (avoiding drawing things that aren't actually visible onscreen).

• Some gambles made on external work-in-progress software that turned out not to be ready at release time, forcing the developer to implement their own solutions.