This is what I did for my current game, I just cut features like crazy but at the point where I am, I think it's not fun anymore.
Well, then you should analyse a little bit deeper why it is not fun anymore.
Maybe you cut a feature too much? Maybe you focused on the wrong set of features? Maybe you did everything right, but after you cut your initial starting point to size, you would have had to "put it back into shape" by refining what was left?
To come back to my example, just taking an FPS and cutting out movement does not make for an interesting game. It might even be a jarring expierience as people EXEPCT movement to be there.
You would need to add small gameplay elements to make it a fun expierience without actual movement (like the ability to duck into cover, or other small features that are inexpensive art and codewise, but could alter gameplay a lot), and maybe some art / story elements to prevent the whole missing movement thing being jarring (explain why the player cannot leave the room hes in... or let him play as a static turret for example).
By doing that, the whole things feels less like an "FPS without movement", and more like something very different.
In the end, if there is one thing I learned as a not-game-designer from reading game design books, its the simple fact that most ideas you implement are not fun initially... no matter how much time an thought you invest into planning a feature, the first iteration will most probably suck. You need to test it, analyze it, iterate on it, and generally try and error until you "find the fun"... there can be a lot of fun hidden in the weirdest ideas, and from what I can tell, the fun is not something that comes from a good idea itself, but from a good implementation of an idea...
Even bad ideas can be fun if someone spent enough time to iterate on it and polish it to perfection.
Of course, if you already spent a lot of time trying to pull off an idea, it might be just too hard to make it fun (though it might warrant a forum post here to see if fellow game devs might come up with a good way to make it more fun).
A fun fact from my own expierience: most of the time, it takes some time and exposure to a prototype to come up with the best ideas... and sometimes you just need a break. In my own case, I really struggled getting cross country driving to work with the limited physics of wheel colliders in Unity last year. I tweaked it until it was usable, but was never really satisfied with it.
over 6 months later I worked on a different topic on the same prototype and was searching the web for answers when I stumbled upon an answer that somehow gave me a good idea of how to make the wheel colliders react more realistically to bumps in the terrain... lo and behold, it worked brilliantly. Suddenly driving cross country feels just right, and it is a quite cheap method in comparison with what I tried before too.
As they say, the best ideas usually come under the shower 
... and sometimes the best idea just take some time, and yes, sometimes some distance from the project to pop up.
).
