My personal opinion is to try to create a vertical slice of your game first before spending ANY kind of money on the real thing.
Ideally you are able enough to hack together code, and create the needed 2D or 3D art... doesn't need to be too professional. If you need to outsource, only outsource THE BARE MINIMUM at first. One model, one SMALL level, the most important game logic systems. Not a finished game. Make it clear to your outsourcer you don't expect a finished product but a prototype and try to lower price and time needed this way.
Ideally, before actually outsourcing 3D models, you have a running prototype of your game with proxy models. Free models from the internet, engine example models, simple lowpoly objects created in blender in an afternoon. ONLY spend money on models AFTER you know the game concpet is viable.
What you want here is get an idea about
1) what are the EXACT specifications of your assets...
(the difference between a rather undetailed low poly model with 512x512 pixel textures compared to a high poly character model with 8kx8k Textures and an extreme amount of detail can be staggering, in time and/or money spent on creation). If you know EXACTLY what you need, you prevent paying too much for details you don't need.
2) what kind of money do you need to pay for the models you order...
Now, this is rather vague to test until you have the exact specs from point 1) and either pay a freelancer to create you one test model or do that yourself, if you have the necessary skills. Even then, time needed and price per hour will most probably vary for each freelancer, not everyone is comfortable working in every style, and if you do it yourself, you will most probably spend way longer than a professional would.
Point is to get a sense of scale. If you find yourself being able to churn out the model in acceptable quality in 2 weeks, its an asset that will most probably not cost you much if ordered from a professional freelancer. If you ask a freelancer to do it and he is ready to provide it within days for lets say 500$, you can expect a similar price (maybe 300-900$, but not 5000$) from any other freelancer.
3) If your idea is actually good and fun. Prototype, prototype, prototype.
Ideally, you get comfortable with using an engine, and with some basic coding. Create a simple prototpye of your game (if you haven't done so), with boxes and spheres as visual proxy models or whatever (most engines come with example scenes and models you could use). Try to code your game logic, and TEST IT!
Make sure you have tested your idea and found the fun in it before deciding if you should spend anymore time and money on your game idea.
Chances are high that even an idea that sounds fun in your mind actually isn't so much fun in practice.
And if coding isn't going to work out for you, ever, do a paper prototype. Create your game logic as a board game. Test it like this. But please don't spend your money on an untested game idea!
Additional things to think about:
4) the market and business realities.
Make sure you understand the business you want to enter. What is your competition doing? Do you have a chance of success? What is your preffered platform?
Make sure you do the math. How much can you spend? What if this sum is not enough? How can you either scale down your idea or build up additional funds? How much do you need to make a return on investment? Is it a realistic sum, or do you need to sell 100k units for that?
5) Marketing... You NEED Marketing skill to be successfull.
As important as a good game is, without marketing you will never, ever sell enough to make a RoI. IF you do this for the money, make sure you either have an idea yourself of marketing, or you set aside a good amount of your money to let others run the marketin for you.
2 Years is nothing when it comes to developing a game.
Trying to rush things is often a recipe for disaster. You probably did skip on the Pre-Production phase... which is the phase where you do all this testing I wrote about before. Usually, studios try to be scrappy and low profile during this phase. Only a very small team of senior devs working on it with all tools necessary. The intent is to find out IF continuing with a project all the way to production is a good idea, with failure to prove it a good idea a very likely result.
Point is, before you spent all this time and money on a game, make sure you spent enough time validating if what you are doing here actually is a good idea.
It would be VERY GOOD if you could try to at least aquire SOME skills related to game development before you pay others to develop games for you. You should at least be able to do some basic prototyping yourself, and be able to talk to devs in their language.
IDK what happened between you and the two studios you worked with, but I find it very believable that misunderstandings had at least some part to play in the failures.
Yes, learning these skills takes time. On the plus side, if you are able to put together a crappy prototype in Unity for example, you might be able to show that to the studio that should produce the finished game for you... no better way to communicate intent than to show a prototype of a running game.
