I have built a number of "dream" cars over the years and have learned some valuable lessons. The biggest lesson has been that if you're careful, you can enjoy a car and eventually sell it for a small profit (or at least not lose money). Even when you have your dream car, your dream may change some day and you don't want to take a massive loss on it. "Loosecannon, can you be more specific?" you may ask. Yes, I can
1.Consider your location-You may desire a slammed Miata but you live in Alaska so even if you get the car you want, good luck selling it when you're bored. Example: I had an Outlaw(ish) air cooled 911 and it was awesome, except I live in Canada and up here, Porsche people want stock, unmodified Porsches and I lost sooo much money. If I was in California, I might have found a buyer for it
2. Age changes you-if this is a car you plan on keeping forever, consider what future you might want. Straight pipes might be cool when you're 22 years old but somewhere in your 30's, that drone on the highway is going to drive you and your fiance' crazy. Example: I had a rusty 1988 Suburban and a super cool sports car with Supertrapps on it and my girlfriend at the time (now my wife) didn't want me to pick her up in the sports car. A simple, well thought out exhaust system cured the noise problem and made it so much more enjoyable for me, and everybody within hearing distance.
3. There is unique-good and unique-bad. You might have a vision for a super unique dream car but if you're the only person on the planet with that vision, that car is going to be almost worthless. Example: a local ordered a new Porsche 911 Turbo in orange. Nothing wrong with that, it looked awesome, except they fit it with a RED interior and YELLOW seat belts. They loved the uniqueness of it but after a few years, they took a loss when selling it
4.Don't build a one trick pony AKA build it to be versatile-If you build a car to dominate any particular thing, it will inevitably be really poor at everything else. Example: The Pink Panther (my race car) has been built to win EMod class in autocross and as a result, it can't do anything else, which is why I'm building a new car which can do everything well
5. Pick base vehicles that have a healthy aftermarket-Some vehicles have very little aftermarket support and maybe you think a Restomod Trabant would be the coolest car in the world but how are you going to put a 6 speed gearbox in it? Example: My 69 GMC Suburban restomod had a very healthy aftermarket and only a few bits were made from unobtanium and as a result, I was able to build it relatively easy and make a nice profit when the decision came to sell it.
6. Consider cost of build VS what the finished project will be worth-I guess this whole post could be boiled down to this one rule, but that wouldn't be fun, would it? In the year or so before I bought my current project car (67 Camaro) my wife and I made a list of dream builds. Top cars on that list were 510 Datsun, air cooled 911, first gen Nova and first gen Camaro. After considering the first 5 rules, we narrowed the list down to a 62-67 Nova or a 67-69 Camaro. This is where it gets interesting. Doing a complete spreadsheet of costs to build each of these cars the exact way we wanted, turns out the COST was identical (roughly $100,000) but a search of BringATrailer auctions revealed that the Camaro would be worth 2-3 times as much as an equally prepared Nova. The decision had been made for us. We have no intention of selling the Camaro but we have learned that desires change with time and if we ever decide to sell, the Camaro will be no problem.
I have to go to work at my side gig now so I can afford to build my Camaro, hope this helps