Watch this answer as a video:
Elizabeth, in Australia, because of our pissweak consumer laws, you need to start by eliminating the brands that like to bend you over and stick it to you unlubricated in the event of a problem. No matter how carnally attracted you are to that Jeep Grand Cherokee, that Holden Captiva or Cruze, that Volkswagen Golf - it’s always a bad idea.
You generally also eliminate the brands that are asleep at the wheel - like Honda and Nissan. And then you eliminate the brands with the crappy warranty and short service intervals - and I’d suggest you’re doing yourself a favour there because there’s too much choice in the market in any case. And I’d be eliminating the cars from companies with nearly no traction in the market - like Peugeot or Citroen, Volvo, MINI - like that, unless of course you’re a tragic Francophile, or something - and also eliminate the obvious dogs, like Ssangyong and Great Wall.
Then I look for the benchmark cars in every segment - like, if you’re looking for a five-seat, medium sized SUV you look at maybe the Mazda CX-5, the Hyundai Tucson and the Kia Sportage. You don’t have to buy one of them - but you’re doing your choice a disservice if they’re not on your list, because benchmarking rocks. So, what you do is assess the car you want - the one you are perhaps carnally attracted to - against the benchmarks.
Personal preference is important - vital - but you need to make a list of the things about your preferred choice that are actually superior to those offered by the benchmarks. And a list of the things the benchmarks offer you, which your option misses out on - and then you have to make friends with both lists. It’s just like being married - you need to love the things you love about your other half, and come to terms with the things you don’t.
If you don’t do this, your choice is simply not being built on solid foundations.
If you don’t do this, you’re going to have 20 or 30 cars on your list, and you’ll be stalled on the grid because you lack any roadmap - the underlying robust logic for saying sayonara to cars that really don’t stack up to objective analysis. If you want to get this right, you don’t have to buy a benchmark car - but you really do need to identify and decide why the car you want is superior for you.
Otherwise, I’d suggest, you’re about to blow it. If you do it minus the logic you end up with a dodgy short list, you buy the wrong car. And then it’s complete social disintegration - Planet of the Apes, Apocalypse Now, Night of the Living Dead, dogs and cats living together. And that never ends well.