You can answer this in the session instead of driving around listening for problems.
Why
Translation is measurable before it is audible elsewhere
Playing a mix in the car, on phone speakers, and in headphones tells you something is wrong but rarely tells you what. By the time you hear it, you are guessing which decision caused it.
The same failures show up again and again: low end that only exists in stereo, width that collapses when summed, and a master pushed so hard that quiet detail disappears on any noisy system.
Check
Four checks, in order, before export
First, low-end weight: how much energy sits below 100 Hz, and is it centered? Bass that lives in the sides disappears on mono-summed systems and car subs.
Second, the mono fold: sum the mix to mono and listen for anything that thins, hollows, or vanishes. Wide reverbs, doubled parts, and stereo-widened bass are the usual suspects.
Third, width risk: check whether the exciting stereo image is decoration or structure. If the mix stops working in mono, the width is doing load-bearing work it should not be doing.
Fourth, push: compare your loudness and true peak against a commercial reference at matched playback level. If your mix only feels competitive because it is squashed, it will lose on any system with a noise floor.