The last mistake, and this is a doozy, is that the same critics who insist on only custom, in-house drivers, are paying for even cheaper drivers!
Makes me think of the argument made by some who’re actually involved in the making of bundled, active speakers, and how they stress that all the innards here are intricately matched. As it is though "matching" also comes down to amp downscaling because the rationale says there’s no need to use 3x200W actively for a woofer, midrange and tweeter when you can get by with 200, 100 and 50W respectively, which also conveniently cuts down cost. Sometimes there’s even a combo of Class A/B and D, but is 3x200W necessarily less of a match with 3 similar amp channels and overall topology other than (possible) excess wattages? And what if you had 3 externally configured amps of your own choice instead of built-in, cheaper variants, not to mention DAC’s? "Matching" can be a dubiously applied term, but implementation is indeed king, and active as configured externally/outwardly can be a playground of vast experimentation and great results from any segment, manufacturer, size, cost or whatever one chooses.
ATC btw. seems to be one of the exceptions to any general observation about custom, in-house drivers that are considered el cheapo, because they’re anything but. Very good amps too. And Mundorf AMT’s are certainly great in the HF region with the wider (i.e.: taller) versions also picking up on sensitivity, but actively configured with high eff. horns it’s less of, if any issue (I prefer a point source from the lower mids on up, but that’s just me).