The other big Q is how many of these bullet bc's are averaged out per vel/ranges vs just being given the high number for the initial velocity, ie marketing at work! ?
In the days of G1 BCs being the norm, many bullet manufacturers published a single BC value for each bullet, and apparently the most favourable value at that - and since G1s produce
very large ranges of values, that could make the figure pretty well useless. Given that few people shot over longer ranges than 600 yards even at targets and even fewer knew their MVs accurately (or even closely in many cases), that piece of inaccuracy didn't make much practical difference to on-range performance / sight settings. (Many rifle sights, both iron and optical, had so much inbuilt error in any case until recently that this was another major factor for the wannabe long-distance marksman.) The main effect as you surmise was on sales as Bullet manufacturer A often produced products with apparently much higher BCs than equivalents from maker B. This was particularly valuable to those firms that stuck relatively pointy impressive-looking plastic tips onto their bullets as that was what the punters ascribed the unusually high claimed BCs to making them more credible.
In those times, and more recently than many might now think, the most honest people were Sierra whose velocity-banded values (usually three bands, but as many as five in more recent introductions) gave as good predictions as you'd likely get while saddled with G1 for a modern BT bullet design. Most people simply took Sierra's values and averaged them, and if their measured or estimated MVs were about right, it worked pretty well on the whole. Problems only arose when people wanted to shoot their 308 SMKs at 1,000 yards and the terminal speeds were well into the trans-sonic band, worse nudging the speed of sound - all bets were off then especially for some bullet designs.
What changed, or at the very least initiated change, was Bryan Litz's invaluable (but very readable to the layman like me) tome
Applied Ballistics for Long range Shooting, now on its third edition and still a must-read for anybody who wants to understand what happens after the bullet leaves the muzzle. Bryan demolishes the case for G1 even with banded values and gave actual measured averaged long-range G7 BCs for most then available US match bullets and a few foreign models from Lapua and suchlike. Since then Bryan has expanded the library vastly with data for several hundred bullet models in his series of editions of
Ballistic Performance of Rifle Bullets and even more so in AB's apps
. Once that data was in shooters' hands, the ability of manufacturers to make excessive claims to gain any market advantage was finished.
Note though, these are still
averaged values. For the typical long-distance shooter in known distance competitions, with an accurate MV to hand, and with 'sighters' allowed, I'd say they are more than good enough. (I had no trouble predicting 308 Win FTR come-ups from my 100 yards zero for August heat Raton in the US nationals / FCWC meeting back in 2013 up to 1,000 yards despite the huge external ballistics differences caused by the very different conditions - temperature and even more so altitude - from my usual venues in the UK.)
For disciplines that don't allow 'sighters', are very long-distance; shoot at estimated distances; etc, etc, something still better is needed which is now provided by the recent advent of professional grade Doppler radar kit being available and relatively affordable which is what all major bullet makers and ballistics consultants now use. As the actual bullet speed is known at every point on a very long-distance trajectory, CDMs (Customised Drag Models) are now available, and if you're into say ELR where a first shot is at 1,500 yards or further, and is a score shot, then Applied Ballistics LLC, for one, will do you a PDM, or Personalised Drag Model over various ranges using your own rifle, scope, and handloads. Scope adjuster accuracy and consistency is in a different league today from those of not that many years ago (turret click actual values being 5-10% 'out' was common even with expensive models), but even so Mr Litz and AB LLC still tell you that checking your actual changes against turret values is essential on a 'Tall target'. Not much good having super-accurate CDM/PDMs, MVs accurate to a few fps, and ambient conditions / wind measurements measured on super-duper expensive devices if the change your scope actually makes to your elevation / wind adjustments are even 2 or 3% 'out', it's over ELR distances, and the first shot is a shot to count!