But isn't that because you've chosen not to fit lower gears? If someone finds 90 rpm an optmum cadence on the flat, I can't see any good reason why it shouldn't also be more-or-less their optimum cadence up a hill. The reason most people climb hills with lower cadence, I've always assumed, is a kind of macho thing - people feel it's somehow unmanly or an admission of failure to have a bike with low gears, hence the frequent references to "granny gears" or "baling out". And because there's this whole culture about not using gears that are low enough to make climbing hills a pleasure rather than a test, there's little demand for them, and manufacturers and bike shops can get away with not selling them, so people don't realise it's possible, and it becomes a viscious circle.
Consider a reasonably fit cyclist climbing our notional 10% hill at 200 W. That power output - 200 W - determines the speed - 4.7 mph assuming rider+bike 85 kg. You can achieve that with different gearing and cadence: 36x28 will give a cadence of 45 rpm or 24x32 a cadence of 78 rpm. So you can, if you wish, still maintain a reasonably high cadence, by fitting lower gears.
But, of course, the gear you need for a given cadence varies with what your power output is. If you're a TdF rider, ouputting 400 W up an Alpine climb, that 36x28 would get you a cadence of (roughly) 90 rpm. I'd suggest that one of the reason many people accept shops selling them bikes with not-very-low gearing is that they, and the shop salesperson, know that that gearing is what keen racers use and assume it must therefore be right for them - without fully appreciating the difference in fitness.
In the other direction, if you're like me, pretty unfit and you just want to cycle up the hill at an enjoyably relaxed pace that you can keep up all day without knackering yourself, say less than 100 W, you'll want even lower gears to maintain a comfortable cadence. Good luck with that in your average bike shop