I've used 10% here, as a bit of a magic number and I appreciate that while a hill may have bits that at >=10%, they typically don't average it for the whole length (with a few notable exceptions[1]). But even if we picked 7.5%, 10kph/80kg rider == ~221w. For a 95kg rider it would be ~255w.
Unless your hills take an hour to ride up (i.e. 1000m in one go), you don't need an *FTP* of those numbers (which is usually defined as what you can sustain for a full hour), you need a peak output of those numbers, which is a very different and much more achievable goal.
(in fact you can probably do it already in short bursts, so what you're looking to do is increase how long you can sustain it for)
Short bursts yes, but surprising just how short those bursts will be.
My best FTP was probably about 230W maybe a little more. I've never done a proper FTP test but I've got ride data from lots of rides. I can see that I did one ride that was about 85 minutes long where my one hour power (not NP or xPower, those give inflated figures) was 214W average.
Using Golden Cheetah to look at all of my other rides I see that my record for sustaining an average of 300W or more is just 3m11s. Maybe I didn't find quite the right hill in terms of length or steepness but over many rides my critical power graph looks the expected shape so I doubt it is too far off.
It's surprising just how quickly the numbers tapered off (well, definitely for a relatively untrained cyclist like me), less than a 50% increase from my FTP and I can only maintain it for about 1/20th of an hour.
What I found is that when I was too heavy, not quite fit enough, and I hit something steep (i.e. Ditchling Beacon) I'd end up stopping multiple times on the way up. No great problem. If I had needed to sustain 300W to ride up it non-stop at 6kph then doing it in chunks with rests means I just took longer and average 4kph. It's no big deal. Ride as long as you can in your lowest gear, stop and rest, repeat.
I need to lose 25kg to get back to a more sensible weight. Given I'm this much overweight I'm not worried about training for power right now (I'd probably do more if I could have a turbo but I live in a converted house [thin floors and ceilings] and have upstairs and downstairs neighbours, no garage and so nowhere to put a turbo - even if it was a quiet one). My main focus is on losing weight. Doesn't mean I couldn't do both, but adding 10% to the numerator part of W/kg is going to have a much smaller effect than reducing the denominator by 25%. Adding 10% to the numerator part once I've lost the weight will also have a bigger psychological effect too; I want to have some good performance gains once I've got rid of the weight, not have all of the easy early performance gains mixed up with the gains from losing weight (if you see what I mean).