Well, since I was around and in the industry back then, I can offer a few random observations.
I did a lot of firmware development around the time that the Apple computer was first introduced (much of it using the same, 6502 chip) as well as quite a bit of development using IBM PCs. Never did anything with Macs back then, as it simply wasn't suitable for our purposes. I was living in Detroit, and so much of my work revolved around the auto industry - statistical quality control, simulation (on S-100 and PC) and a lot of factory floor automation, gauging, etc. (firmware on S-100 using boards loaded-up with PROMs, and on custom boards).
Apple (or, actually, Steve Wozniak, before the formation of Apple) picked a processor chip - the 6502 - from a fairly unknown manufacturer - MOS Technology, which at the time was primarily known as a second-source for Texas Instruments. The chip was revolutionary for a single reason - it's $25 price tag. A Motorola 6800 was close to $200, I don't think the 8080 was out yet, and the 8008 was also costly and needed a lot of support chips.
There's the first reason. There is no way that MOS Technology could have produced chips in the numbers needed for what would be the PC Revolution. They were too small.
In fact, I probably had a 6501 in my hands before Steve Wozniak did. I was working on a firmware project at the time - a controller for a self-serve gas pump - and we were using the Intel 4040 chip. When we heard about the $25 65xx that looked very attractive to us: an 8-bit chip with much of the system on a single chip, vs. a more expensive multi-chip 4040 set.
So, we arranged to visit the MOS Technology development and fab facilities in a converted knitting factory near Boston. We met with the chip designer, who had drawn the rough layout of the chip literally on his office wall. (On butcher paper or whatnot, not actually on the wall.) There is a blank spot on the original chip layout where the electrical outlet was on the wall. He needed to be able to plug in a lamp there, I kid you not.
We met with Chuck Peddle, who later developed the PET computer for Commodore. He gave us a tour of the fab area, we had a conference with a couple of their engineers, and then he showed us a prototype of the KIM-1 - the development board for the 6501/2 - on a bench in the lab area. (The only difference between the two was one pin assignment. The 6501 was pin-compatible with the Motorola 6800. The 6502 was not.) We returned with a single 6501 prototype chip with the metal lid soldered on (they needed to be individually probed for test at this stage), and a 9-track tape with an assembler written in Fortran that I deposited at my University's tape library, and then installed the assembler on their MTS timeshare system. I wirewrapped a prototype system and we rented an ASR-33 teletype and modem to access the timeshare system and puch paper tapes with object code.
Clearly, MOS Technology was willing to talk to anyone and everyone at that point. Intel offered us no such coddling. We were two people working for the owner of a small local chain of discount gas stations. Wozniak didn't get the royal treatment we got, though I'm sure he would have had he made-up a company name, asked, and paid for a plane ticket to Boston. I think he bought his chip off the floor of WESCON, which was the first public sale of the chips - a bit of a publicity stunt for MOS Technology.
Anyway, the chip was way cheap and easy to work with. We had an 8-bit chip that was more affordable than the 4-banger we'd been using. But if Wozniak or Jobs had visited the factory, as we did - and had realized what was coming - I doubt they would have gone with the 6502. MOS Technology just wasn't big enough.
It was fine for our purposes, though, and we were assured that we could obtain the small quantities that we needed.
I did some more work with the 6502, then 6800, but increasingly it was Intel chips that were used. The auto industry couldn't work with anybody smaller than Motorola, and they did like their chips for on-board stuff. Factory automation was almost all Intel. 6502 wasn't ever a thought in the industry.
Bottom line is the 6502 wasn't the right chip for the PC Revolution. It couldn't be. It was easy for hobbyist Wozniak to obtain and use. But, somehow, as his sideline hobby turned into a business, the chip choice seems to have never been re-thought until many, many years later. Apple had to go through two painful chip changes - first to the Motorola 68000 when increasing memory demands made them go to a 16-bit processor and it finally became clear that MOS Technology (later Rockwell) couldn't deliver the quantities needed, nor could they keep up technology-wise - then, later to Intel.
Also working against Apple was their "closed" system. First, they moved to the PC bus, then to Intel processors. But they were also "incompatible compatible". First, because they were using a different processor chip - and then because they choose to bypass the common "BIOS" implementation. Sure you could buy third-party video cards and whatnot. But the manufacturer had to load alternative firmware on the board, and only a few manufacturers had the resources and willingness to do so.
Surpisingly, I don't think Apple reached-out to hobbyists the way that Microsoft did, even though they were grounded in the Homebrew Computer Club. I was a co-founder of a Detroit-area hobbyist computer club (SEMCO) and Bill Gates spoke at our meetings several times. (I'm assuming Rick Inatome - at the time co-founder (with his dad, Joe) of then Detroit-local Inacomp, which later become Computer City and then merged into CompUSA - initially made the connection.)
I gather Gates did this often when he was traveling. Not only did he speak - in a dull, yet somehow still passionate monotone - but he eagerly took questions afterwards - until every last one was answered. (He was as bad a speaker as Inatome was good...) I don't remember if we ever had Jobs or Wozniak, but here's the takeaway - I remember Gates. He was genuinely interested in feedback and wanted to hear from anyone and everyone who was using Microsoft products. He was smart enough to know that it was worth his while to talk to a room full of hobbyists loaded-up with auto company engineers.
Apple never did that kind of outreach. Their appeal was limited. They could never have gotten into banks, manufacturing companies, etc. etc. that had specialized requirements that couldn't be satisfied on Apple's limited platform, and they couldn't deliver the needed quantities.
The turn-around from Apple's initially-limited world view has been brilliant, of course. I hope I've been able to give some insight into some of the reasons why it was IBM and Microsoft, though, that has dominated the general PC market for most of the time since the beginning of the PC Revolution.
