Concentrating on things
I would concentrate less on concrete things and instead include a few policy decisions:
1. Upper-Right — the decision by upper management to fleece customers for as much as possible started early on. Rather than trying to build marketshare, Apple management went for short-term profit. Andy Hertzfeld outlines this from the very beginning in his Folklore.org website. The original Macintosh was to be billed at $1500 and that became $2500 after Sculley became involved.
The likes of Jean-Louis Gassée then promoted the idea that because Macs were desirable that they should be milked for all they were worth. The apogée of this was the Mac IIfx which went for $10K in 1990. Despite the high prices, Macs had a 12% marketshare in 1992. Just think what it could have been if Macs had been more reasonably priced.
The pricing in Europe was also much, much higher than in the US and this hurt marketshare as well.
2. Lack of Direction in Models
Steve Jobs highlighted this in 1998. There were 40 or so different models marketed between 1994 and 1997, many of which were the same. The crassest example of which was the LC475 | Performa 475 | Quadra 605. Each was aimed at a different market and each was priced differently despite being the same machine. Then there was the awful Performa name.
3. Games
Games, especially 3-D shooters, became the killer app in the mid-1990s for PCs. Apple were already on the road to nowhere, did little to make the mac more gamer-friendly and game-developer friendly. I myself migrated from the mac in 1998 for this very reason. Half Life never came out on the Mac and I was not happy to wait for 6 months while other games on the PC were already available.
It is a mindset within Apple, I think, that goes way back to the early 1980s. A developer had written a lovely little game (Through the Looking Glass) for the nascent Mac and Apple were reluctant to market it because they din't want the Mac to be seen as a games machine.
4. Symmetric Multi-Processing
I often wonder why Apple didn't go down the road of SMP with the early Mac IIs. Steve Jobs had purchased 1 million 68000 processors from Motorola and it took Apple 6 years to use these up. The last mac to use one was the Classic in 1990. I would have required rewriting the System and Finder as well obliging companies like Microsoft and Adobe to rewrite their software to take advantage of it, but it would have made already powerful (for the day) computers more so. It would have given Apple a great technological advantage and more firmly established the Mac as a powerful computer. The nearest mac users got to this was the Radius Rocket form ex-Mac Developers Andy Hertzfeld and Burrell Smith