Like the 68000 before it, the 88000 was considered to be a very "clean" design. It was a pure 32-bit system, using separate instruction and data caches (Harvard architecture), and separate data and address buses. It had a small but powerful command set, and, like all Motorola CPUs, did not use memory segmentation.
The first implementation of the 88000 design was in the 88100 CPU, which included an integrated FPU. Mated to this was the 88200 MMU and cache controller. The idea behind this splitting of duties was to allow multiprocessor systems to be built more easily; a single 88200 could support up to four 88100's. However this also meant that building the most basic system, with a single processor, required both chips and considerable wiring between them, driving up costs. This is likely another major reason for the 88000's limited success.
This was later addressed in the 88110, which combined the two chips into a single package. An additional modification made at the behest of MIT's *T project which resulted in the 88110MP, including on-chip communications for use in multi-processor systems. A superscalar version capable of speeds up to 100 MHz was planned as the 88120, but was never built.
A major architectural mistake was that both integer instructions and floating-point instructions used the same register file. This required that the single register file to have sufficient read and write ports to support both the integer execution unit and the floating-point unit. The connections for each port is an additional capacitive load that must be driven by register memory cell. This made it more difficult to build high frequency superscalar implementations.
In the late 1980s several companies were actively watching the 88000 for future use, including NeXT, Apple Computer and Apollo Computer, but all gave up by the time the 88110 was available in 1990. The 88110 made it into some versions of a never released NeXT machine, the NeXT RISC Workstation, but the project was canceled along with all NeXT hardware projects in 1993. Apollo Computer implemented their own derivative (a88k) released under the name Apollo PRISM. The 4-processor OMRON luna88k machines from Japan used the m88k, and were used for a short time on the Mach kernel project at Carnegie Mellon University. A number of similar smaller systems were also built, but none are widely known.
Major users were limited. The only widespread 3rd party computer use would be in the Data General AViiON series. These were fairly popular, and remain in limited use today. Encore Computer built their Encore-91 machine on the m88k, then introduced a completely ground-up redesign as the Infinity 90 series, but it is unclear how many of these machines were sold. In the early 1990s Northern Telecom used the 88110 as the central processor in its DMS SuperNode family of telephone switches. All of these users were forced to move to other processors when Motorola later gave up on the m88k; DG went to Intel, Encore to the DEC Alpha.
In the early 1990s Motorola joined the AIM effort to create a new RISC design based on the IBM POWER design. They worked a few features of the 88000 into the new PowerPC design to offer their customer base some sort of upgrade path. At that point the 88000 was dumped as soon as possible. The whole design was never really as sufficient as what was implied.