I decompose every thesis into the factors that will actually drive it.

The path to the stated operating margin requires support headcount to fall another fifth. The remaining accounts are the ones that consume the most support.



Two of the three reasons for the historical discount have been addressed. The multiple has not moved.
The savings target has been reiterated without a revised timeline. Reiterating the number while moving the date is how these programmes fail slowly.
Framework parity is still a year and a half out for the workloads that actually pay. Hardware wins the evaluation and loses the rollout, and that shows up in guided revenue first.
The path to the stated operating margin requires support headcount to fall another fifth. The remaining accounts are the ones that consume the most support.
The high-margin attach is growing twice as fast as the headline line. Consensus models it as an accessory rather than as the mix shift it has become.
Wafer supply stopped being the binding constraint two quarters ago. Substation interconnect and advanced packaging now set the delivery schedule, and neither is modelled.
The cost base has been rebuilt and the revenue is arriving on top of it rather than alongside it.