They just printed 30% operating margins, admitted lead times are lengthening, and then guided the second half down 19%.
Thesis: This is a 'Beat and Raise Machine' disguised as a cyclical peak. The street is misreading the 2H guidance down-shift as demand softening, when the evidence (lengthening lead times, exploding test content) points to supply constraints and management conservatism. Margins hitting 30% proves they have leverage on the mix shift to AI. You buy the complexity ramp; the unit volume noise is a distraction.
Verdict: LONG — Conviction: HIGH
Catalyst: HBM4 transition driving test intensity higher (more stacks, more insertions), forcing them to beat their own sandbagged 2H guidance.
Key Risk: TSMC CoWoS or HBM yield issues at the foundry level causing a 'dark period' in demand that Advantest cannot control, validating the weak 2H guide.
The Tell: The contradiction in Nakahara's commentary. He claims 'delivery requirement is changing every week' and lead times are 'lengthening,' yet they guide revenue *down* sequentially. You don't guide down when you can't ship fast enough unless you are aggressively sandbagging to manage expectations.
Friction Level: MODERATE_FRICTION — The disconnect between 'lengthening lead times' (supply constraint) and the 19% sequential revenue decline implied in 2H guidance. Bulls call it sandbagging; Bears call it an air pocket.
Report not found
The report data is no longer available. Please return to the archive.