QCOM · Forward model · QCT Handsets · Bear case
What has to happen in QCT Handsets
Model as of
This page changes QCT Handsets inside the complete QCOM model. Every other vertical, corporate programme and valuation assumption stays in the selected portfolio case.
Shares this vertical and portfolio case. Slider and horizon edits stay in your browser.
QCT Handsets
Basis quarter$5.09B
Final quarter$2.33B
Implied CAGR−18%
Final revenue mix24%
Snapdragon platforms, stand-alone modems and the RF front end, transceiver and connectivity content that ships with them into smartphone OEMs. The largest single line and the one in structural decline: Apple is insourcing its modem, and the December flagship build is simultaneously the seasonal peak and the quarter management says Apple halves.
Last four quarters
2025 Q4
$6.96B
Reported
2026 Q1
$7.82B
Reported
2026 Q2
$6.02B
Reported
2026 Q3
$5.09B
Reported
Snapdragon mobile platforms (integrated SoC)Stand-alone Mobile Data Modems (MDM/thin modem)RFFE, transceiver, PMIC and connectivity content attached to handsets
Sequential growth
−5.0%/qtr
decaying toward +0.5%
-5.0% deseasonalised. Prints $5.15B for the guided September quarter, against the $5.14B triangulated from QCT guidance.
QCT Handsets
Latest: $2.33B (2030Q3E)
| Period | Value |
|---|---|
| 2023Q3 | $5.25B |
| 2023Q4 | $5.46B |
| 2024Q1 | $6.69B |
| 2024Q2 | $6.18B |
| 2024Q3 | $5.90B |
| 2024Q4 | $6.10B |
| 2025Q1 | $7.57B |
| 2025Q2 | $6.93B |
| 2025Q3 | $6.33B |
| 2025Q4 | $6.96B |
| 2026Q1 | $7.82B |
| 2026Q2 | $6.02B |
| 2026Q3 | $5.09B |
| 2026Q4E | $4.99B |
| 2027Q1E | $4.03B |
| 2027Q2E | $3.55B |
| 2027Q3E | $3.20B |
| 2027Q4E | $3.31B |
| 2028Q1E | $3.52B |
| 2028Q2E | $3.15B |
| 2028Q3E | $2.86B |
| 2028Q4E | $2.97B |
| 2029Q1E | $3.16B |
| 2029Q2E | $2.84B |
| 2029Q3E | $2.58B |
| 2029Q4E | $2.68B |
| 2030Q1E | $2.86B |
| 2030Q2E | $2.56B |
| 2030Q3E | $2.33B |
Assumptions & reasoning
- The December quarter is modelled as two opposing forces, and both are in the spec. The seasonality array lifts the December quarter 8.3% above the year's mean, and a separate -22% level step lands in that same quarter for the Apple decline Palkhiwala guided: 'we are forecasting approximately 50% decline from September to December quarter'. The step is the larger of the two, so the printed December line falls rather than recovers. Encoding the Apple cut as a growth rate instead would have let the seasonal factor manufacture a December uptick.
- The step is worth about $1.2B: the December quarter prints $4,278M where an unstepped path would have printed $5,485M, inside the $1.0-1.5B the research bounds from the derived ~$8.0B of fiscal 2026 Apple product revenue and the guided 50% September-to-December decline. That $8.0B is the single most consequential inference on this page: it is backed out of Palkhiwala's bridge - 'We expect this growth from non-handset revenues in fiscal '27 to replace total Apple product revenues in '26' - against a derived $13.35B non-handset base, and it is a floor, because the 60% growth it comes from is itself stated as a floor.
- Seasonality is carried and is the tightest in the dataset: fiscal 2024 factors 1.076/0.994/0.949/0.981 and fiscal 2025 factors 1.090/0.997/0.911/1.002 agree to 0.014 on the December peak and 0.003 in March. Fiscal 2023 is excluded as a distorted window (the Apple/OEM inventory correction and the RFFE re-segmentation recast), and fiscal 2026 cannot be a window at all - its 7,824 to 6,024 to 5,086 collapse is a demand and share shock, not a shape. Mechanism is disclosed in the 10-Q: revenues fluctuate 'in advance of and during device launches incorporating our products'.
- This line is NOT split into Apple and non-Apple, and it never should be from this disclosure. The fiscal 2025 10-K names Apple, Samsung and Xiaomi as the customers at 10% or more of revenue and separately reports three unnamed customers at 21%, 20% and 13%, but it never maps a name to a letter. Any Apple revenue line on this page would be manufactured.
- No unit driver exists. Qualcomm publishes no chipset units and no ASP anywhere in the release, 10-Q or 10-K, and discontinued Total Reported Device Sales years ago; the 10-Q gives only dollar-denominated price-versus-volume bridges. A growth rate off a disclosed revenue base is the only honest kind here.