QUALCOMM INCORPORATED (QCOM) — FISCAL Q3 2026 EARNINGS PREVIEW

1. Earnings Preview

Key Takeaway: The setup is a low bar on handsets but a high bar on the data-center narrative — the single biggest swing factor is whether management's Q4 China Android recovery call holds and whether the Investor Day data-center targets (>$5B FY2027 revenue) get any near-term validation.

Heading into Q3 FY2026, consensus sits at roughly $9.66B revenue and $2.23 non-GAAP EPS — both modestly above the guidance midpoints of $9.6B and $2.20, leaving a thin but achievable cushion. Management's posture has shifted from cautious to cautiously constructive since the April print: the explicit Q3 bottom call on China Android, the June 24 Investor Day doubling of the FY2029 revenue target to $40B, the naming of Microsoft and Meta as data-center CPU customers, and the announcement of a second hyperscaler custom-silicon win have all reinforced the diversification thesis. Estimate revisions have been modestly negative since last earnings — Q3 EPS consensus has drifted from ~$2.20 at the post-print baseline to ~$2.23 today, essentially flat, while FY2026 and FY2027 EPS have edged slightly higher — suggesting the Street has largely absorbed the guidance miss and is now pricing in the recovery narrative. The stock has underperformed SOXX materially since the April print (QCOM roughly flat to down vs. SOXX up ~9%), trading at only ~15.6x NTM P/E versus the sector at 25–30x, implying the market has not yet priced in the data-center optionality. The key wildcard is memory supply normalization timing: if DRAM pricing shows any sign of easing into Q4, China OEM build plans could recover faster than guided, creating upside to both Q4 guidance and FY2027 handset estimates; conversely, any further tightening or a delay in the hyperscaler custom-silicon ramp could pressure the stock despite an otherwise in-line print.

2. KPIs & Consensus Expectations

Key Takeaway: Consensus is a modestly low bar on revenue and EPS given the guided range, but QCT Handset revenue (~$4.96B consensus vs. ~$4.9B guidance) is the bigger swing factor — any sign that China OEM builds are recovering faster than feared could drive a meaningful beat, while automotive (~$1.48B, +50% YoY guided) is tracking well and unlikely to disappoint.

Table 1 — Current Quarter Snapshot (Q3 FY2026)

¹ Source: Visible Alpha Consensus and Actuals Data

Table 2 — Beat/Miss History (Last 8 Quarters)

Top 2 KPIs: Total Revenue (Operating) and Non-GAAP Diluted EPS (Operating)

¹ Source: Visible Alpha Consensus and Actuals Data

Pattern: QCOM has beaten consensus on both revenue and non-GAAP EPS in every one of the last seven reported quarters, with EPS beats averaging ~5% and revenue beats averaging ~2–3%; the consistent beat pattern sets a high behavioral bar even against a low absolute guidance level heading into Q3 FY2026.

3. Guidance & Commentary Evolution

Key Takeaway: Guidance has not been formally revised since the April 29 earnings call, but the June 24 Investor Day materially upgraded the long-term financial framework — the FY2029 revenue target doubled to $40B and data-center revenue was pulled forward to FY2026 — shifting tone from cautious to structurally bullish without changing the near-term Q3 range.

¹ Source: Visible Alpha Consensus and Actuals Data

4. Guidance vs. Estimate Revision Tracker

Key Takeaway: Estimates have been remarkably stable since the April 29 print — Q3 FY2026 revenue consensus is essentially flat vs. the post-print baseline, and FY2026 full-year estimates have edged slightly higher — suggesting the Street has absorbed the guidance miss and is now anchored to the recovery narrative rather than pricing in further downside. FY2027 estimates have drifted modestly higher, reflecting early data-center revenue credit.

¹ Source: Visible Alpha Consensus and Actuals Data

Near-term estimates (Q3 FY2026 and FY2026) are essentially anchored at the post-print baseline, confirming the Street has not revised materially in either direction since April 29. The more notable move is in FY2027, where revenue and EPS estimates have risen ~4% since the post-print baseline, reflecting early credit for the data-center revenue ramp (Investor Day $5B FY2027 target) and the automotive pipeline expansion ($65B design wins, $10B FY2029 target). The gap between FY2027 consensus (~$44.2B revenue, ~$10.93 EPS) and the Investor Day long-term trajectory implies the Street is still discounting a significant portion of the data-center and diversification upside.

5. Stock Performance

Key Takeaway: QCOM has significantly underperformed SOXX since the April 29 earnings print — the stock is roughly flat to slightly up from the pre-earnings close while SOXX is up ~9% — driven by multiple compression rather than estimate cuts, as the market has not yet assigned value to the data-center optionality disclosed at the June 24 Investor Day.

Chart: QCOM vs. SOXX vs. SPY — Indexed to 100 at April 29, 2026 (Last Earnings Date)

Note: Indexed to 100 at April 29, 2026 close ($156.00 QCOM, $449.99 SOXX, $711.58 SPY). Source: Stock Price Data (Yahoo Finance).

Key Events Since Last Earnings:

• Apr 30: QCOM surged +15.1% on the day after earnings as the market digested the hyperscaler custom-silicon announcement and the explicit China Android bottom call, despite the Q3 guidance miss.
• May 22 (Bloomberg report): QCOM reported as a ByteDance ASIC customer for AI agent inference workloads; stock hit an all-time intraday high, driving the peak indexed return of ~161.
• Jun 5: Broad semiconductor sector selloff; QCOM fell ~14% in a single session as sector rotation and AI spending concerns weighed.
• Jun 24 (Investor Day): QCOM unveiled the Dragonfly data-center portfolio, named Microsoft and Meta as CPU customers, announced the Modular acquisition, and doubled the FY2029 revenue target to $40B. Stock declined ~3% on the day as the market had partially priced in the news.
• Jul 16–28: Broad semiconductor sector weakness driven by CXMT IPO in Shanghai (Chinese memory competitor), TSMC capex concerns, and sector-wide risk-off; SOXX fell ~14% from its late-June peak while QCOM fell ~18% from its Investor Day level, widening the underperformance gap.

Performance Summary (Apr 29 – Jul 28, 2026): QCOM +9.0% | SOXX +14.7% | SPY +3.9%. QCOM underperformed SOXX by ~5.7 percentage points over the period. The 1-month performance (June 30 – July 28) shows QCOM -14.0% vs. SOXX -19.0%, suggesting QCOM has actually held up slightly better than the sector in the recent selloff. The NTM P/E of ~15.6x (vs. ~12.9x six months ago) reflects multiple expansion from the data-center narrative, but remains at a steep discount to semiconductor peers at 25–30x, implying the market is still pricing in significant execution risk on diversification.

6. Material News & Developments

Key Takeaway: The most important development since last earnings is the June 24 Investor Day, which doubled the FY2029 revenue target and named Microsoft and Meta as data-center CPU customers — a structural re-rating catalyst that the stock has not yet fully priced in. The secondary wildcard is Qualcomm's plan to implement double-digit chip price increases, which could provide a meaningful near-term revenue and margin tailwind if customers absorb the hikes.

• Jun 24, 2026 — Investor Day (Most Material): Qualcomm unveiled the Dragonfly data-center portfolio (C1000 CPU, AI300 accelerator, HBC near-memory compute), named Microsoft (Azure CPU deployment) and Meta (multi-gen CPU supply) as anchor customers, announced a second hyperscaler custom-silicon win, and doubled the FY2029 non-handset revenue target from $22B to $40B (implying 40% CAGR FY2025–FY2029). Data-center revenue target set at >$15B by FY2029 and $5B by FY2027. Automotive design-win pipeline expanded to $65B (from $45B). FY2029 non-GAAP EPS target set at >$18. Implication: Structural re-rating catalyst; the market has not yet assigned full credit.
• Jun 24, 2026 — Modular Acquisition Announced: Qualcomm announced the acquisition of Modular (AI software / unified compute layer), funded with up to 19.2 million unregistered shares. Implication: Strengthens the software stack for the Dragonfly data-center platform; bolt-on consistent with stated M&A strategy.
• ~May 22, 2026 — ByteDance Named as ASIC Customer: Bloomberg reported ByteDance (TikTok parent) as a Qualcomm custom ASIC customer for AI agent inference workloads, with the deal also including chip manufacturing services. Stock hit an all-time intraday high on the news. Implication: First named ASIC customer validates the custom-silicon go-to-market; deal reportedly within current U.S. export control boundaries.
• Jul 24, 2026 — Double-Digit Chip Price Increases: Bloomberg reported that Qualcomm is telling customers prices are going up by double digits. Implication: Potential near-term revenue and margin tailwind if customers absorb the hikes; risk of demand destruction in price-sensitive mid-low tier handset segments already under memory-cost pressure.
• May 27, 2026 — Bernstein Conference: CEO Cristiano Amon explicitly defined 'material' FY2027 data-center revenue as multiple billions of dollars and described the data-center opportunity as disaggregated compute targeting inference sub-workloads (prefill and decode disaggregation). Reaffirmed 30% QCT EBT margin target. Implication: Raised the credibility bar for the data-center narrative ahead of the Investor Day.
• May 19, 2026 — J.P. Morgan Tech Conference: Management reiterated the Q3 bottom call on China Android and confirmed the hyperscaler custom-silicon engagement is on track for December quarter shipments. Implication: No change to guidance; reinforced the recovery timeline.
• Jul 15, 2026 — Q3 Earnings Date Confirmed: Qualcomm confirmed July 29, 2026 earnings release after market close, with conference call at 1:45 PM PT. Implication: No surprise on timing.
• Jul 16–28, 2026 — Sector Headwinds (CXMT IPO / Memory Competition): Chinese memory chipmaker CXMT raised $8.55B in a Shanghai IPO, becoming mainland China's most valuable company at ~$484B market cap. Reports of Chinese progress on immersion DUV lithography machines (competing with ASML) drove a broad semiconductor selloff (SOXX -14% from late-June peak). Implication: Negative read-through for QCOM's handset segment — if Chinese OEMs gain access to cheaper domestic memory, the memory-cost headwind on handset builds could ease faster than expected (positive for Q4 recovery) but also raises longer-term competitive concerns for Qualcomm's China Android exposure.
• Jul 27, 2026 — Broadcom-Samsung $200B MOU: Broadcom announced an expanded MOU with Samsung across memory and semiconductor technologies, expected to be worth >$200B over five years. Implication: Reinforces the scale of AI infrastructure investment; indirectly positive for Qualcomm's data-center and connectivity ambitions.
• Jul 22–23, 2026 — Texas Instruments Q2 Beat & Strong Q3 Guide: TXN reported Q2 EPS of $2.14 (vs. $1.93 consensus) and guided Q3 revenue to $5.65B–$6.15B (vs. $5.61B consensus), with broad growth in industrial, automotive, and data center. Implication: Positive read-through for QCOM's automotive and IoT segments; TXN cited automotive demand led by China EVs/hybrids and depleted tier-1 inventory as key drivers.
• Jul 28, 2026 — NXP Semiconductors Q2 Beat & Strong Q3 Guide: NXPI reported Q2 EPS of $3.61 (vs. $3.54 consensus) and guided Q3 revenue to ~$3.75B (vs. $3.70B consensus), with automotive guided up low-double-digits YoY and industrial/IoT up high-30s% YoY. Implication: Positive read-through for QCOM's automotive and IoT segments; NXP cited software-defined vehicle architecture transformation and physical AI as key content drivers.
• Jul 16, 2026 — TSMC Q2 Beat & Raised Capex: TSMC reported Q2 net profit up 77% YoY and raised FY2026 capex to $60B–$64B (from $52B–$56B), pledging an additional $100B for Arizona. CEO C.C. Wei stated AI demand is 'extremely robust.' Implication: Positive for QCOM's long-term access to leading-edge nodes; TSMC's packaging capacity tightness is a near-term constraint to monitor.

8. Insider Transaction Activity

Key Takeaway: All insider transactions since the April 29 earnings print are 10b5-1 planned sales — there are no discretionary open-market buys or sells, and no unusual clustering or outsized transaction sizes. The pattern is routine and does not signal any incremental insider view on the upcoming print.

Source: SEC Form 4 Filings (Insider Transaction Data)

All seven transactions since the April 29 earnings print are pre-scheduled 10b5-1 planned sales — none are discretionary open-market transactions. The CFO (Palkhiwala) is executing a regular monthly 2,500-share sale plan established in December 2025, and the CEO (Amon) sold via trust on May 4–5 in what appears to be a pre-scheduled plan. There are no open-market buys, no unusual clustering of sales ahead of the print, and no 10b5-1 plan initiations or terminations that would signal a change in insider sentiment. The absence of any discretionary activity is neutral — it neither confirms nor contradicts the constructive management tone expressed publicly.