As of: mid-August 2026 (post FY2026 Q3 print, Jul 30, 2026)
Horizon: FY2026 Q4 (Sep quarter) and FY2027 Q1 (Dec quarter)
Focus: Services gross margin — what filings show vs what the market is pricing
Apple’s Services engine is still compounding — $30.7B in Q3 FY2026 (+12.1% YoY), ~75.6% Services gross margin, and ~26% of FY2025 sales at 75.4% GM — but the story the Street is trading is no longer Services margin expansion. Post-print narrative is almost entirely Products: memory BOM inflation, tariff-refund optics rolling off the 50.1% company GM print (~2 ppt one-time), and a soft FQ4 revenue guide on component shortages [1][2][3].
What consensus appears to be missing:
Bottom line: Consensus is correctly skeptical of company GM near term (refunds out, memory in) but is mis-locating the Services GM debate — underweighting licensing/regulatory mix risk and ad/scale mix upside — while anchoring to a flat ~75–76% Services GM mental model that filings show is already moving quarter to quarter on mix [2][4][5].
| Period | Services sales | Services GM % | Products GM % | Company GM % |
|---|---|---|---|---|
| FY2024 | $96.2B | 73.9% | 37.2% | 46.2% |
| FY2025 | $109.2B (+13.5%) | 75.4% (+150 bps) | 36.8% | 46.9% |
| Q1 FY2026 | $30.0B | 76.5% | 40.7% | 48.2% |
| Q2 FY2026 | $31.0B | 76.7% | 38.7% | 49.3% |
| Q3 FY2026 | $30.7B (+12.1% YoY) | 75.6% | 40.1% | 50.1%* |
*Q3 company GM includes ~2 percentage points from tariff refunds (~$0.11 EPS); clean company GM ≈ 48.1%, in line with the prior 47.5–48.5% guide [1][2][5].
Nine months FY2026: Services sales $91.7B (+14.1%), Services GM 76.3% [2].
Apple discloses Products vs Services sales and cost of sales (hence GM $ and %), but does not break out App Store, advertising, AppleCare, cloud, licensing, or content GM. Deferred revenue is a single current line ($9.5B at Jun 27, 2026 vs $9.1B at FY2025 year-end) — steady, not a step-change [2][8][9].
Implication: Apple has already told the market Services GM is a mix residual, not a one-way escalator. Q3’s −110 bps QoQ move is the first recent data point that the Q2 mix high was not a new floor.
High-margin legs (advertising, platform commissions net of fees, licensing, much of AppleCare) vs lower-margin legs (TV+/Music/sports content, cloud COGS, regulated/lower take-rate App Store paths) determine the blended 75–77% band.
Q2 call: advertising grew YoY; additional App Store search ads; Apple Maps ads in US/Canada for key search/discovery moments [4]. Ads are among the highest-GM Services dollars. Into FQ4/FQ1 (install and discovery season), this is a concrete mix support that consensus treats as background ecosystem color rather than a GM lever.
Street FY2026 company GM ~48.6% and a clean run-rate mentally ~47.5–48.5% after refunds [3][10]. That debate is valid for the P&L and multiple — but it is not a Services GM debate. Q3’s 50.1% headline was Products/tariff math; Services GM must be read from Services sales/COS only [1][2].
2.5B active devices; paid and transacting accounts at all-time highs [4][1]. Supports Services revenue durability at low-double-digit growth; does not automatically expand GM if mix shifts toward content or lower-take-rate distribution.
| Risk | Timing | GM direction | Why it’s under-priced now |
|---|---|---|---|
| Google search/default revenue-share re-pricing under DOJ remedies (exclusivity banned; limits on conditioning RS on long exclusive default) [6] | Contract roll / compliance — live over coming quarters | Down, high $ at near-100% incremental GM | Open narrative post-print ignores it; classic latent Services risk with no Apple-disclosed sensitivity |
| US external purchase links / Epic guideline regime (in force from May 2025) [11] | Ongoing slow burn | Down on effective take rate | Largest geography; friction still high so impact muted — market treats as “solved/old news” |
| EU DMA fee architecture (alt distribution/payments; CTF → single CTC-style model path around Jan 1, 2026) [12][13] | Already live; structural consolidation dated | Mild down / mixed (Apple retains acquisition/store/CTC-like fees even on “external” paths) | Developer uptake still limited; therefore ignored — but any acceleration of steering hits high-margin commission mix |
| Content / sports COGS vs ads/licensing growth | Continuous; multi-year rights | Mild down | Services COS Q3 $7.49B vs $6.70B prior (+~12%), in line with revenue so far — complacency risk if growth slows faster than COS [2] |
| AI / Private Cloud Compute in Services COGS | Ongoing post–Siri AI push | Down if in COGS | Street stresses R&D (+32% YoY to ~$11.7B in Q3) and product credibility, not Services COGS [1][3] |
| Mix mean-reversion from 76.7% | Already started in Q3 | Down toward mid-75% or below | No public Services GM % consensus; models hold flat 74–76% by habit [3] |
Magnitude anchor (licensing): Trial/secondary reporting has long placed Apple search RS in a ~$18–20B/year zip code (not an Apple filing figure). At ~100% incremental GM, a mid-single-digit % rate cut is tens to low hundreds of millions per quarter — enough to move Services GM tens of bps without any App Store drama [6]. That asymmetry is the single largest unmodeled Services GM risk over a two-quarter window if renegotiation lands.
What is not a near-term Services GM cliff (per evidence): DoJ smartphone case remedies (multi-year); binary “EU kills App Store” (fee menus show Apple still monetizes alt paths); tariff refunds (company/Products, not Services) [1][6][12].
Negative / clarifying
Positive / under-owned
| Consensus default | Filings / primary evidence | Miss |
|---|---|---|
| Services GM “stable mid-70s,” not a live factor | Q2 76.7% → Q3 75.6%; CFO says mix drives it and won’t guide path [2][4] | Path dependency and ±100 bps quarterly volatility are real; flat model is lazy |
| Margin risk = memory + tariff refund fade | Refunds are company-level; Services COS line is the clean read [1][2] | Services GM and company GM have decoupled in the narrative |
| Regulatory App Store risk is long-dated / priced in 2024 | US links live; EU fee model still evolving into 2026; Q3 growth headwinds already cite model changes [7][11][12] | Slow bleed into mix, not a distant binary |
| Google deal is background “Apple gets paid” | Remedies reshape exclusivity and RS conditioning now [6] | Highest-GM Services dollars face re-pricing optionality the tape isn’t trading |
| AI is an opex/R&D and product story | True so far (R&D spike) — but inference could hit Services COGS as usage scales [1] | Asymmetric left tail not in Services GM bridges |
| Installed base = automatic Services GM support | Base supports revenue; GM needs ads/licensing/AppleCare > content/regulated [4][5] | Growth ≠ margin mix |
| No need for Services GM estimate discipline | Apple does disclose Services GM every quarter via COS [2] | Street rarely publishes forward Services GM %; debate never forms |
Stance on the Services GM setup (two-quarter view): Hold the “stable ballast” assumption only with an explicit mix bar-bell — lean slightly defensive on Services GM bps, constructive on Services revenue.
Rationale
Monitor next: FQ4 Services GM % and YoY growth; any call mention of advertising, App Store terms, or licensing; clean company GM ex one-times vs Street ~48% handle; and primary reporting on search-deal implementation under the remedies order [1][6][10].
[1] Apple Newsroom, “Apple reports third quarter results,” Jul 30, 2026 — https://www.apple.com/newsroom/2026/07/apple-reports-third-quarter-results/
[2] Apple, FY2026 Q3 Consolidated Financial Statements PDF — https://www.apple.com/newsroom/pdfs/fy2026q3/FY26_Q3_Consolidated_Financial_Statements.pdf
[3] Market/web synthesis of post-print narrative (guide, memory, ratings) — e.g. https://247wallst.com/investing/2026/08/07/prediction-can-apple-stock-reach-400-this-year ; https://www.tikr.com/blog/apple-aapl-stock-dips-analyst-downgrade-glass-iphone-canceled
[4] Apple Q2 FY2026 earnings call transcript — https://www.fool.com/earnings/call-transcripts/2026/04/30/apple-aapl-q2-2026-earnings-call-transcript/
[5] Apple Q2 FY2026 materials / GM bridge context; Q3 secondary call insights — https://www.apple.com/newsroom/pdfs/fy2026q2/FY26_Q2_Consolidated_Financial_Statements.pdf ; https://seekingalpha.com/news/4622392-apple-projects-9-percentminus-11-percent-september-quarter-revenue-growth-amid-rising-supply
[6] U.S. Department of Justice, remedies outcome vs Google search distribution — https://www.justice.gov/opa/pr/department-justice-wins-significant-remedies-against-google
[7] Q3 call themes via secondary (App Store model changes, gaming, FX as Services headwinds) — https://seekingalpha.com/news/4622392-apple-projects-9-percentminus-11-percent-september-quarter-revenue-growth-amid-rising-supply
[8] Apple Form 10-Q, quarter ended Mar 28, 2026 — https://www.sec.gov/Archives/edgar/data/320193/000032019326000013/aapl-20260328.htm
[9] Apple Form 10-K, fiscal year ended Sep 27, 2025 (Services definition, risk framework) — https://www.sec.gov/Archives/edgar/data/320193/000032019325000079/aapl-20250927.htm
[10] Consensus snapshot (revenue/EPS/GM aggregation) — https://stockanalysis.com/stocks/aapl/forecast/ ; https://www.marketbeat.com/stocks/NASDAQ/AAPL/earnings/
[11] Apple Developer, U.S. App Review / external link guideline updates — https://developer.apple.com/news/?id=9txfddzf
[12] Apple Newsroom, EU DMA changes (iOS, Safari, App Store) — https://www.apple.com/newsroom/2024/01/apple-announces-changes-to-ios-safari-and-the-app-store-in-the-european-union/
[13] Apple Developer, DMA / Core Technology Fee & EU alt distribution — https://developer.apple.com/support/dma-and-apps-in-the-eu/ ; https://developer.apple.com/support/core-technology-fee/
[14] Apple FY2025 Q4 consolidated statements (FY2025 Products/Services GM) — https://www.apple.com/newsroom/pdfs/fy2025-q4/FY25_Q4_Consolidated_Financial_Statements.pdf