Broadcom [AVGO]
Every number used in AVGO_Research.md and AVGO_Valuation.md, with its derivation. A figure that was
computed rather than read off a filing is marked derived and the arithmetic is shown.
Broadcom files no Q4 10-Q, so Q4 revenue is derived as the fiscal-year total less the three reported quarters. Each derivation is then cross-checked against the Q4 earnings release.
| Fiscal quarter | Period end | Revenue ($m) | Source |
|---|---|---|---|
| Q1 FY2024 | 2024-02-04 | 11,961 | 10-Q |
| Q2 FY2024 | 2024-05-05 | 12,487 | 10-Q |
| Q3 FY2024 | 2024-08-04 | 13,072 | 10-Q |
| Q4 FY2024 | 2024-11-03 | 14,054 | derived: 51,574 − 11,961 − 12,487 − 13,072 |
| Q1 FY2025 | 2025-02-02 | 14,916 | 10-Q |
| Q2 FY2025 | 2025-05-04 | 15,004 | 10-Q |
| Q3 FY2025 | 2025-08-03 | 15,952 | 10-Q |
| Q4 FY2025 | 2025-11-02 | 18,015 | derived: 63,887 − 14,916 − 15,004 − 15,952. Cross-check: Q4 release states "record revenue of $18.0 billion" ✓ |
| Q1 FY2026 | 2026-02-01 | 19,311 | 10-Q |
| Q2 FY2026 | 2026-05-03 | 22,187 | 10-Q |
TTM revenue (Q3 FY25 → Q2 FY26) = 15,952 + 18,015 + 19,311 + 22,187 = $75,465m. Matches
scan_v3.
Prior-year TTM (Q3 FY24 → Q2 FY25) = 13,072 + 14,054 + 14,916 + 15,004 = $57,046m.
TTM YoY growth = 75,465 / 57,046 − 1 = +32.3%.
Q-annualised run-rate = 22,187 × 4 = $88,748m (+17.6% on TTM).
| Fiscal quarter | EBIT ($m) | Source |
|---|---|---|
| Q1 FY2025 | 6,260 | 10-Q |
| Q2 FY2025 | 5,829 | 10-Q |
| Q3 FY2025 | 5,887 | 10-Q |
| Q4 FY2025 | 7,508 | derived: FY2025 OperatingIncomeLoss 25,484 − 6,260 − 5,829 − 5,887 |
| Q1 FY2026 | 8,563 | 10-Q |
| Q2 FY2026 | 10,788 | 10-Q |
TTM EBIT = 5,887 + 7,508 + 8,563 + 10,788 = $32,746m.
TTM operating margin = 32,746 / 75,465 = 43.4%.
Latest-quarter operating margin = 10,788 / 22,187 = 48.6%.
FY2025 operating margin = 25,484 / 63,887 = 39.9% (the figure scan_v3 used; correct for its period,
stale as a terminal-margin input).
Gross profit: Q3 FY25 10,703; Q4 FY25 derived as 43,294 − 10,145 − 10,197 − 10,703 = 12,249; Q1 FY26 13,157; Q2 FY26 15,415. TTM gross profit $51,524m → TTM gross margin 68.3%. Latest quarter 15,415 / 22,187 = 69.5%.
At 2026-05-03 (all from the Q2 FY2026 10-Q balance sheet unless noted):
| Component | $m |
|---|---|
| Cash and cash equivalents | 19,628 |
| — of which time deposits | 4,003 |
| — of which money-market funds | 2,801 |
| — of which US Treasury bills (at fair value) | 2,985 |
| Short-term investments | none tagged; Broadcom reports no separate line |
LongTermDebtNoncurrent |
(62,655) |
LongTermDebtCurrent |
(2,252) |
LongTermDebt (total, tagged) |
(64,907) — ties to 62,655 + 2,252 ✓ |
| Net financial debt | (45,279) |
Leases, both stale relative to the debt date: finance lease $7m noncurrent + $3m current @2025-08-03; operating lease $1,181m noncurrent + $144m current @2025-11-02. Including all of them gives −$46,614m.
scan_v3 reported −$46,470m, reproducible as
19,628 − 62,655 − 2,252 − 7 − 3 − 1,181 = −46,470 — i.e. it included operating-lease noncurrent but
omitted operating-lease current ($144m), and mixed three balance-sheet dates. Variance to the
financial-debt basis used here: $1,191m, or 0.06% of enterprise value. Immaterial for AVGO; the same
construction is not immaterial on a small cap.
EV = 4,757.6m × $380.46 + $45,279m = $1,810,069m + $45,279m = $1,855,348m.
ConvertibleDebtNoncurrent: not present in Broadcom's XBRL. No convertible overhang.
| Measure | Value | Source |
|---|---|---|
| Common shares outstanding | 4,758m @2026-05-03 | 10-Q (CommonStockSharesOutstanding) |
| Scan value | 4,757,580,198 | scan_v3 — agrees |
| Diluted WA shares, Q2 FY26 | 4,876m | 10-Q |
| Q2 FY26 GAAP net income | $9,310m | earnings release and 10-Q |
| 9,310 / 4,876 | $1.9093 | |
| Filed diluted EPS | $1.91 | ✓ scale confirmed |
Not dual-class. No customer warrants, no convertible preferred, no contingent share issuance — so as-reported and fully diluted are the same number. This is the material contrast with AMD (320m warrant shares at $0.01) and MRVL (21.8m as-converted preferred).
Not an XBRL tag. Sourced from the quarterly earnings releases (8-K Item 2.02, EX-99.1), which state AI semiconductor revenue in the CEO quotation each quarter.
| Quarter | AI semis ($bn) | Disclosure |
|---|---|---|
| FY2024 (full year) | 12.2 | "AI revenue of $12.2 billion … grew 220 percent year-on-year" |
| Q1 FY2025 | 4.1 | "Q1 AI revenue grew 77% year-over-year to $4.1 billion" |
| Q2 FY2025 | 4.4 | "Q2 AI revenue grew 46% year-over-year to over $4.4 billion" |
| Q3 FY2025 | 5.2 | "Q3 AI revenue growth accelerated to 63% year-over-year to $5.2 billion" |
| Q4 FY2025 | ~6.2 | guided in the Q3 release; Q4 release confirms +74% YoY without restating the level |
| Q1 FY2026 | 8.4 | "Q1 AI revenue of $8.4 billion grew 106% year-over-year" |
| Q2 FY2026 | 10.8 | "Q2 semiconductor revenue from AI of $10.8 billion grew 143% year-over-year" |
| Q3 FY2026 | 16.0 (guide) | "we expect semiconductor revenue from AI to grow over 200 percent year-over-year to $16.0 billion" |
Internal consistency check. Back-solving prior-year quarters from the stated growth rates: Q1 FY24 = 4.1/1.77 = 2.32; Q2 FY24 = 4.4/1.46 = 3.01; Q3 FY24 = 5.2/1.63 = 3.19; Q4 FY24 = 6.2/1.74 = 3.56. Sum 12.08 against the disclosed FY2024 total of 12.2 — a 1.0% reconciliation gap across four independently stated growth rates. The series is internally consistent.
TTM AI (Q3 FY25 → Q2 FY26) = 5.2 + 6.2 + 8.4 + 10.8 = $30.6bn = 40.5% of TTM revenue. Prior-year TTM AI = 3.19 + 3.56 + 4.1 + 4.4 = $15.25bn. AI contribution to TTM growth = (30.6 − 15.25) / (75.465 − 57.046) = 15.35 / 18.419 = 83.3%.
Ex-AI: (75.465 − 30.6) / (57.046 − 15.25) − 1 = 44.865 / 41.796 − 1 = +7.3%.
| Quarter | Semi solutions | Infra software |
|---|---|---|
| Q1 FY25 | 8,212 | 6,704 |
| Q2 FY25 | 8,408 | 6,596 |
| Q3 FY25 | 9,166 | 6,786 |
| Q4 FY25 | 11,072 | 6,943 |
| Q1 FY26 | 12,515 | 6,796 |
| Q2 FY26 | 15,009 | 7,178 |
TTM semis $47,762m; TTM infra software $27,703m; sum $75,465m ✓ ties to TTM revenue.
Annual: FY2023 28,182 / 7,637 / 35,819 · FY2024 30,096 / 21,478 / 51,574 · FY2025 36,858 / 27,029 / 63,887. All from the 10-K and the Q4 earnings releases.
Organic semiconductor CAGR, FY2023 (2023-10-29) → TTM (2026-05-03), 2.51 years: (47,762 / 28,182)^(1/2.51) − 1 = 21.7%.
scan_v3 base: cagr_base_revenue = $33,203m = FY2022 revenue, period ended 2022-10-30 (10-K).
Endpoint: TTM revenue $75,465m at 2026-05-03. Interval 1,281 days = 3.5075 years.
(75,465 / 33,203)^(1 / 3.5075) − 1 = 26.4%
scan_v3 reports 24.378426% — bit-identical to scan_v2, whose TTM endpoint was $72,454m. The
CAGR was not recomputed when TTM revenue was corrected. Same fingerprint on AMD and MRVL.
Daily EV/Sales, 2022-03-10 → 2026-07-28, 1,099 trading days. Price from Alpaca daily bars (split adjusted). TTM revenue as known at each date, lagged to the actual 10-Q/10-K filing date so the series is not forward-looking. Current share count and net debt held constant, so the series measures multiple movement rather than capital-structure change.
| min | p25 | median | p75 | p90 | max | current | percentile | |
|---|---|---|---|---|---|---|---|---|
| EV/Sales | 7.84 | 11.37 | 17.53 | 22.99 | 27.12 | 34.24 | 24.59 | 80.3 |
Known limitation, stated rather than hidden: shares rose ~4.1bn → 4.76bn and net debt ~$28bn → $45bn across the window on the VMware financing. Holding today's capital structure constant therefore inflates the earlier part of the series, which makes today's 80.3rd-percentile reading conservative (the true historical multiples were lower, so today would rank higher).
EV/EBIT history spans only 151 trading days — a continuous four-quarter EBIT series does not exist
before the Q1 FY25 filing. Declared UNIDENTIFIED as a valuation anchor per valuation.md rather
than substituting a peer median.
assets/reverse_dcf.py, EBIT basis. Solved: revenue CAGR. Fixed: EV $1,855,348m; revenue $75,465m;
5 years; WACC 10.0%; terminal margin 40.0%; exit multiple 24.2x.
Result 32.5%, implying FY2031 revenue of $308,685m. Full sensitivity in AVGO_Valuation.md.
Terminal value is 100% of modelled EV under this construction (the model discounts only a terminal value), well above the 60% threshold, so the reverse DCF is the primary long-horizon output and the sensitivity is run over the exit multiple — never over scenario probabilities.
revenue_cagr_demonstrated is stale in scan_v3 for all three cluster names — carried over from
scan_v2 after the TTM endpoint was corrected. Understates AVGO by 2.0pp, AMD by 1.7pp, MRVL by 1.2pp.terminal_margin capped at 15.6% for AVGO (from 39.9%) and MRVL (from 16.1%) on an industry
percentile. Sole cause of "reverse DCF unsolvable in range" and the INDETERMINATE verdicts.margin_period is the prior fiscal year on all three, so gross and operating margins are one to three
quarters stale. AVGO's operating margin is understated by 3.5pp on a TTM basis and 8.7pp on the latest
quarter.entity_public_float values are stale 10-K cover-page dollar amounts, not current floats.