Broadcom AVGO

Technology · Semiconductors · Synthos Deep Dive · 2026-07-03

$360.45
Hold
Risk 6Growth 9Exponential 6Fair value $405 $250–$560

At a glance

VerdictHold — systematic Synthos tier
Price (2026-07-02)$360.45 · market cap ~$1.71T
Synthos scores (0–10)Downside Risk 6 · Growth Quality 9 · Exponential Potential 6
Synthos fair value (base case)~$405+12% · full range $250 (bear) – $560 (bull)
Street consensus$498 (high $582 / low $400; 51 Buy · 7 Hold · 0 Sell) — context, not our anchor
Valuation58× trailing EPS · 31× FY26E · 19× FY27E · EV/S 23× · EV/EBITDA 42×
Exponential Potential6/10 · Moderate-High — AI accelerator + networking ramp is still accelerating into a huge TAM, but a $1.7T cap and customer concentration cap the multibagger
TechnicalsMixed — $360, −25% off the 52-wk high, below 50-DMA, at 200-DMA, RSI 41, +36% 12-mo (SPY +21%)
ConvictionLow breadth — only 2 net-bullish voices / 8 claims in the KB; call rests on fundamentals + quant, not a broad expert panel
Position sizingSatellite, ~2–4% — own the AI ramp, but the rich multiple + cyclicality argue against a core-defensive weight
Next catalyst2026-09-03 Q3'26 earnings (Street EPS $3.22)
Single biggest riskAI-capex cyclicality + hyperscaler concentration — a handful of customers drive the accelerator boom

One-line thesis. Broadcom has become the second pillar of the AI-infrastructure buildout — custom accelerators (XPUs) and data-center networking — with FY25 revenue +24% to $63.9B, a 67% gross / 56% EBITDA margin, and ~$27B of free cash flow; the fundamentals are elite, but you are paying 58× trailing / 31× forward for a business whose growth is levered to a small set of hyperscaler capex budgets, so we own it as a satellite, not a core-defensive holding.

◆ Synthos call — Hold AVGO is a solid business largely reflected at ~$405 — fine to keep, no reason to chase; it gets interesting again below ~$344.
Downside Risk (lower = safer)
6/10 · High
Sturdy balance sheet (net-debt/EBITDA 1.1x) but 58x trailing, beta 1.43, -25% drawdown, and AI-capex cyclicality.
Growth Quality
9/10 · Very High
~60%+ forward EPS growth FY26, 67% gross / 56% EBITDA margin, ROIC ~19%, custom-silicon moat.
Exponential Potential
6/10 · High
AI accelerator ramp is still accelerating & the TAM is huge — but a $1.7T cap and hyperscaler concentration cap the multibagger.
⚖ Reverse-DCF cross-check Market-implied growth ≈ 52%/yr To justify today’s $360, earnings would have to compound roughly 52% a year for 10 years (9% discount rate). Analysts forecast ~4%/yr, so the market is pricing in MORE than what the Street expects.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
Buy — CoreOwn it as a foundation — start or add now, size it for years, let dips be gifts.
Buy — TacticalGood price + confirmed trend + a defined exit — buy the setup, not a marriage.
WatchWe want the business, just not at this price/setup — act only when the listed trigger hits.
HoldFine to keep if you own it — no reason to buy more; new money does better elsewhere.
AvoidDon't own it — the problem is the business or the expectations, so a cheaper price won't fix it.

In plain English

Broadcom makes the specialized chips and networking gear that power the "brains" behind AI — the custom processors big tech companies like Google and (per the KB) Anthropic use instead of buying everything from Nvidia, plus the switches that wire giant data centers together. It also owns a big, sticky enterprise-software business (VMware). The company is enormously profitable: it keeps roughly 39 cents of every sales dollar as pure profit and throws off huge cash.

The catch: the stock is expensive and it has been volatile — it's already down about 25% from its high this year. You're paying a premium for a great company that rides the AI-spending wave, and if that wave slows, the stock can fall hard. Our verdict is Buy, but as a smaller "satellite" position — own the AI growth, but size it modestly because the price leaves little room for error.

Here's what our three scores mean in everyday terms:

The one big worry: Broadcom's AI boom depends on a handful of huge customers continuing to spend enormous sums on AI hardware. If that spending pauses or shifts, revenue and the stock could drop sharply.


Price & moving averages 12 months · 50 & 200-day averages · 52-week range

181261342423504Jul '25Sep '25Nov '25Feb '26Apr '26Jul '2652w hi $48250-DMA 409200-DMA 361Price 36052w lo $270

Solid = price · dashed = 50-day average · dotted = 200-day average · amber = 52-week high/low. Price above both averages is an uptrend.

Bollinger Bands 20-day average ± 2 standard deviations

218289360430501Jul '25Sep '25Nov '25Feb '26Apr '26Jul '2620-day avg 384Price 360

The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.

RSI (14) momentum gauge · 0–100

705030Jul '25Sep '25Nov '25Feb '26Apr '26Jul '26RSI 39.6

Above 70 (red band) = overbought, below 30 (green band) = oversold. Currently 40.

MACD 12 / 26 / 9 · trend & momentum

0Jul '25Sep '25Nov '25Feb '26Apr '26Jul '26signal -9.3MACD -11.7

Blue crossing above amber (bars flip green) = momentum turning up; below (bars red) = turning down. Bar height = the size of that gap.

Relative performance vs S&P 500 & its sector (XLK (sector)), set to 100 a year ago

94116139162185Jul '25Sep '25Nov '25Feb '26Apr '26Jul '26XLK (sector) 142AVGO 134S&P 500 120

Solid = AVGO · dashed = S&P 500 · dotted = XLK (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.

Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate

078155233311$39BFY23EPS $27$52BFY24EPS $5$63BFY25EPS $7$105BFY26EEPS $12$174BFY27EEPS $19$232BFY28EEPS $26$275BFY29EEPS $22$139BFY30EEPS $35

Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.

Key stats an RIA wants

Price$360.45
Market cap$1,715B
P/E trailing16×
P/E FY26E / FY27E31× / 19×
EV / Sales23.3×
EV / EBITDA41.9×
Gross margin67.0%
Net margin38.8%
Dividend yield0.70%
Beta1.433
52-wk range$270 – $482
RSI(14)41
50 / 200-DMA$409 / $361
12-mo return+36% (SPY +21%)
Street target$498 ($400–$582)
Analyst grades51 Buy · 7 Hold · 0 Sell
FMP ratingB
Next earnings2026-08-05

What the experts actually said 8 traceable claims on AVGO · showing the highest-conviction voices

“Anthropic's revenue run-rate is exploding toward 10x; it lacks compute, so it inked deals with Google and Broadcom for non-Nvidia compute simply because it's already built and needed now.”
Jordi Visser Aibullishconviction 802026-04-21jordi_visser_ai-Xnsi-_mvrS0:51b5eaa934
“Q3 semiconductor revenue from AI expected to grow over 200% YoY to $16.0 billion on custom accelerator and networking demand.”
Avgo Mgmtmanagementconviction 922026-06-03AVGO-earnings-2026Q2:892297e7a2
“Quarterly cash dividend maintained at $0.65 per share, payable June 30, 2026.”
Avgo Mgmtmanagementconviction 702026-06-03AVGO-earnings-2026Q2:312d447b28

Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.

1. What it is

Broadcom Inc. (Nasdaq: AVGO), led by CEO Hock Tan, is a global semiconductor and infrastructure-software company built by serial acquisition (Avago/Broadcom, CA, Symantec enterprise, and — most consequentially — VMware in 2023). Today it runs two reporting segments: Semiconductor Solutions (custom AI accelerators / XPUs, data-center networking switches and NICs, broadband, wireless, storage) and Infrastructure Software (VMware, mainframe and security software). Fiscal year ends early November.

Revenue mix (FY2025, from filings):

The strategic engine the market cares about is custom AI silicon: hyperscalers co-design accelerators (XPUs) with Broadcom to run their own workloads rather than buying merchant GPUs, plus the Ethernet-based networking that stitches those clusters together. This is the thread the KB's highest-skill voice keys on (§2).

2. The expert thesis — thin coverage, cite what exists (traceable)

Honest disclosure: Broadcom has thin expert coverage in the Synthos KB — 8 total claims, only 2 net-bullish voices. This is not the deep, multi-voice panel behind our highest-conviction names. The verdict here is therefore primarily fundamentals- and quant-driven, with the KB used as corroboration, not the anchor. What exists:

Honest composite note. With only two independent voices, there is no broad-panel signal to lean on and no distilled cautionary voice in the sample beyond the neutral dividend note. Treat the expert layer as supporting the quant/fundamental case, not establishing conviction on its own.

3. Synthos scores & the Bull / Base / Bear cases

The one-glance judgment — three scores, 0–10, each anchored to real metrics (not probabilities we can't honestly calibrate):

Score0–10The read
Downside Risk (lower = safer)6 · Moderate-HighNet-debt/EBITDA ~1.1× and ~$27B FCF make it sturdy, but 58× trailing / 42× EV/EBITDA, beta 1.43, a −25% drawdown already in hand, and AI-capex cyclicality + hyperscaler concentration raise the risk.
Growth Quality9 · Very High~60%+ forward EPS growth into FY26, 67% gross / 56% EBITDA / 39% net margin, ROIC ~19%, ROE ~36%, and a genuine custom-silicon + VMware moat.
Exponential Potential6 · Moderate-HighThe AI accelerator + networking ramp is still accelerating (management guides AI semis +200% YoY) into a very large TAM — but a $1.71T cap and a concentrated customer base limit the multibagger.

The three cases (our own scenario model — assumptions shown; each target is a ~12–18-month fair value). We deliberately do not attach probabilities: the base case is by definition the expected path, so a weighted blend would just restate it with false precision. Instead the cases bound the range, and the scores above summarize them.

CaseKey assumptionsFair value
BullAI XPU + networking ramp sustains; VMware conversion lifts margin. FY27E EPS beats toward ~$21 (vs ~$19.4 cons); AI-scarcity premium holds a ~27× multiple.~$560 (+55%)
Base (our anchor)Estimates roughly hit — FY27E EPS ~$19.4; a durable high-growth compounder settles to a ~21× FY27 multiple as growth normalizes.~$405 (+12%)
BearAI-capex digestion / a hyperscaler in-sources or pauses; software growth stalls. FY27E EPS misses to ~$15; multiple de-rates to ~17×.~$250 (−31%)

Synthos fair value = the base case, ~$405 (+12%), with the full $250–$560 span as the honest range. Our base sits below the Street's $498 consensus — we take the AI-capex cyclicality and the rich multiple more seriously than the sell side, whose consensus already sits near the recent highs. Note the Street's own low target ($400) is right at our base. This is a tracked call — the Forecaster Scorecard grades it once it matures.

4. Exponential Potential

Synthos separates compounders (durable high returns on capital) from exponentials (accelerating, multi-baggers-from-here). AVGO is a rare case of a mega-cap that is still accelerating, which lifts it above most $1T+ names — but concentration and size keep it out of the top tier:

Exponential Potential: Moderate-High (6/10). The accelerating AI leg earns it a notch above a decelerating mega-cap, but concentration risk and the $1.7T base keep it out of the 8–9 range a small accelerating name would earn.

5. Financials (real numbers — FMP annual/quarterly)

6. Valuation — priced in or room?

There is no way to call AVGO cheap on trailing numbers (58× EPS, 23× sales, 42× EV/EBITDA, ~19× P/B). FMP's letter rating is B (overall score 3/5) — dragged down by P/E, P/B and debt sub-scores of 1, offset by ROE/ROA sub-scores of 5. The bull's defense is that EPS grows into the multiple: on live consensus the forward P/E is ~31× (FY26E) → ~19× (FY27E) — the multiple nearly halves at a flat price if estimates hit. The forward PEG (~0.86) looks reasonable because the growth is so steep, but that steepness is exactly what's uncertain. Street targets (context): consensus $498, high $582, low $400. Our ~$405 base fair value sits below consensus and near the Street's low — we discount for AI-capex cyclicality and the concentration risk that the sell side's near-record targets underweight. Not a value buy; a high-growth-at-a-full-price buy where the entry multiple leaves little margin for error.

7. Technicals (from the tech block)

8. Moat & competitive position

Broadcom's moat is a rare pairing: (1) custom AI silicon — deep co-design relationships where hyperscalers embed Broadcom IP into their own accelerators, creating multi-year, hard-to-switch design wins; (2) data-center networking — a leading position in the Ethernet switching (Tomahawk/Jericho) that AI clusters require; and (3) VMware / infrastructure software — sticky, high-margin, recurring enterprise licenses now being converted to subscription. The semiconductor side is fabless (relies on TSMC), so execution risk sits partly with the foundry. The competitive frame: Nvidia (merchant GPUs — both partner and the alternative Broadcom's XPUs displace), AMD (accelerators), Marvell (the direct custom-silicon comp), and networking peers; on software, VMware competes with the hyperscalers' own virtualization/cloud stacks.

Peer set (market cap, from data): TSM $2.25T (its foundry), META $1.48T (a customer/AI-capex bellwether), MU $1.10T, AMD $844B, ASML $682B, ADI $184B, plus smaller comps (QRVO, POWI, FORM, SMTC). AVGO commands one of the richest multiples in the group — justified only if the AI growth persists.

9. Management, capital allocation & guidance

10. Catalysts & what to watch

Thesis tripwires (what would change the call): a hyperscaler pausing or in-sourcing a major XPU program; two consecutive quarters of AI-semi deceleration below guidance; VMware growth stalling; or FCF margin compressing materially from the ~42% level.

11. Key risks

12. Verdict, position sizing & monitoring

Buy — Tactical. Broadcom is an elite AI-infrastructure business — FY25 revenue $63.9B, ~$27B FCF, 67% gross / 56% EBITDA margins, ROIC ~19%, an accelerating AI-accelerator + networking franchise, and a sticky VMware software ballast. But the price (58× trailing / 31× forward), the beta of 1.43, the 25% drawdown already in hand, the hyperscaler concentration, and — honestly — the thin expert coverage (2 voices) keep this out of the core-defensive sleeve. You own it for the AI ramp with your eyes open to the cyclicality.


Provenance & disclosures