Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
The world's second-largest connector maker by Bishop & Associates' 2024 count, Amphenol wins on custom design-in across thousands of customer platforms, run through more than 130 autonomous business units and a serial-acquisition machine; the moat is real but sits in physical parts, and a rising share of sales now rides AI data-center capex.
Amphenol's durability is design-in at scale plus an operating model rivals have not copied — a supplier its customers specify rather than a monopoly:
- Designed In, Not Bought Off a Shelf: Most of what Amphenol sells is engineered to a customer's platform — a GPU rack's copper backplane, a fighter jet's harness, a car's sensor — and qualified before production. Once a part is in, replacing it means re-qualifying the platform, so share is won at design time and held for the program's life.
- Decentralised and Acquisitive: More than 130 general managers run their own P&Ls, and the company says it has acquired more than 50 businesses in the last ten years, including CommScope's Connectivity and Cable Solutions unit for about $10.6B in January 2026. The model lets it chase small niches fast and has lifted the adjusted operating margin to 29.8% in Q2 2026.
- Exposed to the AI Build: IT datacom reached 43% of Q2 2026 sales, growing 63% organically, and Amphenol supplies high-speed copper cable and connectors into NVIDIA NVL72-class racks. TE Connectivity, Molex, Luxshare and Credo's active electrical cables all compete for the same rack content, and a hyperscaler capex pause would hit the fastest-growing line.
Moat Verdict
Amphenol is a net beneficiary of AI: the AI rack build is its fastest-growing market, and none of its moats rest on software an agent could replicate. The strongest pillar is design-in embedding across thousands of platforms, with intact design know-how, defense qualifications, scale, breadth and a hard-to-copy decentralised operating model behind it. The risk is cyclical, not structural — a hyperscaler capex pause or a shift from copper to optics inside the rack.
72.8 resilient · 65.0 vulnerable · 80/20 = 71.2 · = 71
Open a moat to read its note.
Amphenol sells connectors, cables and sensors to manufacturers; there is no user interface to learn.
A physical manufacturer: interconnect design and process know-how is not software logic (M7, the Cameco precedent); Corning rates the same pillar na.
Amphenol does not derive advantage from controlling access to public data.
More than 130 general managers run their own P&Ls under an entrepreneurial, decentralised model that rivals have not copied at the same scale. The engineers and managers still come from the pool TE Connectivity and Molex hire from, so it rates intact (M8).
Amphenol sells connectors, copper and fiber cable, antennas and sensors across nearly every end market, and CCS added structured data-center cabling. There is no attach or retention figure showing the breadth locks customers in, so it rates intact (M9).
Decades of application-specific interconnect designs give Amphenol a head start on each new platform, the same class of process and design know-how Corning rates intact. TE Connectivity and Molex hold equivalent libraries, so it is an advantage, not an exclusive dataset.
Military and aerospace connectors must be qualified to defense specifications and onto approved-parts lists, which forces re-qualification to switch suppliers. TE Connectivity, ITT Cannon and Glenair hold the same qualifications, so it bars entrants, not peers (M4).
Connectors and cables have no network effect; more customers do not make the product more valuable to each other.
Most sales are custom parts designed into a customer's platform and qualified before production, so switching means re-qualifying the platform; Q2 2026 orders of $10.7B (book-to-bill 1.23) are largely that design-in pipeline, across more than 130 business units and every major end market. Strong on the same design-in fact as Corning, Credo and Eaton.
Amphenol makes physical components; nothing downstream defers to it as a source of truth.
Amphenol is one of the two largest interconnect makers, but TE Connectivity operates at comparable scale (Bishop ranked TE first in 2024 connector sales), so the cost position is competitive rather than a structural lead; the margin edge comes from the operating model, not unit cost.
Buyers are engineers specifying parts on performance, qualification and switching cost, which the other pillars rate. The name carries reputation, not a price premium it could hold on brand alone (M10).
Combined average of Moat (AI Resilience), Growth, and Valuation scores.
Moat Score
The world's second-largest connector maker by Bishop & Associates' 2024 count, Amphenol wins on custom design-in across thousands of customer platforms, run through more than 130 autonomous business units and a serial-acquisition machine; the moat is real but sits in physical parts, and a rising share of sales now rides AI data-center capex.
Growth Score
Amphenol grew FY2025 sales 52% to $23.1B (38% organic), with adjusted EPS up 77% and a record $4.4B of free cash flow (103% of net income). 2026 has accelerated: Q1 sales $7.6B (+58%, +33% organic, book-to-bill 1.24) and Q2 $8.8B (+55%, +30% organic, orders $10.7B, book-to-bill 1.23), with adjusted operating margin at 29.8%. IT datacom was 43% of Q2 sales and grew 63% organically; Harsh Environment Solutions grew 22% organically and Interconnect and Sensor Systems 13%. The CommScope CCS deal, closed in January for about $10.6B and expected to add about $4.1B of 2026 sales, accounts for much of the gap between reported and organic growth. Q3 is guided to $9.3–9.4B of sales (+50–52%) and $1.40–1.42 of pre-split adjusted EPS ($0.70–0.71 after the two-for-one split of September 2026). Q3 results are due October 28.
Valuation Score
At $87.24 (October 9 close, ~$215B market cap, post the two-for-one split), Amphenol sits between the $68 bear and $97 base, about two-thirds of the way from bear to base, for a valuation score of 73. That is roughly 33× a ~$2.66 2026 adjusted EPS and ~27× a ~$3.23 2027 consensus — a premium to Amphenol's own history, priced for the AI build to keep going.
The Interconnect Roll-Up
Amphenol's durability is design-in at scale plus an operating model rivals have not copied — a supplier its customers specify rather than a monopoly:
- Designed In, Not Bought Off a Shelf: Most of what Amphenol sells is engineered to a customer's platform — a GPU rack's copper backplane, a fighter jet's harness, a car's sensor — and qualified before production. Once a part is in, replacing it means re-qualifying the platform, so share is won at design time and held for the program's life.
- Decentralised and Acquisitive: More than 130 general managers run their own P&Ls, and the company says it has acquired more than 50 businesses in the last ten years, including CommScope's Connectivity and Cable Solutions unit for about $10.6B in January 2026. The model lets it chase small niches fast and has lifted the adjusted operating margin to 29.8% in Q2 2026.
- Exposed to the AI Build: IT datacom reached 43% of Q2 2026 sales, growing 63% organically, and Amphenol supplies high-speed copper cable and connectors into NVIDIA NVL72-class racks. TE Connectivity, Molex, Luxshare and Credo's active electrical cables all compete for the same rack content, and a hyperscaler capex pause would hit the fastest-growing line.
Moat Verdict
Amphenol is a net beneficiary of AI: the AI rack build is its fastest-growing market, and none of its moats rest on software an agent could replicate. The strongest pillar is design-in embedding across thousands of platforms, with intact design know-how, defense qualifications, scale, breadth and a hard-to-copy decentralised operating model behind it. The risk is cyclical, not structural — a hyperscaler capex pause or a shift from copper to optics inside the rack.
72.8 resilient · 65.0 vulnerable · 80/20 = 71.2 · = 71
Open a moat to read its note.
Amphenol sells connectors, cables and sensors to manufacturers; there is no user interface to learn.
A physical manufacturer: interconnect design and process know-how is not software logic (M7, the Cameco precedent); Corning rates the same pillar na.
Amphenol does not derive advantage from controlling access to public data.
More than 130 general managers run their own P&Ls under an entrepreneurial, decentralised model that rivals have not copied at the same scale. The engineers and managers still come from the pool TE Connectivity and Molex hire from, so it rates intact (M8).
Amphenol sells connectors, copper and fiber cable, antennas and sensors across nearly every end market, and CCS added structured data-center cabling. There is no attach or retention figure showing the breadth locks customers in, so it rates intact (M9).
Decades of application-specific interconnect designs give Amphenol a head start on each new platform, the same class of process and design know-how Corning rates intact. TE Connectivity and Molex hold equivalent libraries, so it is an advantage, not an exclusive dataset.
Military and aerospace connectors must be qualified to defense specifications and onto approved-parts lists, which forces re-qualification to switch suppliers. TE Connectivity, ITT Cannon and Glenair hold the same qualifications, so it bars entrants, not peers (M4).
Connectors and cables have no network effect; more customers do not make the product more valuable to each other.
Most sales are custom parts designed into a customer's platform and qualified before production, so switching means re-qualifying the platform; Q2 2026 orders of $10.7B (book-to-bill 1.23) are largely that design-in pipeline, across more than 130 business units and every major end market. Strong on the same design-in fact as Corning, Credo and Eaton.
Amphenol makes physical components; nothing downstream defers to it as a source of truth.
Amphenol is one of the two largest interconnect makers, but TE Connectivity operates at comparable scale (Bishop ranked TE first in 2024 connector sales), so the cost position is competitive rather than a structural lead; the margin edge comes from the operating model, not unit cost.
Buyers are engineers specifying parts on performance, qualification and switching cost, which the other pillars rate. The name carries reputation, not a price premium it could hold on brand alone (M10).
Growth Analysis
Growth Drivers
Key Risk
IT datacom is now 43% of sales and the main growth engine, so a 2027 hyperscaler capex digestion, or a shift from copper scale-up links to co-packaged optics inside the rack, would slow it just as CCS integration and new capacity land. Falsifiable: organic growth below 10% or a book-to-bill below 1.0 for two consecutive quarters in 2027, or adjusted operating margin back below 26%, would break the thesis. Secondary: net debt of $13.4B after the CCS deal limits room for the next large acquisition.
Score Derivation
80.3 base + 2.7 trajectory + 4 margin − 5 risk = 82
Base 80.3 (13–18% CAGR, midpoint 15.5%) + 2.7 trajectory (IT datacom and Harsh Environment accelerating; Interconnect and Sensor Systems stable) + 4 margin (expanding: adjusted operating margin 29.8% in Q2 2026 vs 25.6% a year earlier, about 29% excluding the $80M tariff recovery) − 5 moderate risk (AI data-center capex is the largest growth engine; moderate to match Corning and Eaton, since defense, industrial, auto and mobile sit outside that cycle, unlike Credo's high) = 82
Price Scenarios (12–24 Months)
Price vs Targets
Dashed lines mark the 12–24 month bear, base and bull targets. Daily closes.
Valuation Analysis
All per-share figures are post-split. 2026 adjusted EPS is about $2.66 if Q4 lands near $0.75 after Q1 $0.53, Q2 $0.675 and a Q3 guide of $0.70–0.71; consensus 2027 is about $3.23 (pre-split $6.45). The $97 base prices ~30× 2027 EPS; the $68 bear is ~24× a ~$2.85 2027 if AI datacom digests. Q3 results on October 28 reset the base. Add on weakness toward the low $70s rather than at the current price. ~$97.
Valuation Multiples
| P/E (2026E adjusted) | ~33× |
| Forward P/E (2027E) | ~27× |
| PEG Ratio | ~1.7× |
| Price / FCF (2025) | ~49× |
At ~27× next year's earnings Amphenol trades at a premium to industrials and roughly in line with Corning, which is fair only if the AI build keeps compounding earnings above 20%. The fall from ~33× 2026 to ~27× 2027 signals a continuing earnings ramp, but the stock sits at an all-time high, so the margin of safety is thin.
Approximate figures as of October 9, 2026.
Where We Are vs Targets
Loading live price…
Hyperscaler capex digests in 2027, IT datacom growth falls toward single digits, and the multiple reverts toward Amphenol's pre-AI range.
- Organic growth falls below 10% and book-to-bill drops under 1.0 in 2027
- Adjusted operating margin slips back toward 26% as CCS and new capacity run under-utilised
- 2027 EPS near $2.85 at ~24×, giving about $68
Amphenol delivers consensus 2027 earnings as AI datacom growth slows but stays strong and the other end markets compound.
- IT datacom grows 20%-plus in 2027 as next-generation racks ramp
- CCS integration lifts its margin toward the company average
- 2027 EPS near $3.23 at ~30×, giving about $97
AI scale-up networks stay copper-heavy longer than expected, Amphenol wins share in next-generation racks, and it is valued as AI infrastructure.
- Copper scale-up links hold against co-packaged optics through the next rack generation
- Book-to-bill stays above 1.15 through 2027
- 2027 EPS near $3.60 at ~36×, giving about $130