Outsourced assembly and test: the back end as a business
Most chips in the world are packaged and tested by companies that did not design or fabricate them. That was a commodity service for decades. As packaging became the step that determines whether high-end products can exist, these firms moved from the end of the chain toward its centre.
In one sentence
Outsourced assembly and test providers are contract manufacturers that package fabricated dies into finished components and test them, serving chip designers, integrated manufacturers and foundries.
The model matches the fabless one: a chip designer who does not own a fab has no reason to own a packaging plant either, and an assembly house serving many customers can keep expensive equipment loaded. That logic built a substantial industry concentrated in Taiwan, China, South Korea and Southeast Asia.
What changed is the top end. Advanced packaging requires capital and process knowledge closer to a fab's than to a traditional assembly plant's, and the leading foundry has kept much of the most advanced work in-house. The result is a two-tier market: high-volume conventional packaging as a competitive service, and advanced packaging as a scarce capability sold under allocation.
How it works
The range of what they do
From wire-bonded plastic packages produced by the billion, through flip-chip parts for processors, to advanced multi-die assemblies. The high-volume end is a cost and throughput business; the advanced end is a capability business where the question is whether a provider is qualified at all.
Where they sit geographically
Assembly is labour- and capital-intensive and has historically located where both were favourable, producing heavy concentration in a small number of Asian economies. Efforts to establish capacity elsewhere face the same difficulty as fabs — cost, workforce and the absence of a local supplier ecosystem.
Why allocation matters
When advanced packaging capacity is scarce, who receives it determines which products reach the market. That gives packaging providers and the foundries that operate their own lines a commercial position they did not previously have, and it is why packaging allocation is now discussed alongside wafer allocation.
What this depends on
2 of these are marked as a chokepoint: a handful of qualified suppliers, a multi-year lead time, or a single geography.
Supply chainChokepoint
Package substrates
Every assembly needs one, and the high-end grades are themselves constrained.
Handlers, sockets and probe cards are per-product tooling the provider has to hold and replace as it wears, and their lead time gates when a new part can be run.
These plants need large numbers of trained operators and technicians in one place. It is most of why the industry sits where it does, and why capacity elsewhere ramps slowly.
What depends on this
Other pages in this map that name OSAT as something they cannot do without.
What each company supplies at this step, and — where a public figure exists — its share of this specific market — with what that share measures, the period it covers and who published it. Some rows also show the company’s own reported revenue for the segment covering this step, which is a different thing: it says how much this business matters to that company, not how much of the market it holds. Not a ranking and not a recommendation.
China's second-largest packaging house, and AMD's packaging partner there.
What would change the picture
Whether advanced packaging capacity outside the leading foundry grows enough to ease allocation.
Whether packaging capacity is successfully established outside its current geographic concentration.
Whether the value split between wafer fabrication and packaging continues to shift.
Questions people ask about this
Why don't chip companies package their own products?
For most of them the economics do not work. Packaging equipment is expensive and specialised, and a company packaging only its own products cannot keep it loaded. A contract provider serving many customers can — the same argument that produced the fabless model one step upstream.
Why does packaging allocation get discussed like wafer allocation?
Because for advanced products it is equally binding. A design with wafers but no packaging slot does not ship. When the scarce step moved from fabrication to assembly, the commercial conversation moved with it.
Each page explains one technology in plain language, states what it depends on, and names companies by what they supply at that step. Company roles are described qualitatively and deliberately carry no market shares, revenue figures or rankings — those change faster than an explainer can, and a stale number is worse than none. Ticker links point at company pages on this site and are provided for reference only.
Nothing here is investment advice, a recommendation, or a forecast. A company named on a page about a technology is not thereby a good investment, and the chokepoints described are structural facts about supply chains rather than predictions about prices. Technology moves; where a page describes something as unresolved or in development, that was true when it was written.