The process can be real and still not be the business
Semiconductor founders often start with a process claim: a better material stack, a clever integration step, a device structure, a memory element, a sensing layer, or a manufacturing shortcut.
The claim may be technically real. It may still not explain the business.
The business begins when the process changes something a customer cares about: cost, power, yield, reliability, size, supply, qualification, security of supply, system performance, or time to market. If the process does not move one of those levers enough to change behavior, it is an invention looking for a wedge.
Process proof is not product proof
A process claim usually sits upstream of the evidence a buyer needs. The buyer rarely purchases the process. The buyer purchases a part, a system improvement, a qualified supply path, or a financial outcome.
That means the process story has to travel: from unit process to integrated flow, from flow to yield, from yield to cost, from cost to product, from product to adoption, and from adoption to margin or strategic value.
Each step can be plausible. The chain still has to hold.
What must be true
For a process claim to support a business, several assumptions must be explicit. The process must fit a manufacturing path. It must be measurable. It must be repeatable. It must not introduce a qualification burden that overwhelms the advantage. It must be tied to a product configuration with a buyer and a reason to switch.
Most importantly, the process must be replaceable in the model with evidence. Planning proxies are acceptable early. They are not a substitute for quotes, measured data, customer interest, or foundry/dev-fab results.
TIGRE lens
TIGRE pressure-tests the chain between process claim and commercial consequence. The question is not "is this clever?" The question is "does this create defensible confidence in a product, customer, manufacturing path, and capital plan?"