Ask what a well-qualified outside replacement would not know on their first day, and ask which internal document or system holds that difference. If nothing can be named, the claim is weak.
One question filters most cases
The classification reaches knowledge of the petitioning organisation's own product, service, research, equipment, techniques, management or procedures, or an advanced level of knowledge of its processes. It does not reach general expertise, however rare that expertise is in the local labour market. The replacement-hire question makes the distinction concrete because it forces the answer to point at something.
Where the response identifies proprietary control logic, an internal specification, or a method developed in-house and documented nowhere else, the case has substance. Where the response is that the person is exceptionally capable and hard to replace locally, the case is describing scarcity rather than specialized knowledge. Asking this before any drafting begins saves the cost of building around an answer that will not hold.
Hypothetical example: a medical-device calibration platform company wants to send the engineer who designed its internal fault-detection routines to a related U.S. team. The first review should produce a definition of the knowledge, the records showing how the engineer gained it, and a comparison with skills obtainable from ordinary training.
Version histories and restricted technical manuals settle the organisation-specific point; an impressive résumé does not. Confirm the corporate relationship and foreign employment year before commissioning technical statements. The common failure is describing the engineer as unusually talented without identifying a proprietary process that the new team genuinely needs.
If an outside hire could learn the same task from public manuals, the explanation should confront that fact rather than rely on scarcity.