Testing demand does not require leaving your job, and it does not require announcing your intentions. It requires specific conversations with people who resemble the customer you have in mind.
Start by writing the assumption you want to test as a sentence that could turn out to be false. "I believe operations managers at mid-sized manufacturers consider a slow monthly close a problem worth spending on." That sentence is falsifiable, which is exactly what makes it useful.
An exploratory conversation is not a pitch. Its purpose is to hear how the person describes the problem, what they already tried, and what that cost them. The moment it turns into an early pitch, it stops working as a test.
Watch one signal more closely than the rest: has this person already spent time or money on this problem? Stated interest is cheap. Past behaviour is more honest.
Professional constraints are part of the design, not an obstacle to it. Review your current contractual obligations, avoid using information or relationships belonging to your employer, and keep the conversations inside your own professional standing rather than your job.
The test may show the assumption was wrong. That is a useful result, and it cost a handful of conversations instead of several months.
Nothing in this test tells you whether a career or financial transition is safe for you. It tells you whether demand is present or absent, and no more than that.