These are the assumptions most likely to break the thesis if left unchecked. Each should move to research/ with real findings before being treated as resolved.

1. Is this a “must-have” or a “should-have”? The forcing functions (regulatory audit, complaint escalation, insurance/franchise requirements) need to be concrete and named, not implied — this determines willingness to pay and urgency.
2. Who actually owns the buying decision at a 5-person garage — owner-operator, franchise head office (if part of a group like a dealership network), or the MTA on their behalf? This changes the entire GTM motion.
3. Willingness to pay at this scale is usually low and price-sensitive — worth stress-testing a per-shop or per-technician price point against what they currently spend (often near-zero) on this problem.
4. Agentic reliability in a compliance context is a double-edged sword — an agent that mis-tracks a certification or mishandles a complaint record has real regulatory/liability consequences. Decide early how much of the workflow is agent-automated vs. agent-assisted-with- human-sign-off.
5. Beachhead discipline — automotive is a good wedge because of Kelvin’s relationships, but resist scoping the MVP for “all regulated trades” too early.