Should you leave a stable job for a startup?

The variables that actually matter, the risks people underestimate, the questions worth asking before you accept, and how to test the decision before you're fully committed.

"Should I take the startup offer?" gets asked like it's one question. It's really four smaller ones stacked on top of each other, and most of the anxiety around this decision comes from trying to answer all four at once instead of one at a time.

Variables. Compensation is the obvious one, but it's rarely the deciding one. The variables that actually move the outcome are: runway (how many months the company can operate at its current burn rate), your own financial runway (how many months you could go without this income if it doesn't work out), how replaceable your current role is if you wanted it back, and how much of the startup's success depends on things you can't see from the outside — the founders' judgment, the market timing, whether the product has found real demand yet.

Hidden risks. The risk people prepare for is the company failing. The risk people don't prepare for is the company surviving in a form they didn't sign up for — a pivot that changes the role, a founder who manages differently under pressure than they present in interviews, or a funding round that dilutes equity before it's worth much. Ask what happens to your role if the company's current plan doesn't work, not just what happens if the company itself doesn't work.

Questions to ask before accepting. What is the runway, specifically, not just "we're well-funded"? What does success look like in 90 days, in the founder's own words, written down? Who did this role report to before, and why isn't that person doing it anymore? What's the equity actually worth in a realistic (not best-case) outcome, after dilution?

Reversible tests. Before resigning outright, there are lower-risk ways to test the decision: negotiate a short trial or consulting period before converting to full-time, talk to two people who left the company in the last year, or ask to sit in on a real working meeting before you sign anything. None of these fully de-risk the decision, but each one converts an assumption into an observation — which is the whole point of testing before committing.

None of this tells you what to do. It tells you what to actually look at, so the decision you make is based on the real variables instead of how confident the offer sounded in the room.