Product Build for a practice everyone agrees is right but nobody does
Builds systems that make hiring decisions better
The richest product territory is the gap between "known best practice" and "what people actually do." Structured interviews are universally accepted as the strongest predictor of job performance, and almost nobody runs them. Double-entry bookkeeping, security key logins, automated tests, pre-commit hooks: same pattern. Look for a practice that experts in your industry agree is correct and operators skip because it is inconvenient, then build the tool that makes the right thing the easy thing. The onboarding pitch writes itself.
Related advice
Flavius D. Ask specific questions that reveal the real role
Generic interview questions ("tell me about yourself," "what are your weaknesses") produce rehearsed answers and almost no signal. Two questions that consistently produce the opposite are "What does a typical day look like in this role?" and "What separates a great hire from an average one?" The first forces the hiring manager to describe the real job instead of the job description. The second forces them to articulate the dimension on which people actually succeed or fail. Use the answers to design the rest of the interview. If a hiring manager cannot answer either clearly, the role is not ready to be filled.
Flavius D. Price against the cost of the mistake, not your competitors
For tools that sit on top of high-stakes decisions (hiring, compliance, security, legal), the wrong benchmark is competitor pricing. The right benchmark is the cost of one bad outcome. Pace is built for hiring, where a single bad hire typically costs one to two times the annual salary once you factor in ramp, opportunity cost, severance, and team drag. A tool that prevents one bad hire a year is paying for itself many times over at almost any price. Teach your prospects to do that math in the first five minutes of the demo, and pricing objections mostly go away.
Mircea Build the simplest version first
Mircea built SingleFax in a weekend by refusing to add anything beyond the core action: upload, enter a number, pay, send. No user accounts, no dashboards, no analytics. If your v1 takes longer than a week, you're building too much. Strip it down until a complete stranger can use it in under 60 seconds.
Mircea Remove all friction: no signup, no subscription
Every form field you add, every account creation step, every subscription commitment is a point where customers leave. I removed all of them. No signup, no login, no monthly plan. Just pay and use. For occasional-use products, this is the difference between making money and making nothing.
Extracted from
Indie TM #7: News as a Traffic Locomotive