Most bad freelance projects were bad from the first call. The warning signs were there. You just didn't want to see them, because you needed the money.
Qualifying a client is not about gatekeeping or being picky. It's about finding out, in the cheapest possible way, whether this engagement will end with you getting paid and both sides being glad they worked together. Five minutes of honest questions saves months of regret.
Ask what their budget is — a real number, not a range. If they refuse to share one, they either don't have one or don't trust you with it. Neither is a good start. Ask when they last hired a freelancer, and how it went. Listen for how they talk about that person. That's how they'll talk about you.
Ask who approves the work. If it's "the team" or "we'll figure it out," you're about to run into the most expensive freelance failure mode: no single owner, so every decision requires three. Ask when they expect to pay. If "net 60" shows up unprompted, plan for net 90.
The clients who pay on time, respect your time, and scope cleanly — they don't flinch at any of these questions. The ones who do are the ones you needed to screen out.
It's not rudeness. It's urgency without clarity. Clients who need it "yesterday" but can't articulate what "it" is are the ones who will keep moving the goalpost and blaming you when the goalpost moves. Decline politely. There's always another client.
You don't have to say no right away. Price the risk in. A client who shows three red flags is not your $X client — they're your $2X client. Charge for the operational overhead their chaos will cause. Most will self-select out. The ones who don't were actually serious.