You shipped fast. Lovable, Cursor, Bolt, Replit, v0, building has never been easier, and that's both the gift and the trap. The faster you build, the easier it is to skip the part where you find out whether what you built actually works.
These are the seven questions worth answering before you spend money on ads, ship a launch tweet, or sink another week into features. None of them are about your stack, your design, or your copy. All of them are about whether what you've built solves a real problem for a real person.
If you can't answer one of them with conviction, that's not a problem, it's a flag. The good news is each one is answerable in days, not weeks.
1. Who exactly is this for?
Not "small business owners." Not "founders." Not "marketers." Those are categories, not people.
Better way to frame it
"This is for a [specific role] at a [specific stage of company] who's currently struggling with [specific situation]."
Common mistake: Most builders define their audience in broad demographic terms (small business owners, B2B teams, e-commerce sellers) and end up building for everyone, which means building for no one. A sharper audience makes every other decision easier, from the landing page copy to who you should be talking to right now.
2. Have 5 real strangers from that group reacted to it?
Not your friends. Not your LinkedIn followers. Not the people in your Slack groups. Strangers, people who don't know you and have no reason to be polite.
Better way to frame it
Recruit 5 people who match the specific audience from question 1, give them a clear task ("here's what this is supposed to help with, try it"), and watch what actually happens.
Common mistake: Posting "what do you think?" on Twitter and counting the replies. People who already follow you have already decided they like you. Their feedback is encouragement, not signal.
If you don't have 5 real strangers from your target audience to ask, that's the part we help with, we recruit verified people from your actual target market, not your network.
3. Can a new user understand what it does in under 10 seconds?
The clarity test. Show your landing page or first screen to someone for 10 seconds, then hide it. Ask: "what did you think this product does, and who is it for?"
Better way to frame it
"Looking at this for the first time, what would you expect this product to do?" Asked before they've read everything.
Common mistake: Asking "Is this clear?" People will always say yes, because saying no feels rude. The 10-second test makes clarity measurable. Most vibe-coded apps fail here, not because the product is bad, but because the headline tries to be clever instead of clear.
4. What's the one thing it does better than what they use today?
Every product competes with something, even if that something is "nothing, they just live with the problem." If you can't finish the sentence "we're better than [current solution] because [specific reason]," your customer can't either.
Better way to frame it
Ask target users "how do you handle this problem today?" Get the honest answer (a spreadsheet, three browser tabs, a workaround, doing nothing). Then ask "what's the most annoying part of doing it that way?" That answer is your real value prop.
Common mistake: Listing features. Features are how. The question is why, why would someone switch from what they're already doing? Your "better than" answer has to be specific enough that a stranger can repeat it back to you.
5. Have you watched someone actually use it, in silence?
Watching is the part most builders skip. They demo the product, narrate what it does, or send a link and read the message back. None of that is watching. You learn more in 90 seconds of silent observation than in a 30-minute interview where you fill the gaps.
Better way to frame it
Share your screen with someone in your target audience. Give them a task. Say: "Talk me through what you're doing as you go. There are no right or wrong answers, I just want to understand how you experience this." Then stay quiet, even when they get stuck.
Common mistake: Jumping in to explain when they hesitate. Their hesitation is the data. The places they pause, hover, or backtrack are the places your product is unclear. If you rescue them, you lose the signal.
6. What happens at the moment someone has to pay?
The pricing page, the checkout, the upgrade prompt, the moment friction becomes real. Most vibe-coded apps are designed and tested up to the paywall and not past it. That's a problem, because the actual decision to pay happens at that exact moment.
Better way to frame it
Watch someone try to convert. Ask them "at what point did this feel too expensive, and at what point would it feel suspiciously cheap?" That's the Van Westendorp price sensitivity question, it works because it doesn't ask the hypothetical "would you pay" question, which always gets a meaningless yes.
Common mistake: Asking "would you pay for this?" during a casual conversation. Almost everyone says yes. Almost no one actually pays. The price question only gives real signal when there's a real purchase decision in front of them.
7. Who would tell their friend about this, and why?
This is the word-of-mouth question, and it's the one most builders confuse with "virality features." Virality is mechanics. Word of mouth is whether your product is good enough that a real person, unprompted, would describe it to someone they care about.
Better way to frame it
Ask target users "if this existed and worked well, who else you know would want it, and what would you tell them about it?" The "what would you tell them" part is the gold, it forces them to describe your product in their own words. If their description matches your pitch, you've got message-market fit. If it doesn't, your pitch is wrong.
Common mistake: Asking "would you recommend this?", another hypothetical that always gets a polite yes. The "who specifically, and what would you say?" version forces honesty.
What to do with the answers
If you can answer all seven with conviction, you're probably ready to launch.
If two or three feel shaky, that's normal, those are the gaps to close before spending on growth. Recruit 5-8 strangers from your target audience and answer them properly. The cost of getting honest answers from 5 real users is less than the cost of one week of paid ads to the wrong audience.
If you don't know how to find those 5 users, or you've tried and your network keeps giving you flattery instead of signal, that's the part we help with at Kosmo.
Get real users to answer these questions, not your network
We recruit verified professionals from your target audience, real people who match your brief, not anonymous panellists or polite friends. Pre-launch is the cheapest moment to find out what's not working.
You can also read our companion piece on the seven mistakes founders make in customer interviews, the things to avoid once you're in the conversation.