Decide on where to eat in 10 minutes or less

I have been working on a project on lovable.

I built a small tool to solve a problem I kept watching happen — and it turns out I wasn’t alone.

Why This Exists

A few months ago, I started paying attention to how my friend groups decide where to eat. Here’s what it usually looks like:

  • Someone sends a message: “Where do you want to eat?”
  • The group chat goes quiet.
  • Then someone says “I don’t care.”
  • Then someone else says “Anything but sushi.”
  • Then a third person pulls out their Iphone notes app with restaurants that they still have not been to but want to.
  • Now you’ve got five people, restaurant lists on notes app, Instagram, Google Maps and zero consensus — and it’s already 7:15.

What should take five minutes stretches into 30. And somehow, you still end up at the same place you always go. I wanted to make sure this wasn’t just my experience. So I texted friends, brought it up in conversations, and watched it play out in real time. The same patterns kept coming up:

Ideas are scattered across Notes apps, Instagram posts, screenshots, and group chats Constraints like price and distance surface after people have already anchored to something No one wants to be the one to decide, so the loudest voice wins

One friend put it best: “We spend more time deciding than actually eating.”

What I Built

A shared session where everyone can submit restaurant options and vote. The organizer creates a session and shares a link. No accounts, no login. Participants open it, add their picks, and rank the options. The app surfaces a winner. I scoped it down deliberately. The hardest part of this problem isn’t finding a good restaurant — it’s getting a group of people to agree on one. That’s what I’m solving first. Reservations and menus stay offline for now.

PM Decisions

A few deliberate calls shaped what this became. Preference tiers over simple voting — Favorites, Top Choices, Okay With, Dislike — because the goal isn’t just picking a winner, it’s finding where the group actually overlaps. A single vote per person tells you what people want. Tiers tell you what people can live with. That’s a different and more useful signal when you’re trying to get five people to agree on dinner.

No login, no accounts. Friction kills group tools before they start. If someone has to create an account to join a session, half the group won’t bother. The session link is the access. No reservations, no menus — yet. It’s tempting to build the whole thing. But the hardest part of this problem is getting a group to agree, not finding a good restaurant. I scoped to the hard part first and left the rest for later.

What I’m Measuring

I’m watching three things right now: Drop-off point. Where do people abandon a session? If it’s at the voting step, the UX needs work. If it’s at the share link step, the onboarding isn’t landing. Session completion rate. A session that gets created but never voted on is a failed session. That number matters more to me than signups. Time to decision. The whole premise is 10 minutes or less. If real sessions are running 20–30 minutes, the product isn’t solving the problem yet. Soft launching with small friend groups — people who’ve lived this problem and will give honest feedback. If you’ve felt this pain or have a take on how your group handles it, I’d love to hear from you.

Built with Lovable. Problem validated the old-fashioned way, by watching people argue about dinner.