Spent months building LazyEats AI. Spent 1 day realizing I have no idea how to get users.

For the first 100 users I’d make the ask much narrower than ‘try my app’: pick one meal-planning pain and go straight at that. E.g. ‘I’m tired and have chicken/rice/eggs, what can I cook in 10 minutes?’ is a much easier hook than a general AI meal app pitch.

I’m doing the same kind of hand-to-hand work with Kinetic Override, my Android utility app, and the useful signal has come from exact-problem threads rather than broad launch posts. Your best early feedback probably comes from people already complaining about dinner decision fatigue, not from people browsing app launches.

Also: the Play Store screenshots should answer the delete-after-5-min question before install — what input do I give, what useful result do I get, and how fast?

The spam flagging is the trap most of us hit. Quora and forums punish anything that reads like a drop. What worked better for me on a similar solo Android app was finding threads where people already complain about the exact problem you solve, and answering the real question first. Mention the app only when it's the genuine answer, and only after you've actually helped.

For a meal app I'd go where the frustration lives: people posting about decision fatigue, hating meal planning, eating the same thing every night. Reddit cooking and budget subs, replies to those exact posts. One good reply on a thread someone is reading right now beats ten cold posts.

And 10 installs after launch is normal, not failure. First 100 is hand to hand. You're not behind.

the fast signal is not how much you built, it is how quickly people can use it without explanation. if users need a tour, the loop is probably too long.

t

For first 100 users, I would stop treating this like an app launch and treat it like a repeated dinner-rescue service.

My 7 day test would be:

1. Pick one narrow user: busy students, gym people trying to hit protein, or parents who hate planning weekday dinners. Do not test all three at once.
2. Rewrite the hook around a moment, not the tech. Example: “send what is in your fridge, get 3 cheap dinner ideas in 30 seconds.”
3. Make 10 short videos using real constraint prompts: “$12 grocery run,” “high-protein meal from leftovers,” “dinner with eggs, rice, and frozen veg.”
4. In every post, ask for a comment like “drop 3 ingredients and I’ll reply with a meal.” Then manually reply fast. That gets you conversations before installs.
5. Only push the app after the person has already received one useful answer.

The thing that would make me delete it after 5 minutes: generic recipes I could get from ChatGPT. The thing that would make me keep it: it remembers my constraints, gives a realistic grocery list, and saves me from thinking at 6pm.

The app idea is useful, but I think “AI meal suggestion app” is too broad for getting the first 100 users.

Most people do not wake up wanting another meal app. They hit a specific dinner problem:

“I have random ingredients and no idea what to cook.”
“I’m tired and don’t want to think.”
“I need cheap meals this week.”
“I keep ordering takeout because planning dinner is annoying.”

For LazyEats, I’d probably pick one of those pains and make the whole first-user push around that, instead of promoting the app generally.

The easiest wedge might be: “Tell me what’s in your fridge and I’ll give you dinner ideas.”

That is much more comment-friendly on Reddit, Quora, Facebook groups, and short-form content than “download my AI meal app.”

If useful, I can put the tighter version in writing. The useful part would be the first 100-user plan: who to target first, what hook to use, where to post without getting flagged as spam, and what 7-day test I’d run.