Where to find startup ideas: 8 sources, ranked

Good startup ideas aren't brainstormed in a room, they're overheard. The best ones already exist as complaints, workarounds, and unmet requests. Here are eight places to find them, ranked by how strong the demand signal tends to be.

1. Reddit and niche forums

The richest source. People describe problems in their own words, at length, with context, and upvotes tell you how many others feel it. Search a niche's subreddit for “is there a tool that,” “how do you all deal with,” and “I hate that.” Recurring threads are gold.

2. Reviews of existing products

One- and two-star reviews on the App Store, G2, and Amazon are a list of features people wish existed. “Love it but it doesn't do X” is a product spec. So is a popular tool with a loud chorus of the same complaint.

3. Workarounds and templates people share

When a community passes around a spreadsheet template, a Notion setup, or a Zapier recipe to solve something, that's demand you can productize. The manual solution proves the need and shows you the exact workflow.

4. Job boards and freelance requests

Repeated postings for the same manual task (“need someone to reconcile invoices weekly”) reveal work that's begging to be software. If companies pay humans to do it over and over, a tool that does it is a business.

5. Regulatory and platform shifts

New rules and platform changes create sudden, urgent, deadline-bound demand, compliance requirements, API deprecations, tax changes. Painful, time-boxed problems get bought fast.

6. Slack, Discord, and private communities

Where professionals actually talk shop. The “does anyone have a tool for” questions in a focused community are high-intent and specific, and less picked-over than public forums.

7. Your own work and inbox

The task you dread every week, the process your team hacks around, the thing you'd pay to never do again. Founder-market fit is real, and you already know one niche deeply: your own.

8. Curated idea databases

The fastest path, if the curation is honest. A good database does the reading for you and keeps only ideas backed by real demand, with the sources and comparables attached. The risk is slop, so favor ones that show their evidence.

The source matters less than the signal

Wherever you look, you're hunting the same thing: a specific person, describing a specific pain, that recurs, with money already moving to avoid it. Find that and you've found something worth building. (Our guide to demand signals is the checklist to run once you do.)