The single most overlooked dimension in idea evaluation is distribution: how the first 50, then 500, then 5,000 users actually find you. Founders obsess over the product and assume marketing is "the next thing." It's not. It's the first thing.
The "name 50 users" rule
Before you commit, write down where your first 50 users come from. Not 50,000. Just 50. If you can't name a community, list, group, or relationship that contains 50 reachable target users, you don't have a launch plan.
Channels that actually work for first apps
- Niche communities you're already in — the Slack, Discord, Facebook group, or subreddit where your users hang out. By far the highest-converting channel.
- Cold outreach to a specific list — manual, painful, very effective for B2B niches.
- Content with a clear search intent — pick 10 long-tail queries, write the best answer on the internet, link to your tool.
- Partnerships with adjacent tools — integrate with the thing your users already pay for.
Channels that almost never work for first apps
- Product Hunt as a primary launch strategy
- Generic "post on Twitter and hope"
- Paid ads before product-market fit
- SEO as a 0-to-1 strategy (it's a 1-to-10 strategy)
The outsider's secret weapon
If you're an outsider founder (see Why Non-Technical Founders Have a Massive Advantage), you almost certainly already have a distribution channel — your professional network, your industry community, your old colleagues. Most outsiders dramatically undervalue this. A 2,000-person LinkedIn network in a specific niche is worth more than 200,000 generic Twitter followers.
The "channel-first" reframe
Instead of "I have an idea, where do I find users?" try "I have a community, what would they pay for?" The second question almost always produces better businesses. Niche communities have shared problems and existing trust — both of which take years to build from scratch.
For a structured score on your distribution plan, the Distribution dimension in the ShiporDrop quiz is exactly this. Most first ideas score under 4/10 here. That's the work.