nowfound

Blog · September 5, 2026 · 3 min · by Bartosz Wlodarczyk

Seven in ten product launches never pass 4 votes. Here is what the other three did differently.

We indexed 206,630 product launches from one community launch board, 2009 to today, and looked at what separates the launches that found an audience from the ones that sank without a trace.

Every maker who has launched something knows the feeling: you post, you refresh, and nothing happens. It turns out that is not a feeling. It is the median outcome.

We keep an index of every product launch since 2009. For this piece we took the 206,630 launches posted to one of the big community launch boards, dropped the last two days so scores had time to settle, and counted.

The distribution

Outcome on launch day Share of launches
Under 5 votes 70.6%
10 votes or more 16.5%
50 votes or more 6.4%
200 votes or more 1.5%

Seven in ten launches never pass 4 votes. Only one in six reaches double digits. The front page, the thing everyone launches for, is a 6% event.

This is not because 70% of products are bad. Launch boards rank by votes per hour, so a launch that catches the wrong hour disappears before anyone has read it. The product never got judged; the timing did. Our Missed on launch day list exists for exactly these.

The hour matters more than the day

We looked at every launch from the last three years and asked how often each posting hour reached 10 votes.

Posting hour (UTC) Reached 10 votes
17:00 13.5%
16:00 13.1%
20:00 12.8%
09:00 8.6%
07:00 7.9%

The gap between the best hour and the worst is 70% in relative terms. 17:00 UTC is early afternoon on the US East Coast, morning on the West Coast, and evening in Europe: the one window where all three are awake.

The weekday barely moves the needle. Sunday is the best day at 11.9%, Friday the worst at 11.0%. Nobody should reschedule a launch for a weekday; everybody should reschedule it for the afternoon.

Write the description

Launches that came with a real description, at least a couple of paragraphs, reached 10 votes 13.2% of the time. Launches that were a bare link reached it 9.1% of the time. Writing 200 words raises your odds by almost half.

That matches what we see in search behaviour on this site: people describe a job, not a name. A launch that explains the job gets found long after launch day. A bare link never does.

Open source has an edge

Launches pointing at a public code repository reached 10 votes 13.8% of the time, against 10.3% for everything else. Part of that is audience, since community launch boards skew toward developers. Part of it is trust: a repository is a product you can inspect before you click.

More launches, same front page

Year Launches
2015 10,315
2019 10,719
2023 14,079
2024 17,625
2025 28,278
2026, to September 34,470

Launch volume has tripled since 2019 and is still accelerating. The front page has not grown. Every year, the same few hundred slots are shared by more launches, which is why "launch day" is becoming a worse bet and "being findable afterwards" a better one.

What to do with this

  1. Post at 16:00 to 20:00 UTC. It is the single largest lever you control.
  2. Write the description. Two paragraphs about the job it does and who it is for.
  3. Plan for the 70% case. Most launches sink on the day. Make sure yours is findable afterwards: a page that explains the job, in a place where people search by job.

That last one is what this site is for. Add your product, free, and it sits in the index next to every launch since 2009, where "tool that does X" queries find it long after launch day is over. The full numbers, updated hourly, are on the Numbers page.

Built something?

Put it where people search for it. Listing is free and takes two minutes.

Add product