Playbook · 2026

How to get your first 100 users – the channel reality index.

Most first-users advice is a list of channels with the failure rates removed. This is the same list with them left in: what each channel realistically returns, what it costs in time or money, and the specific way it goes wrong. Including the channels where paying for a community placement is the wrong move.

Reality check

Three numbers that reset expectations.

Published figures, not network estimates. Read them before you pick a channel.

3.1% vs 23.1%Visitor to signup

Product Hunt traffic converts at around 3.1%. Indie Hackers traffic converts at around 23.1%. Same product, same page, different room. The gap is intent, not copywriting.

Launch platform vs peer communityWhy a smaller room can out-deliver a bigger one
~90%Show HN posts that go nowhere

Roughly nine in ten Show HN submissions fail to gain traction. The ones that land are genuinely transformative for a week, which is exactly why founders keep planning around an outcome they can’t summon on demand.

Hacker News / Show HNA lottery ticket, priced as a strategy
4–6 monthsBefore the best converter pays out

Indie Hackers has the strongest conversion on this page and the longest runway to earn it: months of sustained, non-promotional participation before it returns meaningful traffic. Free is not the same as fast.

Indie HackersThe cost is paid in time, not budget
The index

The channel reality index.

Every row includes the failure mode. The last row is ours, and it fails too.

ChannelWhat it realistically returnsWhat it costsHonest failure mode
Product HuntA one-day traffic spike, a backlink, a badge1–2 weeks of prep, plus the favours you spend asking for supportConverts around 3.1%. The audience browses launches rather than shopping your category, and there is no second launch day.
Hacker News / Show HNOccasionally a flood. Usually nothing.An afternoon writing it, then a full day defending it in commentsRoughly 90% of Show HN posts never gain traction, and the 10% that do can’t be reproduced on demand.
Indie HackersThe highest per-visitor conversion here, around 23.1%4–6 months of genuine, non-promotional participationIt reaches founders, not customers. Excellent if you sell to builders, close to useless if you don’t.
RedditDurable visibility: the most-cited domain in AI answers at roughly 40% of citations in one 150,000-citation analysis, and a climb from 68th to 5th in Google organic visibilityMonths of account history before anyone takes you seriouslyZero control. Moderators remove first-time promoters, and one misjudged post closes a subreddit to you permanently.
Cold emailPredictable volume when the list is genuinely rightDomain warm-up, sequencing tools, and reply rates that assume scaleYou need a precise list before you have a product story, and deliverability punishes learning in public.
Paid adsVolume, once the algorithm has learned$3,000–5,000 before the data means anythingAt a hundred users the budget buys noise, not signal. It is the right tool at the wrong stage.
Paid community placementsThe admin a private room already trusts posts your product, tracked per roomCampaigns from $500, with a target cost per action and a hard capAdmins decline products that don’t fit, and rooms only exist for some audiences. If your buyers aren’t already gathered somewhere, this channel does nothing for you.
Order of operations

Do them in this order.

The index tells you what each channel returns. Sequencing is what stops you from spending money to learn something a conversation would have told you for free.

01

Name the hundred

Write down who they are specifically. If you can’t describe them in a sentence, no channel will find them for you.

02

Find where they gather

Public forums, private servers, group chats, or nowhere. The answer decides which rows of the index apply to you.

03

Earn the first twenty by hand

Manual, unscalable, direct. These are the users who tell you whether the pitch survives contact with a stranger.

04

Instrument the source

A tracked link per room, per post. Without it, every channel reports back as direct traffic and you learn nothing.

05

Pay to repeat what worked

Once one room converts, buy the same motion in ten more rooms instead of posting your way into each one.

Where paid placement fits
Our own losses

Three cases where Torchly is the wrong answer.

Published because a comparison that never loses isn't a comparison.

01

You sell to enterprise buyers

Your buyer is a VP at a company with a procurement process. There is no Discord server of them, no group chat you can be endorsed into. Targeted outbound, events and warm intros beat community placement here, and it isn't close. Don't brief us.

02

You have no working product yet

A placement spends an admin's credibility on something the room can't use. That trust doesn't refill, and the campaign under-delivers on results you're paying for anyway. Twenty direct conversations will teach you more this month than any paid channel will.

03

You sell to builders and you have months

Indie Hackers converts at roughly 23.1% of visitors and charges nothing but 4-6 months of genuine participation. If your runway can absorb that, take the free version. Paid placement is what you buy when the timeline can't wait or the room won't have you as a stranger.

If none of those describe you: browse the communities in the directory · the conversion numbers behind the channel: benchmarks · the same argument for AI products: distribution for AI startups.

FAQ

First-users questions, answered straight.

Weeks if you already know where those hundred people gather; months if you don't. The variable is rarely effort - it's whether the audience is already assembled somewhere. Indie Hackers, for example, typically needs 4-6 months of sustained participation before it returns anything, while a single fitting community placement can move dozens of signups in days. Pick the channel by where your buyer already sits, not by which one feels productive.

There isn't one, and any guide that names one is selling something. Product Hunt traffic converts at roughly 3.1%; Indie Hackers converts at roughly 23.1% but takes months of unpaid presence to earn. Roughly 90% of Show HN posts never gain traction at all. The right answer depends entirely on whether your users gather in public, in private rooms, or nowhere at all - and the third case means outbound, not distribution.

It works for what it is: a one-day spike, a backlink, and a badge. It converts around 3.1% of the traffic it sends, because the audience is browsing launches rather than shopping for your category. Treat it as a PR moment with a known ceiling, run it once, and don't build the plan around it. There is no second launch day.

For visibility, increasingly yes. Reddit is the single most-cited domain in AI-generated answers - roughly 40% of citations in one 150,000-citation analysis - and it climbed from 68th to 5th in Google organic visibility. That makes an authentic Reddit presence unusually durable. The catch is control: moderators remove first-time promoters on sight, and one misjudged post can close a subreddit to you permanently.

Three cases, plainly. If you sell to enterprise buyers, there is no room of them to reach and outbound beats us outright. If you have no working product yet, a placement spends the room's trust on something nobody can use. And if you sell to builders and have several months to spare, Indie Hackers converts at roughly 23.1% and costs nothing but time - take the free version. Paid placement is what you buy when the free version is too slow or the room won't have you as a stranger.

Campaigns start around $500. You set a target cost per confirmed action and a spending cap, so the worst case is known before you pay, and unspent budget is refunded. A weak campaign costs you volume rather than a surprise invoice. Admins review the exact copy and decline products that don't fit their room, which is why the fit question is worth answering honestly before you brief.

First 100

Your first hundred are already in a room.

Communities matched to the audience you namedCampaigns from $500, worst case known upfrontPer-room tracking on every placement
Submit a campaign brief