How to Get Your First 100 Waitlist Signups Without a Launch Platform

Founders treat the first hundred subscribers as a scaled-down version of the first thousand, and reach for the same instruments: launch platforms, paid acquisition, a posting schedule. All three underperform badly at this size, and the failure is misread as evidence that the idea is weak.

The first hundred are structurally different. They are not an audience you reach; they are a set of conversations you have.

Get them from three sources that do not scale — direct outreach, communities you have earned standing in, and one piece of writing only you could produce — and instrument everything so the hundredth signup tells you where the next thousand come from.

Twelve individual messages outperform any broadcast at this stage

Not a post to your followers. Twelve separate messages to twelve specific people you already know have the problem.

The ask is not "sign up." The ask is "tell me whether this promise is clear." That framing does three useful things at once: it is easy to say yes to, it produces feedback rather than politeness, and it converts anyway — people who understand the promise and have the problem sign up without being asked.

This is also the only cheap way to discover that your positioning is wrong. If eight of twelve reply with a variation of "so is it like X?", you have learned that your page reads as a category you did not intend, while changing it still costs an afternoon.

Expect roughly a third to sign up and roughly half to reply with something useful. Six signups from twelve messages is a better rate than any channel you will ever run again, which is worth remembering when the rate later drops to two percent and feels like failure.

Communities convert only after you have spent standing in them

For every product there are three or four places where people describe your problem in their own words: a subreddit, a Discord, a professional Slack, a recurring thread. Find them before you build anything — they are also the best source of the language your landing page should be written in.

Then participate for two weeks before mentioning what you are building. This is the step founders skip and it is the step that determines the outcome. A link from a known name who has answered forty questions lands as a recommendation. The identical link from an account whose entire history is one promotional post lands as spam, gets removed, and occasionally gets you banned from the one place your customers gather.

When you do post, post the problem and your finding, not the product. "We spent three months on why X breaks and here's what we found" earns the right to a link at the bottom. "Hey everyone, check out my new tool" does not.

Two communities done properly beat fifteen done shallowly. The constraint is your credibility, and credibility does not parallelise.

One good piece of writing outperforms a quarter of promotion, and keeps working

You have spent months inside a problem. That produces one or two observations that are genuinely non-obvious to everyone else — a benchmark you ran, an approach you tried and abandoned, a number that surprised you.

Write that up properly. Not a listicle, not a product post: the actual finding, with the actual data, including the parts that did not work. Publish it where your audience already reads.

This outperforms promotion on two dimensions. It converts better, because a reader who finishes a technical piece has already spent fifteen minutes trusting you. And it does not decay — a good writeup accumulates links and search traffic for years, while a promotional post is dead in a day.

Put the waitlist link at the bottom, once, without ceremony. The piece has to stand alone or the trade does not work.

Instrument from the first signup, because the distribution will be lopsided

Tag every link you place. Every message, every community post, every article, distinctly.

When you reach a hundred, the source distribution will not be even. It is routinely sixty to eighty percent from a single channel. That concentration is the most valuable output of the entire exercise — more valuable than the hundred subscribers.

It tells you where the next month goes, and it tells you which four things to stop doing. Founders who skip instrumentation reach a hundred and then spread the following month evenly across everything, diluting the one channel that worked.

Switch the referral loop on before you have anything to refer. Ten people sharing at twenty-five percent compounds; a thousand people sharing at zero does not. The loop is worth more the earlier it starts, and waiting for the list to be "big enough" forfeits exactly the period where compounding matters most.

Read the replies as a set rather than individually. Twelve messages produce six or seven substantive answers, and the signal is in what repeats across them — the same misreading, the same missing detail, the same comparison to a competitor you had not considered. A single dissenting reply is an opinion; the same objection from four people is a positioning defect.

Three things to refuse, however tempting

Do not buy a list. The addresses did not ask for you, deliverability damage is permanent, and in most of Europe it is unlawful.

Do not run ads before you know your page's own conversion rate — the most common failure we see is spending on traffic to a page that has not yet been read by a stranger. You will be paying to learn something a free afternoon of direct messages would have taught you, and you will misattribute a copy problem to a channel problem.

Do not wait for the product. The page validates the promise; the promise is what you are testing. Every week the page does not exist is a week of compounding you do not get back.

Put the page up before you send the first message. Pick a template, write the outcome in six words, publish it on your own domain. Referral links are generated for every subscriber automatically, so the loop is already running by the time your twelfth reply comes back.

A hundred is not a milestone. It is a diagnostic — and it only works if you were measuring.