You launched and nobody signed up: 10 reasons and the fix

Your app is probably fine.

You built a shop on a quiet street, and then you stood inside it. Everything below is about the outside: the board above the door, the door itself, and the street. None of it asks you to change what you built.

The honest framing: no single one of these ten is fatal. All ten together is what zero signups looks like. You do not have to fix all of them either. Three usually moves the number.

The ten reasons, and the move for each

Six of the ten an AI can draft for you today, and the prompts for those are further down. The other four are yours: nobody can record your demo, sit next to your users, decide who this is for, or make the hundredth post for you.

If you only do three this week, do 2, 4 and 5. The headline, the place and the door.

Twenty launch angles you can steal

Reason 1 says launch twenty times. This is what makes that possible, because “post about it again” is not a task, and “post the before and after” is.

Fill in the brackets. One a week is a year of launches.

Story angles

1. “I built [thing] because [the specific afternoon that annoyed me].”
2. “It took me [time] and cost [amount]. Here is everything that went wrong.”
3. “[Number] people told me this was a bad idea. Here is what happened.”
4. “I have used it myself for [time]. Here is what I changed after week one.”
5. “The first version was [what it was]. It was wrong. Here is version two.”

Proof angles

6. Before and after: a screenshot of the old way, a screenshot of the new way.
7. “[Customer type] used it for a week. Here is their number.” With their permission.
8. A thirty-second screen recording of the single best moment in the product.
9. “Here is the ugliest part of it, and why I have not fixed it yet.”
10. Your own number, said plainly: “[N] people signed up this month. Here is what I learned.”

Teaching angles

11. “[Number] ways people do [task] wrong.” Your product is the last line, not the first.
12. “Here is how to do [the thing your product does] by hand, for free.” Then: “or this does it in one click.”
13. The checklist your product automates, given away as a checklist.
14. “I looked at [N] [competitors or examples]. Here is what they all get wrong.”
15. Answer one real question you were asked, in public, at length.

Direct angles

16. “If you [specific pain], this is for you. If you do not, ignore me.”
17. “[Product] is live. [One sentence on the pain]. [Link].” The plain one still works.
18. “I am giving [N] people free access this week in exchange for a ten-minute call.”
19. “What I would pay for this: [amount]. What I charge: [amount]. Here is why.”
20. “Still here, still building. Here is what is new since [month].”

Pick five, write them now while you are thinking about it, and put them in your calendar on five different days. The reason one launch feels like enough is that the second one has to be invented from nothing, on a day when you are tired. Invent them today.

And the thing that makes twenty launches bearable: almost nobody saw the first one. You are not repeating yourself. You are being seen for the first time, by a different set of people, over and over.

Where your users already complain

Reason 4 is the one most people get wrong, and it is the cheapest to fix. You posted where other builders are. Builders are lovely, and they are the four likes.

The rule: go where people describe your problem in their own words, before they know a product exists. Those places are searchable.

Replace [problem] with how a normal person would say it, not your product category. Nobody searches for “invoicing SaaS”. They type “chasing clients for money”.

What to do when you find a thread: not a pitch. Answer the question properly, in full, with no link. Do that five times in a community before you ever mention what you built. It is slow, and it is the only thing in this section that works.

Build a list of ten places, each with a name and a link. An AI can do the first pass (prompt 3 below), but open every one yourself, because a good share of any generated list will be dead or wrong. Expect to keep about half.

You will know it worked when you can quote a sentence a stranger wrote about your problem, in their words, and it does not sound like your marketing. Put that sentence on your landing page. That is reason 2 solved for free by doing reason 4.

Cut signup to one step

Every step loses people, usually a noticeable share at each one.

The order to aim for is: let them use it, show them something good, then ask for an email. Many products do this exactly backwards.

• ☐ Remove email verification before first use. Verify later, or never.
• ☐ Remove the password. A magic link or a single sign-in button is enough.
• ☐ Remove card details from signup, unless you are a paid-only product.
• ☐ Remove every “tell us about yourself” question. Ask in email 2.
• ☐ For each field that is left, ask what breaks if you delete it. Often nothing.

You will know it worked when you can go from your landing page to doing the main thing in under thirty seconds, on your phone, logged out, without typing a password. Time it with an actual timer.

Record the thirty-second demo

Not a polished product video. A screen recording of your own hands.

• Start after login. Nobody wants to watch you log in.
• Show the single best moment, the one where the thing becomes obviously useful.
• No intro, no logo, no “hi guys”. The first frame is already the product.
• Keep it under thirty seconds. If it runs longer, it is showing too much.
• Put it on the landing page, high up, looping, readable with the sound off.

You will know it worked when someone who has never seen your product can watch it without sound and tell you what it does.

Fix the empty first screen

A new user who sees nothing leaves, and they are right to. An empty screen is a shop with bare shelves.

Pick one of these, not all three:

• Sample data. Pre-fill the account with three realistic examples, clearly labeled as samples, removable in one click.
• One guided task. A single arrow to a single button. Nothing else on the screen competes with it.
• The first item, already created. Make their first project or document for them, named after something they told you at signup.

You will know it worked when the first screen shows a new user something to click, rather than something to create. The same instinct runs through 20 things to fix before a client touches your app, which is the demo-day version of this list.

The three onboarding emails

Reason 9. They signed up, got nothing, and forgot. This is the whole fix, and it takes one afternoon.

Three emails in the first week. Plain text, from your name rather than a designed template, because they should look like a person wrote them.

Email 1, within an hour of signup.

Subject: [First name], the fastest thing to try first

One paragraph. Not a welcome: one specific action they can take in two minutes, with a link straight to it. Then one line, “reply to this if it does not work, it comes to me”. That line is the most valuable sentence in the sequence, because real replies are your reason 8 research arriving in your inbox.

Email 2, day 3.

Subject: The thing most people miss

One trick that is not obvious, with a screenshot or a ten-second clip, and no ask. You are being useful to someone who has probably forgotten they signed up. The screenshot does the reminding.

Email 3, day 7.

Subject: Did this work for you?

Two sentences and one question. “You signed up last week. Did it do what you needed, or did you get stuck somewhere?” Nothing else. No features, no upgrade prompt, no link.

That third one matters most, because it turns silence into information. People who never came back will sometimes tell you exactly why, and one honest reply is worth more than a week of dashboards.

Rules for all three: plain text, from an address that accepts replies, under a hundred words, no ask in the first two, one question in the third.

You will know it worked when you get a reply. Not a click, a reply.

Six prompts to paste

Each one needs something pasted in with it. That italic line is the difference between a useful answer and a generic one.

1. The headline (reason 2)

Here is my landing page copy: [paste it].
My product helps [who] stop [the specific pain].
Rewrite the headline and first paragraph around that pain, not around
the features. Give me 5 versions. Each one must name the pain in the
first 6 words. No adjectives like powerful, seamless or modern.

Paste in your current copy, and one real sentence a real person wrote about the problem.

2. Twenty launch posts (reason 1)

Here is my product: [one paragraph]. Here is my audience: [one sentence].
Write 10 launch posts, each a completely different angle: story, before
and after, teaching, proof, direct. Same product, 10 genuinely different
reasons to care. Plain language, no hashtags, no emoji. Each under 80
words.

Paste in the angle list above if you want it to follow that structure.

3. Where they complain (reason 4)

My users are [specific person, for example bakery owners who take orders
on WhatsApp]. Their problem, in their words, is: [the pain].
List 10 specific online places where people like this already complain
about it: subreddits, groups, forums, channels, review sites. For each,
give the name, why they would be there, and one search string I can use
to find the relevant threads. Do not list general startup or builder
communities.

Then open all ten yourself and delete the dead ones.

4. One-step signup (reason 5)

Here is my current signup flow: [list every field and every screen].
Cut it to one step. Tell me exactly what to remove, what to delay until
later, and what genuinely has to stay. For anything you say has to stay,
tell me what breaks if I remove it.

5. The three emails (reason 9)

My product is [one paragraph]. The first useful thing a new user can do
is [action].
Write 3 onboarding emails: one within an hour of signup, one on day 3,
one on day 7. Plain text, under 100 words each, from a founder rather
than a company. Emails 1 and 2 ask for nothing. Email 3 asks one
question about whether it worked. Subject lines under 50 characters.

6. The empty first screen (reason 7)

Here is what a brand new user sees on their first screen: [describe it].
Design the empty state. Give me 3 pieces of realistic sample data for my
product, the one first task to guide them into, and the exact text for
the screen. The user should have something to click, not something to
create.

The four nobody can do for you

Six of these are chores. The other four are the actual job.

Name the one person (reason 3). Only you can decide who this is for, and everything else depends on that decision. “For everyone” is not modesty, it is a decision that has not been made.

Record the demo (reason 6). Your hands, your product, thirty seconds, today.

Sit next to five people (reason 8). In person, or on a call with their screen shared. Give them one task, then say nothing. Not a hint, not a “just click there”, nothing, even when it is unbearable. Write down where each person stops. When the same spot comes up three times, you have found something no dashboard would have told you. This is the highest-value hour here.

Make the hundredth post (reason 10). Spreading on its own is an outcome, not a strategy. It tends to start long after the post where most people stop.

A four-week plan, if you want one

Week 4 is not the end. It is the first week where the machine exists.

What this does not answer

It will not tell you whether people want the thing. That is a different and harder question, and no checklist settles it. A quiet launch and an unwanted product look identical from the outside, which is exactly why it is worth removing the quiet first.

If you run all ten properly for a month and still nobody signs up, that is no longer a marketing problem, and it is worth hearing. Ten people saying “I do not need this” is a real answer, and a better one than silence.

Building is the easy part. Almost nobody says that out loud, because we all quietly hoped the building was the whole job. It is not, and finding that out does not mean you failed at it.