Nobody Signed Up After Launch: How to Tell If It's Distribution or the Idea

Guest post by the team behind PainBase.

You shipped. You posted it in a few places. The board says 31 upvotes, analytics says 84 visitors, and two of your three signups are people who already follow you on X.

So what now? Go harder on distribution, or was the idea wrong?

Nick's been making the same point in his weekly recaps for months and he's right: shipping got cheap, getting found didn't. Distribution is the answer most of the time. But most of the time isn't always, and the two wrong answers cost you differently. Grind distribution for something nobody wants and you lose a quarter. Kill something people did want because you posted it once on a Tuesday and you lose the idea, which is worse, because you never find out.

I've watched this call go wrong in both directions. Founders who grind distribution for six months on something with no complaint signal behind it, and founders who kill a decent idea because launch week was quiet. The second one is more common than people think, and you never find out you did it.

What follows is what I run now instead of guessing. Four checks, in order, because each one only means anything if the last one passed.

Check 1: Did anyone actually show up?

Find the number of visitors who came from somewhere the audience plausibly has your problem. Not total sessions, not the launch day spike.

Under about 100, stop. You don't have a result yet.

At 84 visitors, moving two people shifts your conversion rate by more than two points. You'd be making a build-or-kill call on a coin flip and you'd feel confident doing it, which is the dangerous part.

Most launches die here and never get any further. If that's you, the idea isn't dead, you just haven't run the test. Go get the traffic. The directories catalog↗ (opens in a new tab) is the boring way to do it, and boring is fine. Twenty or thirty free listings in a morning is roughly what 100 targeted visitors costs.

Check 2: Did they bounce or did they read?

Assuming you cleared 100, pull median time on page and scroll depth for the landing page.

Under 10 seconds with a high bounce means they read the headline and left. Whatever got them to click promised something the page didn't deliver. Copy problem, fixable this week.

Thirty to ninety seconds, scrolled past the fold, still no signup? That's the useful one. They understood the offer and passed on it. You're learning something real, though not necessarily about the idea. Could be the price. Could be that there's no working product behind the page and they clocked it.

Long time on page with no scrolling is a tab someone left open. Ignore it.

Here's why this check earns its place: the first pattern and the second give you the same signup count and mean opposite things. If all you look at is conversion rate you can't tell them apart, and the people who read "zero signups" as "nobody wants this" are usually staring at the first one.

Check 3: Does anyone search for it in your words?

Wait 30 days past indexing, open Search Console, look at impressions. Clicks can wait.

Any impressions at all is a pass. Even 40 impressions and zero clicks is decent news, because Google matched you to real queries and you're only losing the click to a weak title. Cheap fix.

Zero impressions after a full month is where people get it wrong. They read it as no demand. Usually it means nobody describes the problem the way you described it.

That distinction sounds pedantic and isn't. Builders name problems in system language. People with the problem name them in complaint language. You wrote "workflow automation for agencies." She typed "how do I stop forgetting to send invoices." Same problem, no shared words, and Google has no reason to connect them.

So before concluding anything, read the actual queries in Search Console, including the ones sitting at three impressions, because the phrasing is the data. Then run the problem through the First Users report↗ (opens in a new tab) and look at what communities and queries come back. If none of it resembles your page copy, there's your answer.

Rewrite in their words, wait another 30 days, and don't change three things at once.

Check 4: Is anyone complaining about this right now?

This is the only check that tests the idea rather than your execution of it, which is exactly why it goes last. The first three explain most failures and cost almost nothing to fix.

The question: in the last 90 days, have strangers complained about this specific problem in public, unprompted, without knowing you exist?

Five complaints, five different people, none of them asked by you. Upvotes don't count. Neither does a friend saying "oh that's a good idea," because you framed the problem for them before they answered, and everyone's generous when you're sitting right there.

By hand it takes about an hour:

• Reddit, searching the problem in their words, using phrasing from Check 3, sorted by past year
• Hacker News through Algolia search, same phrasing
• Comment sections on launches of adjacent products, where people evaluating a competitor say what's missing and say it bluntly

Save the permalink every time. Five links or you don't have five complaints. It's easy to convince yourself you found five when you actually found two and a maybe.

I got tired of running that search by hand, which is why I built PainBase, where founders find and validate startup ideas backed by market data. It watches public discussions on Reddit, Hacker News and Dev.to, scores how strong each complaint signal is, and links straight to the thread. The manual version above works fine and costs nothing. It just costs the hour.

Reading it

Failed Check 1 and you never ran the test. Get traffic, come back. This is the answer most of the time and it isn't the satisfying one.

Passed 1, failed 2 on fast bounces: positioning. Wrong people showing up, or the right people showing up with the wrong expectation.

Passed 1 and 2, failed 3: vocabulary. The demand might be sitting right there under words you didn't use.

Passed 1, 2 and 3 and failed 4, now we're talking about the idea. People find you, read you, decline, and nobody out there is complaining about the problem unprompted. That's the one where rewriting the headline again is procrastination.

Passed everything and still nothing? Then the problem's real and reachable and your offer is off. Price, scope, or the credibility gap of a landing page with nothing behind it. Cheapest next test costs money on their side: a preorder, a deposit, a paid pilot. Emails are free to hand over, and the distance between an email and a card is where a lot of validated ideas quietly go to die.

What this won't tell you

Whether anyone will pay. Everything above measures recognition and interest, and neither is revenue.

Whether the market's big enough. Five complaints prove the problem exists. They don't prove it exists ten thousand times at a price that supports a business.

Whether you can reach these people again. Reaching an audience once during launch week and reaching them every month are different skills, and the second one decides whether this becomes a business or stays a project.

So

After a flat launch you'll reach for whichever explanation matches your mood that day. I still do it. Then I make myself run the checks.

Usually you stop at Check 1, realize you never ran the test at all, and go get traffic. Less romantic than "my idea was ahead of its time." Considerably more useful.

Questions I get

How long before I run this?

At least 30 days past indexing, since Check 3 needs Search Console to have collected something. You can read 1 and 2 sooner, but running them alone tends to produce a confident wrong answer.

What counts as a targeted visitor?

Someone from a source where the audience plausibly has your problem. A thread about the exact problem, yes. A general "new products this week" feed, mostly no. If you can't separate the two in analytics, that's your actual first problem, and it's a UTM tagging fix.

Is 100 really enough?

Enough to stop reading noise. Not statistically clean, not trying to be. It's roughly the point where a 2% result and a 12% result stop being the same result.

I have complaints but no search volume. Bad sign?

Not usually, sometimes the opposite. Complaints run ahead of search volume, occasionally by a year. It means SEO is the wrong first channel and community is the right one. Doesn't mean the idea's dead.