What to do after a quiet product launch

Do this before you start explaining the result to yourself. For each place you launched, note:

• how many people visited your site from that listing (your analytics referrer report)
• how many signed up or installed
• how many did the main thing your product is for, at least once
• how many came back a second day
• every comment, copied somewhere you can sort it

If you launched on Nukkadly, the analytics page in your maker dashboard counts website clicks (one per visitor per product per 24 hours, with your own clicks left out), votes and feedback for each product.

Small numbers are fine. Three signups out of forty visitors tells you more than "it flopped".

Every launch is a short funnel: people see the listing, click, sign up, use the product, and come back. A quiet launch is quiet at one of those steps. Find the first step where the number falls off a cliff and work on that one only.

Few people saw it

Signs: low listing views or website clicks, and almost no referral traffic.

This is a distribution problem, not a product verdict. The people who would want the product never saw it, so the launch tested nothing. Before you rebuild anything, list where your users already spend time. For a developer tool that is usually the place they install from, not a launch board. The guides on where to launch a developer tool and product launch websites cover the options.

People saw it but didn't click

Signs: a decent number of views, few website clicks.

The name, logo and tagline didn't tell the right person the product was for them. This is the cheapest problem to fix. Rewrite the tagline so it names the object, the action and the person, using the patterns in how to write a launch tagline.

They clicked but didn't sign up

Signs: clicks arrive, signups don't.

Look at your landing page as a stranger. Can they tell what it does in one screen? Can they try it without an account? The Show HN guidelines ask makers to let people try things "ideally without barriers such as signups or emails" because that is where launch visitors give up. A demo mode, sample data or a short video of the product working can help.

They signed up but never used it

Signs: signups, then empty accounts.

This is onboarding. Something between signup and the first useful result is too long, confusing or broken. Watch one person set it up on a call without helping, and note every pause. The post on getting your first users after a launch covers setting people up by hand.

They used it once and didn't come back

Signs: real first use, no second visit.

This is the hardest one, because it points at the product's value, not its packaging. Either the problem isn't painful enough to return for, or the product doesn't solve it well yet. Talk to these people before you change anything.

Two comments can still be useful. Sort every comment, email and reply into four piles. These are the same tags Nukkadly commenters use on launches:

• Confusing. Someone didn't understand what it is or how to use it. Fix the words or the first screen.
• Missing. Someone wanted a feature or integration you don't have. Write it down, but don't build it yet. One request is an anecdote.
• Pricing. Someone questioned the price or the plan. Check that the price is visible and that the free path is clear.
• Useful. Someone said what worked. Keep it, and use their words in your copy.

Ignore comments that are only congratulations. They feel good and tell you nothing.

If you have fewer than five real responses, get more before you decide anything. Message each person who signed up, one at a time, and ask what they expected and where they stopped. The guide on getting feedback on a SaaS has a simple way to run that with five people, and the beta feedback questions list gives you questions to send.

Once you know the step and have a few real conversations, pick one of three moves. This is a judgment call, so write down why you chose it.

• Fix and launch again when people who saw it wanted it but something got in the way: an unclear tagline, a signup wall, a broken setup step. These are fixable in days or weeks.
• Change who it's for when the people who showed up were the wrong people, or the few who stayed share something you didn't plan for, such as one job title or one use case. Rewrite the page and the listing for them.
• Stop, or park it when the people you built it for saw it, understood it, tried it and still didn't come back, and conversations confirm the problem isn't painful enough. That is a real answer, and you got it cheaply.

Change one thing at a time where you can. If you rewrite the tagline, the onboarding and the price in the same week, a better second launch won't tell you which change worked.

• Don't buy votes or ask friends to upvote. Product Hunt says asking for upvotes breaks its rules. Hacker News says it penalizes or bans submissions, accounts and sites that solicit votes. Nukkadly's guidelines ban vote buying and coordinated fake accounts, and votes from accounts younger than seven days count at 0.25x, as the ranking page shows.
• Don't delete and repost. Hacker News asks people not to delete and repost the same story.
• Don't dodge a duplicate check. Product Hunt removes launches that change the URL to get around its two-month window.
• Don't relaunch with the same product. Every platform below expects something new to try.

Each site has its own rule. These are current as of 9 November 2026. Check them again before you post, because they change.

On Nukkadly, duplicate listings aren't allowed. If you rebuilt the product from scratch, the guidelines say to contact the team about an exception. The FAQ covers the major update option.

A relaunch has one thing a first launch doesn't: people who already looked once.

• Lead with what changed. Product Hunt's own relaunch guide says your tagline, gallery and maker comment should focus on the update, not repeat the first launch. The guide on writing a maker comment covers that note.
• Say what the feedback taught you. For a made-up example: "The first version needed a Stripe key before you saw anything. Now you can try it with sample data." A line like that closes the loop for the people who commented.
• Link the first attempt. On Hacker News, the moderator's tips ask makers who repost to link the previous Show HN and explain what is different.
• Tell the people who signed up last time. Write to them yourself and explain what changed. Don't ask them to vote.

A quiet first launch usually costs a week, and you come out of it knowing which step to fix. When the fix is in, post the update on your listing or submit your product to Nukkadly, and run through the SaaS launch checklist again before the week starts.