Guide 3 of 3
How to launch a browser game
The tactical version: choosing a day, writing a tagline that earns a click, screenshots that convert, what to do in the first hour, and the mistakes that quietly cost you the front page.
On this page7 sections
A launch is one day of attention that you cannot get back. The game is the hard part and you already did it; what follows is the hour of work that decides whether two hundred people play it or eleven do.
None of this is marketing in the uncomfortable sense. It is mostly being specific about what your game is, which is a thing you already know and have got too close to describe.
Pick a day
Everything that launches on the same UTC day competes in the same bucket, so choosing a day is choosing your competition. Two questions settle it.
How busy is that day usually? The launch archive shows every past day and how many games were on it. Midweek is the biggest audience and the biggest field; weekends are quieter on both counts. A weekend launch with a good game frequently finishes higher than the same game would have on a Wednesday, and the top spot on a quiet Sunday still sits on the front page all day.
Can you actually be there? This matters more than the calendar. The first hour is worth several later hours because the ranking divides by age, so a Tuesday you spend in meetings is worse than a Sunday you spend at your desk. Pick a day where you can sit with the comment thread for two hours during your own evening.
The tagline
Your tagline is doing more work than anything else on the card. It sits under the title on the front page, in the daily email, in search results and in every share preview. It gets about one second.
The rule that fixes most of them: describe what the player does, not how the game made you feel. Under ten words. One concrete noun and one concrete verb beats every adjective you were considering.
The mood board
Before
An atmospheric journey through a forgotten world.
After
A hand-drawn walking sim where the map redraws itself behind you.
The first one could be four hundred games. The second tells you the medium, the length and the one mechanic. “Redraws itself behind you” is the thing somebody repeats when they share it.
The feature list
Before
Roguelike deckbuilder with 60+ cards, 4 classes, procedural maps and permadeath.
After
A deckbuilder where discarding a card teaches it to the boss.
Counting content answers a question nobody asked. One rule that sounds like a problem answers the only one that matters: what happens when I play?
The maker's feelings
Before
My first game! Made in 2 weeks, would love any feedback 🙏
After
Park a car using only the handbrake and the horn.
Apologising in advance costs you the click you needed to get the feedback. Put the story in the description and the comments, where people who already like the game will read it.
A test that works: read your tagline to somebody who has never seen the game and ask them what they think they would be doing. If they describe a genre, rewrite it. If they describe an action, you are done.
Screenshots that convert
The cover image is the second-biggest decision on the card and the one most often wasted on a title screen. Nobody has ever played a game because of a logo.
Shot one: the game, mid-action
The one that mattersSomething happening. A frame from real play, not a posed screenshot with the HUD hidden. If your game has one moment people will talk about, this is that moment. Crop tight. Cards are small and a full 1920px screenshot reads as a grey smear at thumbnail size.
Shot two: proof it has depth
The one that convertsA later level, a full inventory, a busy board, a scoreboard with real numbers on it. Shot one earns the click; shot two is what stops someone bouncing when they think it looks thin.
Shot three, if you have it: the twist
OptionalThe mechanic your tagline promised, caught in the act. Skip it rather than pad. Three good screenshots beat six, and a weak third shot drags the average down.
Two practical notes. Capture at 2× and let the site downscale, because a soft screenshot looks like a soft game. And check what your image looks like at card size before you submit. The preview panel in the submit flow shows exactly that, which is what it is for.
The first hour
This is the part people skip and it is the part that works. The hot score divides by age to the power of 1.5, which means votes in hour one are worth many times what the same votes are worth in hour six. It also means a comment thread that is alive early pulls people in while the game is still climbing.
Post the first comment yourself
Minute zeroThree or four sentences: what it is, one thing you are unsure about, and one question you actually want answered. “The difficulty curve in the middle third is the bit I could not settle. Tell me where you got stuck” gets ten replies. “Hope you enjoy!” gets none.
Answer every single reply
First two hoursEvery one, including the negative ones, and especially the detailed negative ones. A player who wrote four paragraphs about your jump physics is the best thing that will happen to you today. Threads with a responsive maker keep growing; threads with an absent one stop within twenty minutes.
Share the link once, wherever your people are
EarlyYour followers, your Discord, the subreddit for your genre. Link the listing, not just the game. The votes and the conversation both live here, and the game link is one click away from the listing anyway.
Fix anything fatal immediately
As it comes inIf three people report the same crash, ship the fix during the launch and say so in the thread. A maker patching live on launch day reads as competence, not as a game that was broken.
What not to do
- Do not launch and leave. The single most common reason a good game finishes tenth. If you cannot be online, move the day.
- Do not lead with the making-of. How long it took, what it was built with and how many all-nighters you did are all interesting after somebody has played it. Put them in the description; the tagline is not the place.
- Do not apologise in the tagline. “My first game, please be nice” lowers the click rate and does not lower the criticism.
- Do not argue with negative feedback. Ask a follow-up question instead. “What were you trying to do when it felt unfair?” turns a bad comment into a bug report, and everyone reading watches you do it.
- Do not buy or trade votes. It is detectable, the votes get reversed, and it pushes an honest launch off the page. The rules spell out what happens next.
- Do not tag your way onto busy hubs. Listing a single-player game under multiplayer gets you clicks from people who immediately leave, which is worse for your ranking than not being there at all.
- Do not relaunch the same game. Once, for a genuine rebuild. Otherwise edit the listing. It stays live forever and edits are instant.
After the day
The listing does not expire. It keeps its votes, it stays in the archive for the day it launched, it remains eligible for the all-time leaderboard, and it can be picked up by a collection months later. So the day after is worth twenty minutes:
- Reply to anything that came in overnight. People check back more than you would think.
- Ship the two bugs everybody found. Then say so in the thread. That comment often gets more votes than the launch comment did.
- Update the description with anything the thread taught you about how to explain the game. Launch day is free copywriting research.
- Put a badge on your own site linking back to the listing. It keeps the thread findable for the people who arrive later.
Then start the next one. The makers with the best results on this site are almost never the ones with the best single launch. They are the ones on their fourth game, who got a bit better at all of this each time.