Soft Launch: How to Test a Game Before You Spend Big

A soft launch is releasing your game quietly, in a small market, to find out whether it works before you spend real money telling the world about it.

It is the cheapest mistake you will ever make. Skipping it is the most expensive.

What a soft launch is for

It answers one question: do players come back?

Everything else is secondary. Not downloads, not reviews, not whether your friends liked it. Just whether a stranger who installed it yesterday opens it again today. If they do not, no marketing budget will save it. You will pay to fill a bucket with a hole in it.

How to run one

Pick a small market. Somewhere with players who resemble your real audience but where a failure is invisible and cheap. Canada, Ireland, New Zealand, and the Philippines are all common choices for different reasons.

Release quietly. No launch post, no press. You want honest behaviour from people who found it cold, not sympathy from people who know you.

Buy a small amount of traffic. You need enough installs to see a pattern rather than noise. A few hundred players is usually enough to see the shape of the truth.

Then leave it alone and watch.

The numbers that matter

Day 1 retention. Of the people who installed yesterday, how many opened it today? This is the headline number. If it is weak, stop and fix the game.

Day 7 retention. Whether there is a reason to keep playing after the novelty is gone. This is where thin games are exposed.

Session length and count. Are they playing once for a long time, or often for a short time? Neither is wrong, but it should match what you designed.

Where they quit. The most actionable number you have. If most players stop on level 4, level 4 is your problem, and it is fixable this week.

Retention tells you whether you have a game. Everything else tells you how big it could get. Do them in that order.

We wrote more on why this metric rules the rest in why retention beats downloads.

Reading the results honestly

This is where it gets hard, because by now you love your game.

If retention is poor, the honest options are to fix the core loop or stop. The dishonest option, which almost everyone reaches for, is to blame the test: wrong market, wrong players, bad luck, we just need more installs.

Sometimes that is true. Usually it is not.

What to do with bad numbers

Bad numbers are not failure. They are the cheapest information you will ever buy.

• Weak day 1? The first minute is not landing. Cut the tutorial, get to the fun faster.
• Fine day 1, weak day 7? No reason to return. You need meta progression, not more levels.
• Good retention, nobody installs? The game is fine. The store listing and icon are not. That is a far nicer problem.

Fix, test again, and only then spend.

Questions we get asked

Do small games need this? The smaller the budget, the more it matters. You have less money to waste on the wrong game.

How long should it run? Long enough to see day 7 from a full group of players. Two to four weeks is typical.

Can I soft launch an unfinished game? Yes, and you should. It needs the core loop and enough content to reach day 7. It does not need final art.

Test it properly

We build soft launch into how we work, and we tell you what the numbers say even when the answer is not what you hoped.

Share your idea.