Every game is its own startup

When I ask how many copies a game has to sell to cover its cost, most people freeze. It is not really a plain maths question, it is the question of whether that project should exist at all, and the answer is usually uncomfortable.

The reason the question goes unanswered is this, a game is taken for a project. But every game is its own startup. It has its own market, its own cost, its own break-even and its own chance of dying. The success of your previous game does not carry over to the next one either, it only leaves you a bit of an audience, a bit of experience and maybe a bit of money.

I wrote before that making games is a business, but that if you put the business part up front you end up a failed entrepreneur, and I stand behind that. Thinking like a startup is not building a pitch deck and chasing investors and publishers or sitting in an incubator, it is writing down each game’s cost and revenue separately and looking at whether that game stands on its own. Not covering one game’s hole with another game’s money, and if you are going to cover it, doing it knowingly. In Turkey this got understood the other way around, unfortunately. Keeping the books interests nobody, but giving presentations became an industry of its own, and schools glorify it on top.

People usually think the business side starts after the game is finished. But almost every decision you make before the first line of code is a business decision. Which genre you work in, what the scope will be, which price band you play for, which platform you launch on, how many people and how many months you work with. All of it stands there dressed as a creative decision, and all of it is paid for in money and time.

That is also why “I will learn it later” does not work. By the day you learn it those decisions are long made, and the only way to take them back is to throw the project away. Working out two weeks before launch that you aimed a game you spent a year on at the wrong audience is not a gap in knowledge, it is a direct loss.

Even Valve looks at each game as a separate business. Steam takes 30% of revenue, that rate drops to 25% at 10M dollars and 20% at 50M dollars, but the threshold counts per game, not per catalogue. So a publisher with five games making 3M dollars each never gets down to 25%, and past success does not buy you a discount on your next game. The practical result for you is one thing, that rate is 30% and it is staying there.

You also need to know the cuts between the price you list and the money that reaches your pocket. Because of regional pricing and discounts the realised average price usually lands around 60-70% of the sticker, so you sell a 10 dollar game at 6-7 dollars on average. Steam’s 30% sits on top of that. There are refunds, any game played under 2 hours and bought within 14 days can be returned, so your first two hours are as much a money question as a design question. There is also the 100 dollar Steam Direct fee you pay to open the page, which is credited back once you reach 1k dollars of revenue. Once all of it stacks up, the money that reaches your pocket from a 10 dollar sticker falls below half.

In a startup the company moves whether you work or not. The team gets paid, the server runs, the rent accrues. A game does not move a single inch unless you sit down at the desk. If you stop for two months the project stops for two months, nobody carries on in your place. There is only one thing that accumulates while you are away, your store page and wishlists. The game itself stops without you, but the page works without you. That is the real reason to open the page early, it is the only thing running on the days you do not work.

This has two consequences. First, the project’s calendar is your calendar, so if you write your estimate on ideal conditions instead of illness, work, burnout and life, that calendar is wrong from the start. Second, the main cost item is not money, it is time, so if your hour has no value inside the calculation you have not really calculated anything, because the salary you did not take is not free, it is just invisible. As a reference, work out what you would earn a month at a game company, the simplest version of it, and divide that by 180. That gives you your hourly rate for the project.

Once you know all these items, the break-even calculation comes out on its own. Write down your price, take out Steam’s share, account for regional distribution and refunds, then divide your cost by the net figure you are left with. What you are holding is how many copies you need to sell. That single number tells you at the same time whether your genre is realistic, whether your scope holds and whether that feature is worth building.

Do not leave that copy count hanging either, put this next to it. In the games Zukowski scanned, the median weekly wishlist count came out as 23, and do not forget it is a median, so half of the games scanned are below even that. At 23 wishlists a week, a year of development leaves you with around 1,200 wishlists. You can also roughly project your lifetime sales from your wishlist count at launch. Say you managed to sell half of that 1,200 in a year, 600 copies. If the game is 10 dollars, gross is 6,000 dollars. Going from gross to net you usually keep between 52% and 56% of that, take the pessimistic end at 52%, which is 3,120 dollars. That figure covers one year of development and one year of sales, so two years. Divide it by 24 and it comes to 130 dollars a month. The net minimum wage in Turkey today is around 590 dollars, so two years of your life come to less than a quarter of the minimum wage. And that is before taxes, Bağ-Kur self employment contributions and the rest.

What changes this picture is not marketing, it is the business decisions you made at the very beginning.

The part of the comparison people skip most is risk. Most startups die, and that is not an accident, it is the definition of the job. It is the same with games, what you are making will most likely not work. So the question to ask is not whether this game will work, it is what happens if it does not. If your answer is that nothing is left, you should not start that game at all, because when you have no second move you have to get everything right on the first one, and nobody gets everything right on the first one.

Build your budget around how many attempts you can make, not around one game working. Whoever puts their whole savings into their first game is not making a game, they are buying a lottery ticket.

And here comes the classic answer, “I am not doing it for money, I am doing it because I love it”. You cannot run a hobby you pay for out of your own pocket and a job you plan to live on with the same calendar and the same scope. The bad part is not failing to make money, it is starting without knowing which of the two you are doing. If you are not doing it for money then it really has to be a hobby, because the moment you have any expectation at all you need to earn enough to put bread on the table. Games do not get made on an empty stomach.

I owe being able to look at games purely as a business to mobile. Every game there was a separate bet and most of them died before they even saw a soft launch. That filter felt brutal to me, but it taught me this, the decision gets made anyway. If you do not make it, players make it for you.

So what do you do? There are five things to write down, the first three are done in an afternoon, the last two take some research.

One, write down your break-even copy count before you start the project. Do the calculation above once with your own price, your own timeline and your own hourly rate, that is all you have to do.

Two, pick 10 games from your own genre released in the last two years and look at their review counts. There are rough methods for turning reviews into sales and all of them have very wide bands, so do not look for an exact number, look only at the order of magnitude. Is the ceiling of your genre 3 digits or 5 digits, the answer is there.

Three, decide your price before your scope. Most people make the game and then go looking for a price, when the right order is to pick the price and work out how much content you can fit into it.

Four, once you have decided which country you will set the company up in, learn the costs and taxes that will come up during the project. If that country is Turkey, write down expense items like Bağ-Kur, accounting and a virtual office. Look at the grants if there are any. Never build your projection on grants, because grants are here today and gone tomorrow. Look at them as nice when they show up.

Five, once you have laid out all the money items, calculate the opportunity cost. Look at what your money would bring in if you simply put it in an interest account, and only make the game if what you expect from it is above that. Interest is risk free, a game is not, so beating interest on its own is not enough, the gap has to pay for the risk you are taking as well. Otherwise it makes more sense to put your money in an interest account and keep the time you would spend making a game for yourself.

If that 130 dollars surprised you, the problem is not the number, it is that you have never calculated it until now.