What game studios can learn from the startup world?
I’ve been involved in tech startup businesses for a significant part of my life and I’ve been running a gaming (startup) service for the last couple of years.
After talking with a lot of people from both industries I see some significant gaps between them. My purpose is to merge these two worlds and highlight learnings game studios can use to make profitable games.
So, let’s dive in!
For innovation think like a startup
Startups all over the world are playing a very difficult Olympic game. It’s called creating something that doesn’t exist but when it’s built, their bet is that it will have a significant impact on people’s lives. Founders and their teams are like explorers in the open sea trying to find new land to conquer, but since no one has ever put their feet there, every resource found it’s theirs to keep.
Ok, I am getting too emotional with this but this is how risk-reward works in the world.
This is how the gaming industry works, and entertainment in general. Once you create a hit game (Product-Market Fit in the startup terminology) you can scale up and return your investment many times over. Riding existing trends means that someone else is ahead and THEY are getting all the rewards.
I hear studios are getting worried about how expensive CAC has become and it is becoming really hard to afford it without a publisher or investor. Marketing math says that you can’t afford to get all the eyeballs you need. By creating an innovative game that people love, this becomes easier because the virality is higher and you spend less acquiring new users!
Learning is the only KPI before launch you should care about
Innovation though has a higher degree of failure and the goal pre-launch should be to minimize that risk. You do that by having more data points about its potential success sooner. Eric Ries’ book, Lean Startup, called it “Validated learning”.
“How much validated learning are you getting for your efforts?” Eric Ries - Lean Startup
If after launch, businesses measure the growth rate of revenue, the only way to measure our progress in the uncharted waters we are sailing before the “big” soft-launch is the growth rate of learning.
How many users enjoy level 2? How clear is it for the user to complete on-boarding? How hard is it to beat the end-boss in level 5? How possible is it to recommend it to their friends?
These are some of the questions you can get answers to before even polishing the game. By running multiple experiments (against a potential audience) you get new distinctions about what’s their ideal game and how close you are to that. And then you can change things and run a new experiment.
Change things, not fix things
The definition of change is replacing. When you run a new experiment you replace a block of experience with some other block. But when you fix something, the dictionary says that you repair it, probably after it’s done. And that’s the most costly way to build a game.
I don’t know how many stories I have now collected from people telling me that they’ve built a game that it didn’t hit its goals on engagement and monetization on soft-launch and they now have to go back and fix the game which means more development time required, frustration, crunches and many more bad things happening. Plus, all that marketing money you gave must be spent again! The other solution of course is to accept your fate and move on to the next game!
The point here is that failing early and often is a much better outcome than 1 big failure in the end. You might even have to kill “whole games”. Supercell is very proud of that! To make 4 games, we killed 14, because they weren’t good enough for you
Users are always right
Which leads us to the next point: “Good enough for you”.
You are building a game for a very defined audience. Your opinion about your game doesn’t matter! Unless you are your target audience. And even in that case, you are too close to look at it objectively.
So the best thing to find out what they want is for you to talk to representatives of that audience about your game as often as possible. And I know some of you might already be thinking, “I know my audience!“. Although I don’t know you, you probably don’t. You don’t know them THAT much to have a clear understanding of what they are going to think, feel and then how they are going to physically react to your game. So many startups and studios I know have made this mistake.
Build, test, measure, learn, (repeat)
By giving them a prototype, vertical slice or whatever you can give them that they can play, you will find their opinions about it and then adjust. And then test it again. Real knowledge comes from the distinctions between 2 different pieces of work.
The Lean Startup Methodology is the less stressful way I know to build a game or anything new.
Big studios have the resources to be more user-data driven and they are learning faster from their audience and that’s why they are winning. By including testing as often as possible along the whole development of the game, you avoid being too wrong, too late and you have a chance for a home run.
Our playtesting.games platform has been created with that intent. It enables you to run test cycles often, collect as many data points you can from your audience sooner, and as a result learn tons more with each iteration you put out. We’ve had studios running cycles of the same game for months and with each version they are improving their learning rate by double digits!
If you are doing this often enough, you can have a true masterpiece. Both artistically and commercially!