Why you should make an Idle Game

I wrote last week about auto-play being one of the trends for mobile games currently. When pushed to it’s extreme, auto-play becomes an idle game. That is, a game that the player hardly plays at all. By then, we are simply left with a Skinner box.

If you are unfamiliar with Skinner boxes, go ahead and have a look at the Extra Credits explanation of them here. He will also explain to you why core gamers loathe the concept.



If Skinner boxes are so clearly bad, then why are auto-play features and idle games a trend? I’m actually late on commenting on the trend already. About a year ago, it was already all the rage, as both Gamasutra and Pocketgamer pointed out.



First, notice the “good” game mechanics that the Extra Credits video suggested we should use instead of what he calls Skinner Box Operant Conditioning. These were:

  • Mystery,
  • Narrative,
  • Novelty,
  • Mastery,
  • Mental Challenge and
  • Flow.

Really, that’s only 2 things, with 6 different names. Mystery, Narrative and Novelty are 3 different names for Stories, while Mastery, Flow and Mental Challenge is just 3 versions of, well, Mastery.

There are 2 things we can conclude here: the guy making the video is likely to be an Explorer/Achiever sort of Bartle type. 

And, games made with his suggestions can actually never work as (mobile) F2P games. Which is a shame, but also very hard to change. I believe Dimitar Draganov is right when he points out that making mobile F2P games for Explorers is very hard. Get his book from Amazon.

First off, no team in the world can create Novelty, Mystery and Narrative at the pace required to keep the game’s superfans happy. Players will simply burn through the story in days, while it will take you months to create more. A game relying on these just cannot keep the players coming back for years.

The same is very likely to be true for Mastery, Flow and Mental Challenge. In an earlier post, I examined why traditional puzzle games with Mental Challenges cannot work as F2P. 

Simply said, when the players get stuck, they will quit playing. With these sorts of challenges, there is no luck or grinding that can help you. When you’re stuck, you are stuck and will leave the game.

Other sorts of mastery won’t get your players stuck, but they are likely to feel stale after a few months of playing. It’s rare that a pure core without any progression on top can keep players long-term. Sure, for some player types, Chess and Go will keep their interest. The vast majority of the population will not make even those games into a daily habit. They will need some form of progression in the meta game.

Even core gamers often degrade into pure progression players eventually. When I have stopped playing the core game, I notice that I still keep playing the meta game for a while. When attacking starts to feel formulaic, I will still go back to the game just to collect resources, and upgrade something. I no longer bother with actually playing, but I do log in to progress. I have noticed the same pattern in colleagues – both in mobile games and in PC MMO games.

All of this means that progression mechanics are really super important for successful games that can be played for years.

This also ties into the evolution of the collection mechanics in F2P games. Early on, Tamagotchis demanded you to respond at a specific time. If you did not, they would punish the player by dying.

When F2P really took off in the west, appointment mechanics were popular. In FarmVille, I could choose what crops to grow, each with a different time until I needed to come back to harvest. If I missed the harvest window, the game would still punish me by having the crops wither.

The next step has been the Collect Anytime mechanic found in e.g. Clash of Clans. They just keep producing gold up to some limit. I can collect the gold at any time, but I will not lose it if I do not. Once I am at the gold limit, it will just stay there, without growing or shrinking. Of course, Clash of Clans still kept some of the punishing mechanics as well, since other players attack you and steal your gold.

Idle game mechanics are one more step towards being nice to the player. There is no longer any cap on how much you can earn without coming in to collect your resources. You will simply keep earning. Of course, if you come in often, you will be able to re-invest your resources into new production, and hence earn even faster.

Tamagotchis to Idle games becomes a continuum between using carrots or sticks to get the player to come back. Are you threatening the player with something bad happening if the do not keep playing, or are you rewarding them with something good?

I would guess that keeping some threats involved can keep players returning more often – until they miss a week and are severely punished – at which point they never ever play that game again. In contrast, a game like AdVenture Capitalist will keep on earning resources for me even when I’m not playing. If I get curious again after a month of being inactive, there will be a huge reward waiting for me in the game!

Back to why we should make Idle games. Most game designers are core players, and find it harder to grasp the retention/meta game. By forcing us to make a game without a real core, we have to get it. Which means that should make an idle game as a sort of F2P masterclass. If there is nothing else in the game, we have no excuses. We have to make a good meta game when that’s all there is.

With the recent successes of Idle games, I would bet that a small wave of them are now about to hit the App Store. As always, most of them will fail to find significant commercial traction. However, I think that even those games will teach valuable lessons to the teams that made them.

Trends in Mobile Games

The very first post I wrote on this blog was about test playing and deconstructing games. It is really, really useful to do, but we never seem to have enough time to do it as much as we should. Of course, you can learn a great deal by reading through game deconstructors that others have done. For instance, over at deconstructoroffun.com you can find a lot of good stuff.


Now, there’s something even better available. How about systematic deconstructions of all the top grossing games, as well as a lot of the lower ranked ones? Over at gamerefinery.com you can get that. The data there is down to a very detailed feature level analysis, together with trend-lines for the features and for different game genres.


They have analysed all these games, and are then running some sophisticated statistics to figure out which features are most important for getting a game to rise up the charts, and which features work well together. Of course, making a good game is not simply about ticking the box for a number of features. Nonetheless, it helps a lot to see what the successful games have in common. As an analogy, reading a recipe will not make you into a Michelin-star chef, but understanding which ingredients the best chefs often combine will still make you a better cook.


Also, bear in mind that this data will tell you how to rise on the App Store top grossing charts. Ad revenue will not show up on that chart, only IAP revenue. This will tilt the numbers in favour of complex games with large in-game economies, rather then the quick ad-driven volume games like what e.g. Ketchapp is doing. That said, below is a very brief summary of what the data tells us.


It will come as a surprise to absolutely no one when I tell you that the single most important area for App Store success currently has to do with publisher and brand. Branding a game with a well-known IP will help it break through the noise on the App Store. Something published by King, Supercell or Machine Zone is way more likely to break into the top charts than something published by small indie companies like our Tribeflame. On the one hand, it looks like this supports indies publishing through larger publishers. On the other hand, none of the companies who publish third party games have a really consistent track record – the three publishers mentioned above will only ever publish games developed in-house.


A second trend that you can see is that games are becoming ever larger in scope. According to this data, I helps a lot that the game is both broad – with lots of different troops, buildings, enemies, puzzle types, etc. – and deep with lots of levels, upgrades, etc.


To further stretch out the mass of the in-game economy, gacha mechanics are trending. This means that the player cannot directly buy some of the things they want. Instead they can buy lottery tickets that might give them something really cool, or might give them something useless. It’s sort of like buying a pack of trading cards in a physical game like Magic the Gathering. With all else equal, it will take the players a lot longer to complete their ambitions when there is an element of luck involved, compared to when they can just buy directly what they want. It is also more fun and engaging.



A third big trend is the rising importance of a number of social features. Broadly speaking, the more space you give to player interaction, the better your game will do. User generated content, trading resources between players, stealing said resources, helping clan members out, seeing others’ progression, and player-vs-player (PvP) battles. All of these will help your game. And yes, synchronous PvP is a trend, and according to the data, I might have been wrong in criticising it as a hype wave a few weeks back.


A forth category that sticks out for me is slightly surprising. It is that part of the gameplay is automated, or can be automated. As an example, think of the battles in Game of War. You don’t really do anything as a player in the battle itself. It just happens, and you’re told the result. In some games where the player actually plays the battles, they are also given the option of skipping this gameplay and letting the computer figure out what happened. My interpretation is that long-term, it is the meta game that keeps players retained and spending in the game. The core game might by then become repetitive and boring. Hence it can be automated to let the players concentrate on the progression.


There you have it: the broad recipe for success on the App Store today. Let’s recap:

  • Build a huge in-game economy with some random elements (gacha).
  • Give it meaning by having as much player-to-player interaction as possible.
  • Automate any section of the core game that can get boring in the long run, and let your players focus on their progression.
  • If you can, brand it with something popular and release it through a successful publisher.


That shouldn’t bee too hard, should it?

Actually, I think that the data is slightly over-stating how complex successful games need to be. You see, a successful game will get a team to maintain it, and will thus become larger and more complex over time. The causation thus goes both ways: larger games become more successful, but more successful games also become larger over time.


Of course, this is all just my reading of the data, where I group together several of GameRefinery’s features into four broad trends. You should go have a look for yourself to see what you can take away from their analyses.

How to choose who your players are

Make games for players who have their own credit cards!

You should not choose who your players are. If someone you thought would never like your game turns out to love it, be happy and try your very best to make them happy. However, it is a good idea to have a customer in mind when you start developing your game. And picking yourself might not be the best answer. The developers of Sunset even said that “whatever we enjoy is never, ever, what the gaming masses enjoy.” http://tale-of-tales.com/Sunset/blog/index.php/and-the-sun-sets/

Most game devs are from a subset of the population. We like certain things. If everyone builds games for themselves, 5% of the population ends up with an oversupply of games tailored for their tastes, while 95% of the population will have no games at all to suit their preferences. Even if the typical game developer is also a heavy consumer of games, we cannot all just be building games for ourselves.

Another common misunderstanding, at least among the general population, is that games, and mobile games in particular, are made for kids. Looking at the early App Store success stories – and some of the current ones – you might very well get that impression. They have very cute graphics with garish basic colours. Surely, they are made with kids in mind?

Let me tell you that they are not. We’ve made that mistake, and advice you to learn from our mistake. If you want to earn a living from a game, please make sure that your customers have their own credit cards. If your main target group is too young for that, you will earn just about nothing from them.

You see, if a kid wants to spend in games, they will need to ask their parents to do it for them. By the time the kids explain it to their parents, it really does not matter anymore how great a value you have constructed around your In App Purchase (IAP). All the parents will hear is that their children want to spend a few dollars on something completely useless in a garish phone game.

On the other hand, if you get the parents themselves hooked on your game, we’re in a completely different place. Now, they really want to achieve that cool thing your game shows them. And since they have a little money, but are pressed for time, it suddenly makes complete sense to spend only the equivalent of a cup of coffee to jump ahead a bit in the game!


So, where should you start when deciding who to build your game for? There are a few deciding factors:

your team’s skills and interests. Yes, this does play a role, but it’s not the only factor you should take into account.

your existing fan base, if you have one. We’ve learned the hard way that some cross promotions work, while others are completely useless. If you jump too far from one customer group to another, you will have to either partner up with another company who already has access to those customers – or spend quite a bit on User Acquisitions (UA) to get a foothold.

competition already in the market. It might not be a good idea to go head to head against King or Supercell. Just saying…

player types and demographics. The standard way of grouping players is the Bartle model that divides players into four groups: Achievers, Killers, Socializers and Explorers. http://www.gamasutra.com/view/feature/6474/personality_and_play_styles_a_.php?print=1

It’s useful, but also limited. Dr. Bartle originally did this grouping based on how players behaved in one particular online game. Which means that it should be based the mother of all skewed sampling methods. I, personally, would probably never have played that game (a MUD, which is an exceedingly nerdy all-text adventure), and thus I, and people like me, would not have been part of the sample when the model was constructed. Dr Bartle himself has said that his model does not apply to all games: http://gamedevelopment.tutsplus.com/articles/bartles-taxonomy-of-player-types-and-why-it-doesnt-apply-to-everything–gamedev-4173

For most mobile games, we can do better than Bartle’s basic MUD-based model. Actually, the University of Turku, who is just next to us, has recently done research on how people play games. Their research covers the whole population, rather than only people interested in specific online games.

With a sample of 3400 people (general population, rather than just self-identified gamers), they were looking into what kinds of game mechanics people prefer to play, and which ones go well together. After running the answers through some statistics, people seemed to group into 5 categories:

  • assault (killing, destroying, stealing, sneaking etc),
  • manage (developing cities, managing resources, directing troops, manufacturing vehicles etc),
  • journey (collecting rare items, developing character’s skills and abilities, exploring the story and the gameworld etc),
  • coordinate (jumping on platforms, staying in the rhythm, match3 etc),
  • care (kissing, hugging, having sex, taking care of pets etc)


The study is not published yet, but I’ll let you know when you can get the details of it. (And, yes, I realize that this is not a huge way off the 4 Bartle types.)
One last tip: a few raving fans (and a bunch of haters) are better than merely OK for everyone. Remember that it’s the superfans that pay your salary. Just Google the most disliked YouTube video ever if you want proof. It’s by Justin Bieber. I hear he’s doing alright financially.