Don't miss
  • 12
  • 6468
  • 6097
  • 20

Why episodic game will never work

By on November 24, 2011
Print Friendly

Why has episodic gaming never taken off? It seems so simple: make part of a game, build an audience, earn the money to make the next episodes and repeat ad infinitum.

Somehow it never seems to quite work that way. The reason is basic maths.

Let’s imagine that you can keep half of your audience from one episode to the next. (Given how few people finish a traditional AAA game, this seems like a high retention rate to me).

Let’s imagine that you start with an audience of 1 million players.

  • Episode 1: 1,000,000
  • Episode 2: 500,000
  • Episode 3: 250,000
  • Episode 4: 125,000
  • Episode 5: 62,500
  • Episode 6: 31,250

In other words, by the time you are 6 episodes in, you have only got 3% of your initial audience buying your game. It’s going to be hard to make money from an audience that small.

It doesn’t have to be that way. You could be gaining followers with each episode, not losing them. Your audience could be loyal. You might be able to reduce development costs so much on the later episodes that you can be profitable with only a fraction of the initial audience.

But the habits of game developers make that unlikely. They focus on initial launch. They are poor at designing for customer retention rather than customer acquisition. They save their best for last (when few people will see it).

The maths is not an immutable law of nature. It can be beaten. But it is a long, hard and challenging slog compared to the ongoing, iterative development of free-to-play games.

About Nicholas Lovell

Nicholas is the founder of Gamesbrief, a blog dedicated to the business of games. It aims to be informative, authoritative and above all helpful to developers grappling with business strategy. He is the author of a growing list of books about making money in the games industry and other digital media, including How to Publish a Game and Design Rules for Free-to-Play Games, and Penguin-published title The Curve: thecurveonline.com
  • Pingback: Episodal Games | Crystalline Green()

  • Pingback: How many people finish a game | Nerd Gamers()

  • Andrew Eades

    Having released an episodic game, I can say that you do not lose 50% of your audience each episode. And they do and can work. We are very pleased with Blue Toad Murder Files and the episodic model we used helped make it a success.
    However, I think we are not talking about the kind of audience building model that Nicholas describes as a broadcast model. Games are not broadcast so it’s always possible to start at the beginning so you will always sell more early episodes than later ones.
    I’ve only just got around to watching The Sopranos. I like hour-sized chunks of entertainment and I buy it 24 episodes at a time. This is the episodic model we look to emulate.

  • One thing I’ve been wanting to ask: what about episodic content in books? It has been proven to work, look at Wheel of Time or whatever.  Books take as long to finish, you can’t understand the later books if you haven’t read the ones before, all things you mention apply here as well. And yet they work. So why shouldn’t the model work in Games?

  • hitOrRun

    So you base your entire argument around this 50% carry-over per episode conjecture? If you hadn’t pulled it out of thin air, then maybe you would have a point. How about you provide us with some sales data on how actual episodic games perform, such as the sam and max series?

    Also, referencing the “game finish” statistics to say that 50% would already be optimistic does not seem very fair.

    From these numbers it is evident that those games that provide the most
    “content” (for lack of a better word) are the ones least likely to be
    finished. (Most likely because finishing those will take the most time)

    If those six episodes would end up delivering an amount of content equal
    to GTAIV, would it not be fairer to then take GTAIVs completion rate of
    25% instead of 3% as the end result?

    Maybe, but it would still be completely made up. Way too many factors
    weigh into this, so you can’t really make a projection like this.

  • James Coote

    Going back to TV shows again, they monetize primarily using adverts. Only the more popular shows will make it to box-set.

    Imagine sitting through a 30 second advert at the end of each boss battle? I can because I play on a Counterstrike server that already does this (shows adverts when the server changes level). The map cycles on this server are set to about 10 minutes, so over the course of an hour of play, you are exposed to an equivelant number of adverts as a TV program

    Interestingly, you can close the video of the advert (which is in the MOTD), so that it doesn’t hinder or interrupt gameplay, but the soundtrack of the advert continues in the background.

  • There are some fascinating stats on how few people finish a game over on Gamasutra

    http://www.gamasutra.com/view/feature/4171/staying_power_rethinking_feedback_.php

  • Thanks for commenting, James.

    My issue is that most episodic games gate the content. In other words, you can’t just pick up the game at episode 3 and have fun, in the way you can with a TV show. This isn’t a fundamental issue with games, more in how they are done.

    I also don’t think that a virtual goods game is episodic. Content drops in an MMO/Virtual world to extend the story line is quite different from asking players to stump up to get access to content, which for me is the heart of the episodic model.

    My issue, basically, is that it will get harder and harder to sell content to players, and that is the heart of the episodic model.

  • James Coote

    What you are basically saying is that most developers make AAA games like films. They invest a lot of money up front, and it is an all or nothing proposition. It has to be finished or they lose everything, and even if successfully made, it could still end up being a flop

    However, It is a fairly ludicrous assertion that an episodic game would halve in number of players after each episode. It didn’t take me long to find two different TV shows and their ratings. One, a cancelled show, http://www.gateworld.net/news/2011/05/an-open-letter-to-stargate-fans-from-syfy/ still did not drop in anything like the manner you describe before it was canned. For comparison, http://en.wikipedia.org/wiki/File:LOST_TV_ratings.png a highly successful tv series. Clearly, numbers do not halve every episode.Just as you would not have a TV episode run for 1h 30m, if your players are not reaching the end of the episode, then you as developers are doing something wrong.

    Also you assume you are selling the game / content, something you’ve argued against doing in some of your other articles, rather than say selling virtual goods or advertising space.

    If we can have successful, episodal TV in the same world as successful film industry, why can’t we apply the same ideas to games? 

    TV studios have a big selection of props, clothes etc that they recycle from old shows or hire from a props company. They make a set, grab some actors, some crew and film a pilot. If successful, the studio boss will give the green light (and funding) to make a season, and if it starts, but ratings slide, they can cut losses and ditch it early

    For games, we can use an off-the-shelf engine and buy or recycle from previous projects all the prop and character models. If we aren’t worried about innovative game mechanics, we just use all the best techniques for story-telling to make our game.

  • Andrew Eades

    You lose a lot of people after the first episode but it tends to straighten out more than you think. There is an upside to episodic but it is not in the sales of single episodes. You have to think about the whole ecosystem of episodes, bundles and the long tail of a full game many months after you tried to sell the 6th episode separately for the last time. For instance, episode 1 sales will spike every time you release a new episode or pricing structure. We’ve now sold nearly 300k episodes of Blue Toad Murder Files but you can only buy them in a bundle these days.

  • I don’t agree that any of your examples are episodic in the sense usually used in games. They are sequels, expansions and annual iterations, all of which are proven to work. The key is that each game can be played on its own, and there is no expectation that you played the earlier ones.

    Most of the time I hear people talking about “episodes”, it is by people (either creators or players) who love story-led gaming and believe that episodic gaming is the secret to enabling the development of more story-led games.

    My maths doesn’t apply to games that have sequels. i agree. It applies specifically to games where game #2 makes no sense if you haven’t played game #1, game #3 makes no sense if you haven’t played games #1 and #2 and so on.

    The post was aimed at games developers considering the “episodic” model. I hope that they learned from it.

  • James Gallagher

    This is a disappointing article not up to your usual standard Nicholas. You seem to be using an exceptionally narrow definition of episodic focusing on just one specific failed example in order to justify a provocative headline.

    Just three examples of successful episodic models:

    1) Games that are pitched as trilogies, ex. Mass Effect and the upcoming Halo trilogy.

    2) DLC packs that expand and continue the main game, ex. GTA IV and Oblivion.

    3) Sports games that get yearly updates, ex. Madden, FIFA, etc.

    In reply to your comment …
    I think that expecting people to pay for an enduring story over 6 or more purchase decisions is very hard. 

    I agree, but when people buy games, and their sequels, they are not paying for an enduring story. They are paying for an enduring experience. And as a wealth of examples show at this time of year, every year, it is EASY to sell “Sequel of Sameness 3”. People like to stick with what they know, and I agree with other commentators: any game that spawns a sequel can and should be included in the definition of episodic.

  • The facts that TellTale produced three consecutive seasons of Sam & Max as episodes and that they continue to use the episode model for their other games are evidence enough that your theory is wrong. It is clearly working for them.

  • To me, that is not episodes, that is sequels. The movie model, not the TV model. I don’t know anyone who thinks of MW3 as an “episode”. That is reserved for games like Sam and Max or Blue Toad Adventures.

    I don’t think that there is no value in building an enduring franchise; I think that expecting people to pay for an enduring story over 6 or more purchase decisions is very hard.

  • Chris

    By that logic, every new version of Call of Duty or Angry Birds should lose audience. This is not happening yet – quite the opposite. Your assumption that audiences always shrink is fundamentally wrong. The challenge is to make sure that each episode works as a jumping-on point for new players as well as keeping the established fans happy…

  • Angry Birds?

    I don’t think selling per episode works for the reason you specified. However I do think over-delivering episodes to customers who already exist as a marketing vehicle to drive new purchases (as Angry Birds does) can.

  • I dont agree with you that ME2 is episodic. These are stand-alone epic games. Sure, it makes more sense if you’ve played the first one, but isn’t it more like watching X-Men 1 before you watch X-Men 2. In essence, I don’t anyone has cracked the TV-style episodes in games that Dom suggest works so well for Dr Who.

    Similarly, I dont think level games would work. I suspect that Nintendo would make less money, not more with a content-selling model. The free-to-play model focuses on giving content away for free, and selling items of self-expression, of power, of status, of progress or of time-for-money. I think that the level unlocks would run into the law of diminishing returns very quickly.

  • Nzen

    I think the time has never been better for episodic games. Although, having “Episode: …” in the title may be a bad idea. I don’t want to wait 2 years for someone to finish a game I probably won’t even complete. I’d rather play it in early alpha form.

    You could say Mass Effect 2 is Mass Effect: Episode 2, etc, but people still buy it. If each episode seems standalone then people will be willing to buy it out of order, as Dom said in his comment.

    Generally, only sandbox games are possible to release in an iterative fashion. An alternative that I can see is that they release a “hub” game/client and patch “episodes” or content into that and charge for it inside the app. This is hard for linear story games, but for “level” games it would be doable. For example, if Nintendo released a Mario 64 hub which was just the castle, then patched new “doors” into the castle once every few months, making you pay $5-10 to open each door, they would make a killing.

  • Dom Conlon

    This is good. Very good. It does seem that you are suggesting episodic is about continuation; that episodic gameplay demands that the player finish one before continuing to the next. I think this is because games are often see in terms of movies. The Call of Duty franchise is seen as a series of movies rather than a TV series with episodes. Maybe that’s why we only get one a year with such massive marketing spends being pushed to a single shot rather than a steady spend urging us to join in with the adventure as and when we feel like it. Plenty of people will dip in and out of Dr Who for example and a good writer will ensure that this sort of casual access is possible.

    But I do think that the approach in gaming takes a single story and then splits that into episodes. Maybe if smaller, self contained games were created for a franchise then it would be possible to avoid the diminishing returns issue you highlight.