Episodic Content: Big Questions and Answers...

[.:|Physalis|:.]

By Mark Thomas


Part 1:

First and foremost I have some real concerns about consumer expectations. As we in games increasingly compete with traditional media for those dollars our success will likely be determined based on how good a job we do of providing more entertainment for the dollar than those other options. 

Currently I think as an industry we are doing really well here. A new game today costs, per hour of entertainment, a good deal less than a movie or some other ways that customers could spend their money. However, as we move into products that are set on similar time scales (i.e. a few hours of gameplay) customer expectations could change.

Episodic content is still new and different as is any sort of online distribution, so customers will be likely to approach it cautiously, and a bad experience, or unmet expectations will likely result in lost customers. Failures in the early episodic content models could result in a chilling effect on other companies planning to take the plunge. The two major factors in the expectation setting process are price and scope, and it ties back into that same question of 'what can I buy with my entertainment dollar?'. My expectations of what I get for $5, $10 and $20 are different and my expectations probably don't match yours. 

From a gameplay perspective I look at how people play games, and how much game I would want in an episode. I think about three hours makes sense.

Gameplay time

I suspect if we went out and asked potential customers they would think of an episode as a bit longer, and if we asked the folks who had to make the content on some sort of regular schedule they might shoot for something a bit shorter. Also users obviously proceed at their own pace, and in fact during every hour of gameplay there is some portion of users that stop, get stuck, or lose interest.

The key will be to set a pricing model that the customers will feel like meets their expectations, while still delivering enough in the way of profit to continue the project. I suspect that the first pioneers in this space will set some expectations we all have to live up to for a while, for better or worse as users decide if their money was well spent customers will react quickly and vocally as to their perception of the deal.

On a positive note, episodic content has the potential to help us as an industry escape some of the pressures that have been limiting our ability to take risks. As games have become increasingly expensive to make, the choices for developers and publishers have become increasingly difficult. 

This pressure is one reason episodic content is so attractive in a lot of ways. If you were to conduct a survey of your customers, what percentage would you find had played your title through to completion?  I don't have access to a giant research team, but I think it would turn out to be significantly less than 50%, probably in the 30% range. Obviously that is going to vary game by game, but that gap represents a massive dollar investment in content that gets made, but never seen.

Pretty compelling

We are building and paying for content that never gets seen by our customers every day. The idea that we could build pieces of games and sell them directly to gamers is pretty compelling, as is the idea that we could evolve and build a larger story and world over months, and years. 

At this point many companies are still investing heavily in technology for every title. This big up front cost is then compounded by the content creation costs. If we can reuse more and more technology across episodes, and eventually across multiple titles, one of the major cost factors could be dramatically mitigated. Add on to that the ability to build only as much game as the majority of players need and one can imagine a scenario where there are more opportunities for experimental and riskier content to make it to market and at least get a chance to prove its value.

Calculating the risk reward tradeoff is one of the toughest things publishers have to do on a regular basis. While in some ways episodic content can help make this an easier calculation, one thing it can't be expected to do is make us smarter. We will learn from doing over time, but this won't be a silver bullet.



Part 2:

True Episodic Content

In most of the cases I've seen thus far where people have shipped additional content it has taken the form of an expansion, or a few extra levels for established properties.Over the next year I think we will see a few cases of true episodic games, either shipping fully in an episodic format, or having major planned single player plot advancing segments scheduled for after their retail release. 

This is an important step forward in getting customers comfortable with the idea, and the structure, but does less than we would like to alleviate the real risk for either publishers or developers. There is no reason to think that we will as a group be any better about picking out hit games as episodes than we will with boxed retail product. This means that we should assume a similar success/failure ratio for episodic games as we see today with traditional titles.  Our failures will be less expensive perhaps, but still for one reason or another some titles will simply fail to capture the attention or interest of customers. Simply taking your game and splitting it into 10 planned episodes might make it easier to get out of the title if it is failing, but it won't make your game any better.

What we need is some more effective risk management models. The possibility exists to hedge our bets with episodic content in a way that we can't with traditional games. Keeping costs under control, and understanding that everything we produce won't be embraced by the public. This problem has been carefully managed for years in film, and especially in TV.

Similar strategies

Episodic content in the gaming space is our chance to adopt some similar strategies. If done right this will mean that there will be more projects cancelled, but more projects overall and studios should be able to make a good living even on moderately successful titles. Additionally, many more titles will be judged by the consumer rather than five guys in a board room somewhere. If we look to TV, which probably has the closest parallels to episodic content we find that in general, the networks, and studios create only as much content as they can sell, in this case to advertisers. Additionally the model is structured such that the big financial decisions are back loaded rather than front loaded.  We could learn a few lessons here.

Getting a TV show on the air is a pretty significant process. At every step along the way there is careful analysis, feedback, and research to judge the worthiness of the particular project. Most scripts never get pilots, most pilots never get larger orders, and many shows never make it past the first season, or even half season. 

With many modern shows costing upwards of a million dollars an episode to produce deciding to commit to twelve or twenty four episodes is a big commitment, so you can bet that the shows that make the cut have been carefully scrutinized. Tons of potential customers, advertisers, etc., all get to weigh in with their opinions. Significant re-toolings often take place before the show is ever shown to a larger public. In the games business today, by the time we have a really good idea of what the experience is going to be like, we have already spent a significant percentage of the total costs.

There are also a ton of early outs in TV, early outs before the big money gets spent. Don't like a script and you are only out the cost of the script, don't like the pilot and sure you had to pay for that and the script, but compared to the twenty four episode season it turns out to be pocket change. 

Even more interesting, is that once a show is in production, the math becomes fairly straightforward on if/when to let it end. Will the ad revenues cover the production costs? Is there something more profitable that can be put in that slot? There are fewer variables than we deal with when making mid production decisions, because they are able to see how the customers react to the product all the way along. In games our current model is the equivalent of approving and producing an entire season of a TV show without ever showing it to the customer. 

Perhaps the most important thing that the folks that make TV shows have figured out is predictability. While in season they manage to release a new episode every week, or at least on some predefined schedule.  I don't hear a lot of stories about ER slipping a month, frankly I'm not sure how long it actually takes to make a TV show. The thing is, it doesn't matter. They understand their technology, their process, and their constraints to the point that they have a sufficient buffer for the randomness of life, and consistently deliver on their commitments to the customer.  This predictability allows them to keep customers engaged from episode to episode, something that gets a lot harder if you have 6-12 months between episodes, or years between sequels.

Technology investment

Episodic content gives us the opportunity to bring some of those benefits across into our medium. If we can leverage the technology investment across multiple titles, the costs of producing an episode will drop. Our up front costs will still be more than a back end episode, but technology and content sharing will help here. I imagine a model where a publisher (or a group of developers) secures a set of technology and starts production on a number of titles, contracting with developers to build the first few episodes of a unique concept. In addition to an engine and tools, I'd expect the publisher to provide a variety of best practices and specifications for asset creation. The goal here is that as much as possible is usable across multiple titles, the technology, and potentially even objects and set pieces.

In the early stages I'd expect to see some additional work done on story arc, and future content, i.e. you know you are going to need a car in episode 4, might as well build it now. The development process should focus on predictability. Exactly how long did it take to make the first episode, how long for the second, how long for the third. With the existing tools and technology, how long is the entire production pipe for a new episode from paper to execution. The publishers goal is to pull this all together. Once the first few episodes of each game are ready, by taking ten to twelve first episodes and putting them on a disc or online distribution system you give the consumer enough reason to pick up that first disc or go to the web to check out the new games.  On day one, the second episode for each game should also be ready to go.

Here is where the risk management portion comes in. At this point the outlay for the titles is hopefully fairly constrained. Technology was covered, and development costs for three episodes should compare pretty favorably to the costs for a full game. The consumer is going to pick up this disc (think publisher x's fall season) and start playing the different episodes.  Some of the games on the disc are going to capture the market's attention, others quite simply won't.

The key is that we can use that second episode to actually know the answer to the question of which one is which.Downloading that second episode shouldn't be free, either the planned retail price of an episode long term, or maybe a discounted rate. Perhaps buying the disc gets you 2 second episodes for free, or perhaps all the first episodes are free, but you pay for continuing. There are lots of business options here, but you have to assign a real value to the consumer for these things, because it is about predicting future behavior. Likely consumers will want options for single episode purchase, and some sort of discounted “subscription” option.

Over the first few weeks as second episode purchases role in, I suspect that it will be very obvious which titles are capturing user attention.  Simple tracking technology should tell us a bunch of information about how many users are playing each title, as well as how many paid for episode two. If we do a good job of determining the production costs per episode the math of number of subscribers to fund an episode will be blindingly obvious. I suspect that out of 10-12 games we will see 2-3 that generate enough interest to keep making episodes. 

Now this is going to mean that we need to move into full production on episodes for those 2-3 games pretty quickly. The question of how frequently people need to see episodes in order to keep them interested is an open one. I’ll bet we will find that customers would consume the right kind of content on a monthly basis, or even more frequently.

The networks seem to have no problem getting people to tune in weekly, so the idea of frequent smaller chunks of content is already pretty well ingrained in consumers. In order to hit this kind of timeframe we will probably need to have multiple teams working on different episodes which means that some of those teams whose games didn't get picked up will still have work to do. This will require a different kind of mindset about content ownership, but we have plenty of teams doing expansion work for others people's games today, so with some coordination effort this should allow a bunch of studios to deliver a relatively seamless evolving experience for customers.

Over time, people will naturally begin to lose interest in games, and story arcs will be taken as far as they can. Simply keeping an eye on the subscriber numbers should tell us when the business case for a particular game is getting weak.  As those numbers start to drop below or approach the break even point, simply make the last episode in production a wrap up episode, and stop working on new ones.  That should allow a smooth ending to the game, and leave customers feeling good.

The math on this model actually works out pretty well even when you assume that we will continue to produce games that are less than hits.  One or two strong titles will more than compensate for the failures, and if done correctly we will get more shots at a hit for far less than we spend today to build complete games, and we will focus on building only as much game as our customers demand.

[Turrican3]

Sarò conservatore, sarò quello che vi pare... ma l'idea del videogioco-telenovela, prospettiva peraltro già accennata qualche giorno fa da un pezzo grosso di Sony, mi piace pochissimo.
E quel che è peggio, non riesco a vederci altro (e del resto l'articolo non è nemmeno pensato per nasconderlo) che vantaggi per le software house anzichè per l'utente finale.  ::)

Verò è che, a ben pensarci, già adesso possiamo vedere degli abbozzi di strategia simili a quella delineata nell'articolo: inutile girarci attorno, ma a prescindere dal giudizio sui titoli specifici e ben conscio del costo odierno dello sviluppo di un VG, non posso che constatare che 4 episodi in 5 anni di un Jak & Dexter o di un Ratchet & Clank sono dovuti, oltre ad un buon riscontro di vendite, anche alla necessità di "spalmare" su quanti più titoli possibile l'investimento iniziale umano/tecnologico ("motore" 3D et similia).

Comunque se proprio devo scegliere preferisco quest'ultiimo approccio, ovvero anche 4/5 sequel a generazione, ma completi, che diamine!