Recently, 1Up’s Chris Pereira penned an article examining how Bethesda’s Dishonored bucked the notion that new IP will not sell well at the end of a console cycle. The article brings up an interesting argument and, perhaps, draws the right conclusion, but misses the underlying issues.
The gist of Pereira’s argument is expressed thusly:
[A successful launch for Dishonored] is encouraging news. As wonderful as seeing refinements of the franchises we already know and love can be, it’s the completely original titles that are often the most exciting to see.
But considering what we’ve been led to believe, this doesn’t make much sense. This month marked seven years since the Xbox 360 was released, and six years since the PlayStation 3 and Wii debuted. With it only being available on 360, PS3, and PC, the launch of the Wii U (coming just over a month after Dishonored’s launch) can’t be used as the reason for why Dishonored has done well. If anything, it signals either the end or fast-approaching end of this generation of consoles, so surely Dishonored had no business performing the way it did.
In logical terms, Pereira’s argument through the bulk of the article goes like this: if Dishonored sold well, then the assertion that new IP will not sell well late in a console’s life cycle has been rendered false. There is a kernel of truth in his assertion, but the fault in his logic is that publishers do not believe that new IP is absolutely guaranteed to fail during a console’s twilight years, only that there is a better return on investment for sequels.
Okay, to be fair, he does quote Ubisoft’s Yves Guillemot making exactly that claim about new IP: “Transitions are the best times, are the best ways, to make all of our creators take more risks and do different things. When a console is out for a long time … you don’t take as much risks on totally new IPs because even if they are good, they don’t sell as well.” But, public statements aside, the argument is one of risk aversion rather than an absolute certainty.
The Thought Process of a Publisher
Think about this topic from the perspective of an investor, namely a publisher. A simple example: you are a VP or C-level employee. You have $80 million left in your annual operating budget. You can apply that money to a sequel for a well regarded brand, or you can invest in a new intellectual property.
Let’s examine the sequel. You have a high degree of confidence that production will yield a viable game. You know from historical data for the franchise that $80 million is sufficient to cover production, distribution, and marketing, and you can anticipate $100 million to $350 million in revenue – or an average of $225 million – with an even probability. This means that you can forecast that every $1 of investment is worth $2.81 in revenue with low risk.
On a weighted average, this forecast indicates that every $1 invested is worth $0.52…a projected loss, with only a 30% chance of establishing a franchise
Now, let’s think about the new IP. It is important to remember that developing a new IP is an exercise in research and development. This means that the path to market is not guaranteed: at the end of production, you may not have a viable product to sell, and the entire investment could go up in smoke. Again, going by historical data, the possible outcomes you face are:
1. 9% chance of $100 million in sales: a kick ass release and a new mega-franchise – good return on investment
2. 21% chance of $80 million in sales: a solid product that only breaks even, but paves the way for sequels – longer term investment, but still viable
3. 60% chance of $40 million in sales: a product that sells poorly against the cost to develope and is unlikely to spawn sequels – recoup some of the investment, but an overall loss
4. 10% chance of project failure: a project that fails to find a path to market – investment lost, no chance to recoup*
On a weighted average, this forecast indicates that every $1 invested is worth $0.52. This is a projected loss, with only a 30% chance of establishing a franchise. So which option do you pick? And, before you answer, remember that you need to turn a profit this fiscal year or you could lose your job.
The Numbers Behind This Rule of Thumb
If a 30% chance sounds melodramatically bleak, it shouldn’t, judging by recent history. In the chart below, I have attempted to map out a crude success rate of new IP on the Xbox 360. For the sake of simplicity, I have narrowly defined new IP as completely lacking a preexisting brand – like Dishonored – so games like Battlefield: Bad Company, and Batman: Arkham Asylum are not included. I have made several assumptions in this analysis†, but the two most important are that the success rate for new IP seen on the 360 is comparable to other platforms, and the success of a new IP can be judged by whether or not a sequel was developed.
The short version: of the 131 new IP launched on the Xbox 360 between launch and February 2011, 44 were successful enough to warrant a sequel. That’s a success rate of around 33% of games that actually make it to market. If you factor in some degree of project cancellation, the odds get even worse (I picked 10% arbitrarily; it would likely be impossible to determine an actual number as many projects are canceled without being announced).
The inference that one could draw from this data is that there is a prime window to launch a new IP
Obviously, every franchise has to start somewhere. But how and when? Three correlations are present in the data:
- All of the new IP that sold at least 600k units on 360 got sequels
- Most games that scored an “A-” or better at 1up.com got sequels
- The greatest proportional success rate of new IP occurred in the period starting almost a year after the launch of the 360, and ending two years after launch
The inference that one could draw from this data is that there is a prime window to launch a new IP: after developers have learned about how the new platform works and after competition from the previous generation has started to decline, but before a massive influx of new gamers and competing software puts the focus on established brands. This analysis would seem to support Guillemot’s espoused strategy.
But such data always invites the possibility of mistaking correlation for causation. Is there another factor at play?
Another Version of the Truth
Of the data I listed above, one should stick out to the reader: games with high scores from 1up.com got sequels. In other words, people buy games that are really good. What I find most interesting about the ninth month through second year window described above is not just the improved success rate during this period, but the actual titles that emerged during that time, namely some of the most successful new IP of this generation: Gears of War, Mass Effect, and Assassin’s Creed. So were new IP especially successful during that period because that is the most fertile window or because developers were making great games? Are Guillemot and other executives creating a self fulfilling prophecy? If they feel that new IP only has a 33% success rate once it hits the market – or whatever low value their internal number suggest – does that cause them to hedge their bets? Does risk aversion mean new IP gets short shrift when it comes to production and marketing budgets, and, thus, lacks the resources to get a new franchise off to a strong start?
Let’s try another data set: marketing and quality. If we assume that a GameInformer cover article is a fair indicator of a strong marketing push by a publisher, and we assume that a review score in excess of 8.5 by IGN is an indicator of quality (IGN tends to give lower review scores than 1up.com), the graph looks like this:
In this case, 64.52% of new IP that landed on the cover secured a sequel, a big improvement over the overall pool of new IP. But, for games that landed a cover article AND scored above 8.5 on IGN, the odds of success are even better: 81.25% of these games secured sequels. Granted, this a VERY small data set, but the correlation is striking: developers make a good game, publisher puts marketing muscle behind it, and gamers will buy enough copies to make a sequel an attractive investment.
Are Guillemot and other executives creating a self fulfilling prophecy? Does risk aversion mean new IP gets short shrift when it comes to production and marketing budgets?
Incidentally, I would have included 2011 in the chart above, but the only original IP to get a GameInformer cover that year…was Dishonored. And while Bethesda calls it a success, it does not yet have an announced sequel. I’ll let the reader decide whether to include that particular data point.
Some Miscillenia from Pereira’s Article that Deserves Discussion
In the end, the biggest objection I have to Mr. Pereira’s article is not his conclusion, but his rationale.
One of Mr. Pereira’s arguments against reserving new IP for the start of a cycle relates to development costs. He points out that the learning curve flattens out over the life cycle of a product, which should, he argues, make sequels cheaper to produce, and, thus free up more money for new IP. This is not actually the case. Yes, developers do become more familiar with a console’s in’s and out’s as the years go on. But to say that this should make games cheaper to produce assumes that all of the other costs of production stay flat.
This is clearly not the case. Games released at the end of a console’s life cycle are, by and large, bigger, prettier, and more feature rich than those at the beginning. This may be due to more ambitious production teams; it may be because developers anticipate that fans will be dissatisfied with a game or, in particular, a sequel that is simply more of the same; or the publisher might have a feature wish list to make sure a product stays competitive in the marketplace. Whatever the motivator, the familiarity with a particular hardware platform manifests in more sophisticated products, not lower production costs. The current console generation started with Kameo, and is winding down with Assassin’s Creed III.
Developers do become more familiar with a console’s in’s and out’s as the years go on, but to say that this should make games cheaper assumes that all other costs of production stay flat
Another quote from Pereira’s article I disagree with:
Or perhaps publishers are mistaking the lack of competition early on in a cycle as a higher level of interest from gamers. Less competition no doubt makes it easier for, say, ZombiU…to be noticed by being launched alongside the Wii U, but that doesn’t have anything to do with gamers’ desire to play something new and original.
No, publishers do not think gamers are more interested; they know gamers have fewer games. To some extent, early adopters are a captive audience. This is a key point. Granted, with a smaller consumer base at the start of a console, your potential revenue base also shrinks. But, the lack of competition provides more opportunity to establish a franchise beachhead on which sequels can capitalize. Beyond the lack of competition, games that make a console’s launch get a crap ton of free marketing: platform makers and retailers tend to sell hardware at a loss and at cost, respectively, and have to make up the margin on software sales, so they market the hell out of any available game. Publishers can draft on these marketing pushes, further reducing risk. All of this makes launch windows very attractive. Unfortunately, this fertile period manifests more often as a willingness to shove half-baked products out the door to be day-and-date with a console launch than it does as an opportunity to launch a new IP in the best possible light.
So, maybe there is something to Pereira’s conclusion, and perhaps Guillemot and other major publishers are being myopic in their due diligence. Perhaps the question is not so much when to launch a new IP, but how: how can the ideas with the greatest potential be identified and cultivated, how can development teams work efficiently and responsibly to launch the strongest possible game while maintaining a reasonable budget and scope, and how can publishers make sure that new games get their fair day in the court of public opinion? Perhaps publishers should be focusing more on quality and efficiency than on timing.
*If the project failed, it would likely fail before the entirety of the $80 million budget was spent, but I am simplifying the example by calling it a total loss. This is not entirely unreasonable, as, if the remaining money was not spent by the end of the fiscal year, it would be “lost” in the strange financial reset that happens at the start of each new fiscal year.
†In assembling this data, I have assumed that:
The success rate of new IP on the Xbox 360 is comparable to the success rate of new IP across hardware platforms
The unit sales of a game on the 360 are indicative of the overall sales
That US sales are indicative of global sales
A reasonable measure of success for a new IP is whether a sequel gets green lit
The score given to the game by 1up.com is a reasonable indicator of the game’s quality
The estimated unit sales as listed on vgchartz.com is sufficiently accurate for the purposes of this example
If a game released before March of 2011 does not have an announced sequel, it will not get one