The Problem With NPD Numbers and the Incomplete Picture They Draw
The NPD Group’s monthly games industry reports continue to be widely covered in the media, and it’s easy to understand why. With game publishers hesitant to make a habit of sharing sales figures, the NPDs, as they’re often called, supposedly give us somewhat of a glimpse as to which games are succeeding and which are failing. Criticisms of the NPD Group’s numbers have becoming increasingly common over time, to the point where the question is now being asked whether the media ought to ignore the numbers outright.
That’s the position taken in a new piece by the Penny Arcade Report, and it’s one I find myself agreeing with. Regular readers of 1UP’s news section may have noticed a distinct lack of coverage of the NPDs as of late, dating back to before the change in how news was handled. I personally don’t give a second thought to the numbers anymore — a far cry from the days when I would be highly anticipating them every month. Rather than having multiple pieces per month analyzing and dissecting whatever could be deduced from the numbers, the NPDs are something that rarely cross my mind at this point. And there are a variety of reasons for that.
The first thing contributing to this situation was the NPD Group cutting back on the amount on information it freely shared with the public. It is a market research company, first and foremost, and it makes its money selling its extensive reports to, in the case of the videogame industry, retailers, videogame publishers, and console manufacturers. The data it shared with the press starting in 2006 was only a cursory glance at what it had to offer, but it was a tremendously helpful resource for providing additional insight.
In late 2007, it announced plans to reduce what it distributed. It wasn’t until three years later, in 2010, that it actually implemented this change, stripping out the hardware and software numbers from its monthly reports. A year later, it followed up by putting an end to analysts sharing its data. Now, beyond a numberless-top-10 list (which combines platforms, hurting its usefulness) and a few numbers (including total sales and amounts spent on videogame hardware and software), it’s been left up to publishers to share any numbers with us. More often than not, they’re disinclined to do so. Even with the occasional bone we’re thrown, that mitigates this only so much.
Much more importantly, the face of the industry has changed dramatically. Anytime NPD numbers are covered, there is a major point that is often not highlighted strongly enough: These figures account only for sales of new, physical games sold at retail in the United States. While it’s easy enough to understand the numbers only cover what’s sold in the United States, the fact that they are retail-only means they can be awfully misleading.
One example of this is in the way PC games have always been underrepresented in the NPDs. The Penny Arcade Report points to the recent example of XCOM: Enemy Unknown; by all accounts, it’s an excellent game, but it didn’t make the NPD’s top 10 list for October (the month it launched in) and sold an unspectacular-by-today’s-standards 114,000 units. PC is the platform you’d expect the game to do best on, and public Steam data would seem to suggest the PC version alone exceeded that mark. But just look at the NPD data, and you’d have no idea that was the case.
This is true of any game with a strong PC or PC-only presence, particularly those which are available through Steam. In the past few years I can think of only one or two PC games I’ve bought at retail, and are therefore accounted for by the NPD’s numbers (which, mind you, don’t actually get sales data from every retailer in the country). The dozens of digital PC games I’ve bought, most of which I’ve gotten through Steam, might as well not exist, as far as the NPDs are concerned. Steam accounts for a major portion of digital PC game sales, and no doubt of PC games in general. Particularly considering that Valve is a private company, it has no incentive to ever share sales figures with the NPD Group, or anyone at all outside of the developers and publishers it works with. Even if other digital storefronts, like Origin, were to share data with the NPD, it would still be a very incomplete data set without numbers from the goliath that is Steam.
Beyond the lack of comprehensive sales figures of your average PC game is the hole which represents things like free-to-play and subscription-based games. World of Warcraft may be past its peak, but it still commands a significant number of subscribers representing a huge amount of revenue, and yet the release of a new expansion is the only way the monthly NPD numbers will ever provide any indication that interest in WoW remains. Other huge hits, meanwhile, have no chance of their success being reflected by the NPDs, because they are free-to-play games. Whether it’s Team Fortress 2, a mobile game, League of Legends, or FarmVille, these games are nowhere to be seen in the NPD lists irrespective of how they compare to those in the NPD’s top 10, which are treated as the top moneymakers in any given month.
The major problem with the NPD numbers is not so much what they claim to be when you examine the fine print, but the way they are portrayed. I do believe the figures can have some value, but only with a great deal of context — and even then, that value is quickly eroding as the industry becomes more and more digital-centric. (When even Nintendo is providing downloadable versions of its games, you know it’s a significant movement.)
Despite this, NPD numbers continue to receive far more attention than they should based on all of these qualifications that have to be made about them. The problem, of course, is the lack of a superior alternative to turn to; if such a thing existed, NPD numbers would have been left behind long ago. But ditching the NPD numbers, or at least paying less attention to them, means having less data — even if that data is less-than-ideal, to say the least — with which to work. But it can be very misleading to operate under the assumption that the entire industry is struggling when NPD numbers can only show that part of it — new, retail sales — is experiencing any difficulties. So while we might have to deal with having less information, in my estimation it would be better to move on and acknowledge our ignorance rather than carry on as we are.