Game sales data is going the wrong direction. It’s not going down or up; it’s disappearing altogether. We’re losing the information we need to understand the industry’s performance, at any level, and it’s a dangerous, unacceptable trend.
I’m far from the first to note this. Accomplished journalists like Ben Kuchera and Brendan Sinclair have written great articles recently (read them!), and analysts like Matt Matthews have been dissecting sales data and questioning its quality for years. I’m just adding another voice to the rising chorus.
Movies have box office returns. Music has the Billboard charts. Books have the New York Times bestseller lists. They not may be perfect, but publishers and creators have come to depend on them for independent, consistent, good enough data that they rely on to make business decisions.
For a long time, games have had NPD. NPD collects retail sales data for game hardware and software and estimates what they can’t collect. They charge for their results, releasing only a few highlights publicly, and their coverage isn’t perfect, but for a while they were good enough.
That didn’t last. First, holes in NPD’s coverage have become more and more glaring. They’ve only just started to measure countries outside the US, used games, and Walmart. All three have grown hugely – used games alone are a multi-billion dollar market, and Walmart contributes 25-30% of US game sales – and the consensus is that NPD’s initial efforts may be too little, too late.
Second, they drastically curtailed the data they release publicly a couple years ago. Low-barrier-to-entry platforms like web, mobile, and digital have been feeding an explosion of indie developers who can’t afford to buy expensive analyst reports, and this hit them below the belt.
Finally, and most importantly, digital is the future, but digital sales data is non-existent. There are exceptions here and there, but none of the large online stores – Steam, XBox Live, PSN, iTunes, Google Play – report actual sales numbers. Rankings, sure, but not numbers. NPD claims they’re estimating digital, but if the platform holders continue to withhold their data, there’s not much NPD can do.
What’s worse, as bad as NPD is right now, it doesn’t have much competition. Firms like EEDAR and Screen Digest do some data collection and performance measurement, but their focus is analysis and research, and they’re up against the same digital blackout as NPD. It’s increasingly falling to major publishers themselves to report numbers, which some do, and to journalists who dumpster-dive SEC filings to fill in the rest. Neither of those is a sustainable solution for the long term.
It’s not just game industry journalists who decry this. People on the business side are equally blunt. Frank Gibeau, head of publishing at EA, recently said “[NPD] is totally irrelevant. We don’t even really look at it internally anymore.” Stardock CEO Brad Wardell described the current state as “very destructive.” Outspoken analyst Michael Pachter went even further regarding the lack of digital data: “Oh my god, that’s horrible. There is none. We have no idea what anything does.”
So, what do we do? We lean harder and harder on NPD, but they’re ineffectual at best. We demand that the digital platform holders publish data, but they don’t. We encourage the publishers to release more of their own numbers, but they’ll never make up the difference. We look to alternative methods and third parties like Google to build new market research tools and publish their own research, which are great, but will never replace actual sales data.
For those of you who work in the business, what do you think? Give me some hope!
5 thoughts on “The slow death of game industry sales data”
cc Sam Z. Glassenberg , Michael Mandel via Facebook
SuperData Research is doing compelling new work to fill in this gap. They’re analysts doing traditional market research; in their words:
The teasers linked from their blog look solid. Definitely worth watching!
another recent development: Valve stopped exposing numbers by default, which shut down third party tracking group SteamSpy. whee. ugh.
I asked Michael Pachter to consider answering this on his next episode of Pachter Factor:
he replied! episode 170, just today. he agreed that specific sales numbers from these services are generally all unreliable these days. evidently he hears a lot of data firsthand from publishers and developers themselves, which he can use to extrapolate other numbers more reliably, based on historical physical to digital percentages and breakdowns by region. still, he said he only still pays for NPD, and only because he can supplement it.
he also said mobile sales data is even worse. rankings themselves are correct, but numbers from services like App Annie are all over the place because they’re all estimates and guesses. whee!