Why GameStop failed

Specialization was the whole advantage. Before e-commerce, buying a game meant Best Buy, Circuit City, Walmart, Toys R Us, Blockbuster or a local shop — and at every one of them games were one aisle. GameStop carried the newest titles, the consoles, the largest used collection, a trade-in counter and a direct line into the industry.

The scale that produced was absurd. Over 3,500 US locations by the mid-2000s and 4,400 by the early 2010s — four times Abercrombie and five times Toys R Us and Best Buy, and about half as many stores as Starbucks had in America. Overseas it went from 866 stores in 2006 to 2,177 by 2012.

It scaled because the box was tiny. The average store was roughly 1,500 square feet — smaller than a tennis court, smaller than a Starbucks, about the size of a Subway — needing four walls, a roof, air conditioning and electricity, so it fitted cheaply into strip malls and shopping centers anywhere in the world.

The trade-in program was a known ripoff and used anyway. GameStop offered $3 to $5 in credit or cash for a game and put the same disc back on the shelf at $15 to $30 the next day. Driving across town for a better offer was not worth the trouble, eBay and Craigslist meant handling shipping and payment on every title, and no big-box retailer thought pre-owned was worth the hassle.

The pre-owned disc was a perpetuity. No matter how many owners a title passed through, as long as the disc worked there was another margin to take on every exchange. New titles averaged $41 and pre-owned $18 — on games bought back for $5.

The business monetized both ends of the adoption curve. Sell the newest hardware to early adopters, repossess the previous generation through trade-in at rock-bottom prices, then resell that same hardware and software at a markup to the budget-conscious majority arriving late.

It was a promoter as well as a retailer. GameStop owned Game Informer, a subscription magazine that was one of the few places to find gaming coverage in the print era, and sold ad space in it and in its stores to publishers, studios and manufacturers. Its reach made it a partner every publisher wanted to stay on the right side of — worth higher allocations of new consoles and titles, plus price protections and return privileges other retailers did not get.

The golden era did the rest. Revenue rose 200% from $3 billion to $9.5 billion in six years as gaming went mainstream, and the average store went from $688,000 a year in 2006 to $1.44 million by 2012 — a 110% increase per location.

But retailing is a thin business at every layer. New consoles were a $1 billion line by 2010 at 6% gross margin, because everyone from Nintendo to GameStop wants the hardware barrier as low as possible. New software averaged 41% of revenue at 21% gross margin — of a $60 game, about $15 reached the retailer. Only pre-owned cleared 47%, and it was what pulled company gross margin to 27-28% and operating margin to 6-8%.

Three shifts ended it, and none of them were about GameStop. Broadband replaced physical distribution with digital delivery. The iPhone and the App Store absorbed casual handheld gaming and handed publishers the blueprint for a storefront of their own. And single-player titles gave way to multiplayer-first games with longer lifespans, post-launch DLC monetization, and DRM that made each copy non-transferable — which is what actually killed resale.

GameStop's response inside gaming was as good as it could be. It sold physical prepaid Xbox Live and PSN cards and DLC codes for customers paying in cash or store credit, and claimed to be *"the only significant brick and mortar retail seller of DLC."* Filling cracks in someone else's storefront is not a strategy.

The diversification is where the real damage was done. Every business it bought to reduce reliance on gaming was an equally fragile middleman facing the same consolidation: Kongregate in 2010, Spawn Labs in 2011, Simply Mac for $10M in 2012, Spring Mobile for $60M in 2013, ThinkGeek for $140M in 2015, and Cellular World and Red Sky Wireless for $400M in cash in 2016.

The Kongregate purchase was indefensible at the time, not just in hindsight. Two months before the deal, Steve Jobs published an open letter saying Flash would never be supported on iOS. GameStop bought a desktop Flash game portal anyway; one year later Adobe put Flash on maintenance and killed it on Android, taking browser games with it.

Spawn Labs is the other one that stings. GameStop bought a game-streaming startup in 2011, the same year Twitch was founded, abandoned it in 2014 with a $20 million write-off citing *"lack of consumer demand for video game streaming services"* — the same year Amazon bought Twitch for nearly a billion dollars in cash.

It worked, and it did not matter. Simply Mac and Spring Mobile grew from $300M to $800M combined and from 200 to over 1,300 locations in four years, and the stated goal was that 50% or more of operating earnings would come from outside gaming. Wireless peaked at 9% of revenue. By 2019 Simply Mac, Kongregate, Cricket Wireless and the entire AT&T division were sold, roughly at one times annual revenue, and the company was on its fifth CEO in two years.

The present bet is the same mistake in a new medium. Revenue is back to gaming alone at $5-6 billion, the average store takes about $1.3 million — less than a decade earlier before adjusting for inflation — and reporting has been collapsed into software, hardware and collectibles so the NFT marketplace cannot be evaluated separately. That marketplace competes with hundreds of exchanges where Blur overtook OpenSea by cutting fees to zero. Software commoditizes exactly like retail does.