Deciding The Fate of Dante and 'Phoenix' -- How Capcom Predicts Game Sales

Phoenix WrightRemember how difficult it was to find the first "Phoenix Wright: Ace Attorney"? Contrary to message board conspiracy theories, that wasn't actually Capcom's fault.

When I tried (and failed) to buy "Tetris DS," it made me wonder. How does a publisher predict demand? How do they figure out how many copies to manufacture and ship, and when (and why) do they halt production lines?

I've been speaking to different companies to figure that out. One of them was Capcom, a company with strong Japanese roots and a large Western audience. They, along with the rest of the industry, wrestle with this every day.

Christian Svensson, Capcom's VP of business development and strategic panning, admits it's extremely complicated. Not only does Capcom have to convince retail to order en masse at launch, but the sales cycle itself is changing as the industry grows bigger and bigger.

"The sales life cycle of a product is shrinking," said Svensson. If a game undersells its first month at retail, it's finished at retail. Often, though, it's quicker: if a game underperforms its first week, it is more than likely "dead on arrival."

Here's how Capcom tries to prevent being in that position.

Devil May Cry 4The original "Phoenix Wright" proved problematic for Capcom. Despite rabid anticipation among fans for its English release, retail didn't care. No one besides GameStop ordered copies at launch. Wal-Mart, Target and Toys 'R Us all passed.

Despite this, the game performed well. Capcom went through 9 or 10 production runs of 3-4,000 units before Toys 'R Us took notice and asked for 15,000 copies out of nowhere. Of course, that didn't stop fans from complaining that the game was impossible to find.

Capcom didn't really have a choice. They weren't creating artificial demand. "We have to be more conservative because, believe it or not, these games cost money," said Svensson.

On the Nintendo DS, for example, the challenge is significant. It takes 50 to 75 days for a production order on a DS title to reach the US, longer during the holidays. Capcom must anticipate demand in the first week, first month and beyond. If a game is successful and supply dries, it could be two months before another copy shows up.

Long before the big retailers passed on "Phoenix Wright," however, Capcom had -- right or wrong -- internally predicted its sales cycle. That's the case with every Capcom title. Before "Devil May Cry 4" was green lit for production, Capcom had already assigned it sales projections. Years before the title would even be announced, Capom had a number.

Each of Capcom's titles has sales projections before development. At this early stage, they're taking into account a variety of factors: market preference, if it's a brand (i.e. sequel), the outlook of hardware platforms, the game's desired feature set and more. Whatever the case, it gets a number. Sometimes it's broad; most times it's very specific.

Capcom does the math and projects the game's budget and potential revenue to determine how much profit they hope the game will produce. If it clicks, the game's green lit.


That sales number is going to change, though, as development moves forward. Take "Dead Rising," widely considered a surprise success for the company. Six months prior to launch, Capcom wasn't sure what to make of it. They knew "Dead Rising" would sell simply because it was their Xbox 360 debut, but was it any good?

For a while, admits Svensson, they honestly weren't really sure. It wasn't until E3 2006 that Capcom saw the game finally start to come together. As a result, they threw their old sales projections out the window. That sales forecast continued to change before release.

Okami"Dead Rising" and "Phoenix Wright," however, are smaller titles compared to some of Capcom's bigger franchise releases, but Svennson said the company is even more cautious when finalizing orders for a sure-fire blockbuster. Being accurate on big releases is "that much more critical because you are producing so much more." Being off by 10% on "Okami" is much different than missing the mark on "Resident Evil 4."

Every time Capcom orders a production run, there's a risk those games will either never be ordered by retail or arrive at retail and stall on shelves, forcing price reductions.

Capcom keeps a warehouse of additional units in case a game has a surprisingly strong launch or sudden surge in sales. They have extra "Devil May Cry 4" inventory just sitting around in case a retailer calls up and says they need another 30-40,000 units.

If they don't have those units immediately available, the retailer may just cancel the order entirely and move onto the next big game. "Demand may not be there [anymore] if a request goes a week ignored," said Svensson.

As for what happens to extra copies of games that never make it to retail...well, ever seen the weird three-for-one bundles that appear at Sam's Club and Costco? Bingo. Extra inventory is eventually passed on to liquidators, whose job is to get rid of the excess inventory, even if it's not through the usual channels or pricing scheme.

And that, readers, is how Capcom releases a video game. Tomorrow, I'll look at how Atlus, a publisher dealing exclusively in the market of niche games, does the same. I feel like I've just solved a case in "Phoenix Wright."

Have a hot tip? Is there a topic that Multiplayer should be covering and isn't? Maybe you have a "missing game" story to share. Drop me an e-mail.