The AOL Model: Do You Need to Actually Sell the Client?

For a retail hybrid, selling the client is a given; that is the business model and there is no other choice at present. For web-based games, the model is the "free to the player" model and distribution is automatic. For subscription games such as PWs, whether or not to sell at retail becomes an interesting question. If the cost of goods to create a retail unit (typically somewhere between $2 and $6 per unit, depending on the size of the print run) and paying MDFs bother you, you need to ask yourself if you should sell the unit or give it away instead.

Consider: If you sell 100,000 units during the launch period, at an average return of $15 each (wholesale to the distributor minus cost of goods, development funding, and marketing expenses), you'll make $1.5 million. This is hardly an insignificant amount of revenue, right?

If you expect your PW to do even as well as Asheron's Call ( AC ), that $1.5 million is equal to 10% of your annual income. Each player who stays with you for 9 “12 months will be paying you the equivalent of three games at full retail price, or 7 “10 games at your average return per unit, depending on the amount of the monthly subscription fee. The predominant revenue source here is obviously the subscription fees, not the retail unit. In most cases, the revenue from retail sales won't even pay for development costs.

The real money is in the subscriptions, not in selling the unit at retail. The real objective should be distribution, not sales, to get the client in the hands of as many potential subscribers as possible. Obviously, not all publishers can afford to bypass the retail revenue, but alternative distribution means should be considered .

And those alternative methods exist. Consider the AOL distribution model, developed by AOL Marketing VP Jan Brandt back in the early 1990s. Ms. Brandt decided that the best way to increase subscription numbers for AOL was to literally put an AOL client disc in every home in America ”not once, but several times and through several means. Remember: This was a time when everyone else was marketing almost strictly online, and maybe including a client disc in one or two magazines per year. She took that an order of magnitude higher and mailed them to US homes several times over a period of years , put the disc as cover mounts on every relevant magazine she could, bundled the client with computers, printers, peripherals, and other software, and put up point-of-sale (POS) displays in grocery stores, software stores, and every other brick-and-mortar retail shop that would hold still for it; in other words, it was impossible for you not to have access to an AOL disc, whether you wanted one or not.

It was an expensive campaign, to put it mildly. By most reports , AOL went $1 billion in debt paying for the campaign over five or six years. And it worked. By the time Ms. Brandt was through, there were no viable US competitors to AOL. It remains the top Internet service provider (ISP) in the world with 34 million subscribers as of March 2002. Compare that with Microsoft Network's estimated 4 million or Earthlink's estimated 2.5 million.

You don't have to go $1 billion in debt to distribute your game client. Alternative means exist, but they require highly organized advance planning and a set-in-stone launch date. If you plan to give away the client, you have a number of viable options with which to create a multi-pronged distribution strategy.

Low-Cost Retail POS

In November 2001, NCSoft pioneered a new concept for distributing online-only game clients by selling them as low-cost POS units at Electronics Boutique (EB). The cost of each unit is $2.99, and one assumes that EB gets to keep all or most of that money. The POS version is simply a disc- sized package in a POS display. This no doubt costs NCSoft some money to create the package, duplicate the discs (probably in the $2-per-unit range), and get them to EB retail outlets, but it is an effective way to get in front of the eyeballs of thousands of potential subscribers each day.

Magazine Cover Mounts

Game magazines such as Computer Gaming World , PC Gamer , and Computer Games not only mount CDs on their covers, but they also sell access to them. Prices range from $7,500 “$15,000, but the magazine takes care of the disc duplication and, if you "own" that month's disc, your art goes on the disc and the disc holder. This is a cost-effective way to reach 125,000 “400,000 hard- core gamers. Remember, however, that there is usually a minimum 90-day lead time on reserving these.

Original Equipment Manufacturer Bundling

Original equipment manufacturers (OEMs) love to bundle software with their hardware as a value-added attraction for potential customers. This runs the full range of hardware, including joysticks, video cards, and on the hard drives of new computers. Some OEMs do huge buys, in the four- to eight-million-unit range. Traditionally, they pay small fees per unit for Class AAA products and take demo versions of only a few very popular products. If you're willing to take no fee or an at-cost fee for duplicating and shipping the discs to them, however, many OEMs would love to talk to you. Bear in mind that the lead time on OEM bundling is quite long, as much as six to nine months.

Downloads

If you're going to offer the client for free, there is no reason not to have a downloadable version available on your web site. You do need to understand that if your client is as large in size as the average online game client ”that is, a full CD-ROM of 500 “600MB compressed ”the bandwidth needed to download it will cost you somewhere between 50 cents and $1.50 per download. With the cost of goods for a shelf unit now running between $1.50 and $3.00 for most games, this is still significant cost savings, but it is a hard cost that needs to be figured into the equation.

You also need to take into account that you must have registration codes, even for free downloadable clients, or you'll find that the 1% of people in any group who tend to abuse any process will start creating multiple accounts simply because they can, and this can become a significant cost.

Free downloads in general create their own set of problems. For example, how do you limit anonymous downloaders to one download and one account to keep them from just creating new accounts and playing for free on the back of your bandwidth costs? The only way to curb most of this is to insist they not be anonymous and provide a credit card number for registration and issue CD codes, but that also provides a block to entry; many people are still leery of giving away a credit card number in general, and even more are hesitant to do it for something they aren't actually purchasing. Online "pocketbook" initiatives such as Microsoft's .NET and Passport may provide some relief in the future, but only if they become widely accepted.

And that is only one of the problems presented by free downloads today. Like the issue of the commercialization of the Internet in the mid-1990s, companies are rushing in without thinking through all the potential problems.