After a long blogging break, I’m back with a short post. The Virtual Goods Summit 2008, including the conversations I had leading up to the event, gave me a lot to think about. So what’s on my mind these days? Three major things related to free-to-play games and virtual goods:
Free-to-play gaming today looks is a lot like the “freemium” web model – Today’s free-to-play gaming models look a lot like the classic freemium model on the web. In both cases you generally have a relatively small percentage of relatively enthusiastic users who pay and a larger percentage of uses who do not pay to consume your service. This is not a complete analogy – freemium tends to sort people based on their willingness to pay for additional features while free-to-play gaming tends to sort people based on their own personal tradeoffs between time and money or their willignness to invest hard dollars to get access to powers, levels, or portions of the game not easily accessible to non-paying players.
Merchandising virtual goods is still hard and not well understood – Figuring out how to merchandise virtual goods is still in its infancy, even for the most sophisticated players in this space. I mean this in the most complete expansive sense of merchandising possible – everything from what should be offered to players (product assortment or mix), the period of time for which such products should be made available to users (shelf life), packaging, pricing, bundling, and art of the upsell are all still, by and large, in their infancy for the space. There are a handful of people who excel at one or two parts of the merchandising formula described above, but I don’t think anyone has yet tied it together in an elegant way. There is much money to be had by figuring this part out and I suspect traditional retail models can teach us all a thing or two about how to raise the bar here.
What do you do about the low ARPU trap and the meaningless average? – In the free-to-play world, there are quite a few games where you end up with a small slice of users who spend and a large number of users who don’t spend, you end up with what looks like a bimodal average and a virtually meaningless average – few users end up spend anything near the average. People are either spenders or non-spenders in many (but not all) games. So what is a game developer to do in this case? If your ARPU is $1, should you try to acquire users at $0.75? Well, as Daniel James and Andrew Chen mentioned in their presentation at the Virtual Goods Summit 2008, you sometimes get what you pay for – very low-cost traffic might end up turning into very low ARPU traffic. If your ARPU is low, is there an inherent danger in trying to find traffic that makes economic sense at a low ARPU level? I don’t have the answer here, but I do think it can be a dangerous downward spiral to chase.
Actually, I’ll toss in a fourth thing and it’s really related to the first point. At the end of the day, so long as free-to-play gaming looks a bit like freemium, there is much to be learned from classic subscription games. Users who pay in the free-to-pay context can pay, in some cases, amounts on par with what a subscriber would pay. Heck, there are quite a few free-to-play games who make quite a bit of money from subscribers or people who behave like subscribers (they spend a fairly consistent amount of money from month to month without it being called a subscription). The advantage of free-to-play games, though, is that you don’t have the subscription wall as a binary choice for users – instead of forcing users to declare their willingness to pay up front at the time of registration (or at the end of the trial period), you can allow them to reveal it over time based on their engagement and willingness to spend. If free-to-play games can get better at converting browsers to buyers, there is a really large opportunity to be had and one that will be problematic for some pursuing a purely subscription model.