The Genesis of the Virtual Goods Model

Guest post by Matt Mihaly on how the virtual goods sales model was born.

Matt Mihaly is a CEO and Creative Director of Sparkplay Media and Chairman (formerly founder, CEO, and Creative Director at Iron Realms Entertainment.

The virtual goods/microtransactions model dominates Asia and is
now far and away the most popular business model for online games in
the West with the meteoric rise in popularity of games on Facebook
and MySpace. Twelve years ago, in ’97, I pioneered this model
at the company I had founded two years earlier – Iron Realms
Entertainment – out of pure necessity. This is the brief
history of how the virtual goods model, and soon thereafter the dual
currency model, came to be.

In 1995 when I started developing Iron Realms’ first MUD –
Achaea, Dreams of Divine Lands – my plan had been to charge for
the game the same way that most online games of the day did –
by the hour/minute. Bandwidth costs had historically been quite high
and it made sense to charge players based on their activity. In 1996,
however, AOL went from charging for each minute of use to charging a
flat, monthly rate for its services, including, at the scale of the
time, popular MUDs like Gemstone from developer Simutronics. This
certainly put the kibosh on my plans as once AOL went to unlimited
use consumers stopped accepting hourly models, at least in the US.

Achaea opened to the public in September 1997, and based on what
players were telling us there was no way we were going to be able to
charge a mandatory subscription and retain many of them. We just
didn’t have anything at that point to get a good deal of
players to take out a credit card and subscribe. No big IP, no money
for marketing, and frankly the game was pretty rough at launch. I
believed in a “get it out there and iterate” approach,
but there hadn’t been enough iteration/polish added to convince
players to pay a mandatory fee to play.

What I did instead was decide that I’d run Achaea for free,
which is the norm in the MUD world since 99%+ of MUDs are run as
hobbies, and sell a virtual currency called ‘credits’.
Players would use those credits to buy ‘lessons’ in-game,
which are used to acquire new abilities – everything from creating
voodoo dolls of other players to mastering the use of religious
shrines as combat weapons. Lessons were also gained by leveling up,
but it wasn’t possible to max out your potential without buying
credits. Interestingly, almost all of our abilities in Iron Realms
are focused entirely on PvP. The great majority don’t even work
on NPCs and involve, by mainstream MMO standards at least, some
pretty strange mechanics. Agoraphobia (panics you when indoors),
asthma (stops you smoking pipes, which are used to cure certain
ailments), epilepsy, implanting subliminal suggestions, secreting
dozens of poisons into your serpentine fangs, all the way to the
familiar swordplay, martial arts, magic, and so on.

Optimizing pricing

This model proved reasonably popular but we were constantly
getting requests from some of our most dedicated players, asking to
purchase virtual items or services beyond the lessons. For instance,
players were asking us to custom-code special houses for them with
one-of-a-kind features like a butler that would circulate serving
drinks, or to create pets with unique behaviors such as a koala that
would hang out in trees and drop down on those passing by, startling

I decided to try an experiment and held an in-game real-time
auction in which we sold a range of custom services/items like the
aforementioned dwelling. To my great surprise, players started
bidding up the items from $25 or so to hundreds of dollars. In the
space of an afternoon the auction brought in around $5,000 –
more than the game had made the previous month. I was the only
full-time person at that point so $5k went a long ways.

Thereafter, we held repeated auctions of custom items but that
quickly became untenable. Creating new functionality (in text MUDs,
cosmetic things are worth a lot less than they are in a graphical
environment generally speaking) becomes problematic on a
player-by-player basis even in MUDs with their very low populations
as compared to graphical MMOs. As a result, we stopped running
auctions and started selling the items within in-game shops for
credits, which is the model all the Iron Realms games maintain to
this day.

The range of items we sell is pretty large. Most are, frankly,
expensive by MMO standards, but it’s how we can afford to
operate and continuously add new content and systems to our games on
a tiny audience – less than 10,000 active players between all four
Iron Realms MUDs, though well over a million registrations….showing
the futility of using registrations as a meaningful metric. Achaea
sells, for instance, a Logosian Broadsword (better than all but the
very best player-crafted broadswords) for 1600 credits –
approximately $480. A Torc of Telepathy that gives you a couple of
offensive telepathic powers goes for 800 credits – about $250. A
magical Box of Chocolates goes for 200 credits – about $65 – and will
generate 18 or so types of chocolate gifts. A Cane of Quickening lets
you walk with a broken leg – breaking limbs is an important
combat mechanic in Achaea – and goes for about $100, while
increasing your friends limit by 5 costs around $8.

As far as the logic of pricing items goes, it’s not
particularly complicated though I’m sure that someone with
serious retail experience could price our virtual goods more
efficiently than we do. In essence, we use a rough calculation that
takes into account whether an item is mainly functional or mainly
cosmetic, and what the magnitude of its effect is. In the case of
cosmetic items, we look at how impressive the cosmetic effect is and
price accordingly. In the case of functional items we look at whether
the item is mainly about convenience or whether it increases a
player’s power over NPCs or, especially, other players. The
greater the effect on PvP, the more expensive the item is. Our games
are very PvP-focused and it’s crucial not to sell items that
will by themselves unbalance PvP to the point that having that item
means you just win a fight without having to employ any skill.

The dual currency model

Even with the introduction of the virtual items that players had
been clamoring for, there were still problems. At that time in 1998,
players couldn’t transfer credits to each other. There were
frequent requests to be able to gift credits, which we’d
occasionally assist with manually, but there was no way to transfer
credits to another player on your own. I can’t remember now why
I didn’t allow it. We hadn’t had serious fraud problems
yet at that point, and I can’t think of any other rationale
that would make sense.

What I did was create the first virtual currency exchange. This
allowed players to trade credits, which were purchased for real
money, for gold earned by playing the game in a sanctioned, in-game
environment. The gold price/credit fluctuated in real-time based
purely on player-driven supply and demand. It’s now referred to
as the ‘dual-currency model’. Overnight, Achaea’s
revenue almost doubled. Suddenly, the untapped demand of the
non-paying players was able to be satisfied via paying players.

In effect, everyone won. Player A who has money but not enough
time can purchase credits from Iron Realms, giving Iron Realms what
it needs – real money. Player A can put those credits on the
credit exchange at whatever price in gold she wants and wait until
someone buys them at that price, or can put them for sale at a price
slightly below the current market price and sell them almost
immediately. The player buying the credits (Player B) spends gold to
get the credits, as he has more time than money. Player A receives
gold – the product of time – which is what she wants, while
player B receives credits – the product of money spent with the game
operator – which is what he wants. Player A, Player B, and the
operator have all received what they want. It’s a win-win-win

Of course, it’s not as if all players feel this win-win-win
situation is how some players view it. We never really had issues
with it at Iron Realms because the business model self-selects for
players who like it, but from almost day 1 there was a lot of
criticism of the model. People claim it breaks the magic circle, that
it diminishes non-buyer’s achievements and that’s it just
feels unfair. I disagree with most of the complaints but that’s
not really relevant. The player perceives what the player perceives,
and it’s not very easy to change that perception. Just as the
subscription model selects for people who are willing to pay
subscriptions, so does the free-to-play w/ virtual goods model select
for people who like playing games for free. It shouldn’t be a
surprise to anyone that far, far more people are interested in
playing games for free than are in paying subscriptions, though that
doesn’t speak to the value of an individual player to a
publisher. At Iron Realms, we average more than $15/month/player but
in most free-to-play games, the average player is worth less revenue
than a player in a $15/month subscription model is.

Compared to the volume of early complaints over the virtual goods
model however, there was almost no dissent whatsoever over the
introduction of the dual-currency model. It was met with almost
universal praise from our players. This should be expected though,
because with the dual-currency model there are no losers –
everybody wins by getting what he/she/they want, from publisher to
paid currency seller to those who buy the paid currency with gold.

The single most important piece of advice I’d give about
running the virtual goods model is that selling near-guaranteed
victory is a bad idea, with how bad determined partially by who your
audience is. It works better in Asia (see ZT Online) than it does in
the US, and tends to work better with a less connected audience than
more connected, such as in the Mafia or Farmville-type games on

Similarly, if you’re going to use the dual-currency model
the most important thing you can do is ensure that you’ve got a
range of desirable things that can only be purchased by your
purchased currency and another range via your earned currency. You
need players to want both types of currency, else there’s no
reason for one side of the gold/credit exchange to trade.

That is, in a nutshell, the genesis of the virtual
goods/microtransaction model. Today, of course, there are many
companies that have employed the model to much greater effect than I
have and it’s great to see the big success stories out there
from Habbo to Tencent to the social games currently eating up
everyone’s free time on Facebook. I don’t think the
‘traditional’ (if something that’s only about 12
years old can be called traditional) subscription model is going to
disappear, but I do think it will increasingly be something
attractive only to very hardcore gamers, who are a distinct minority
of the worldwide gaming audience.

Matt Mihaly is a CEO and Creative Director of Sparkplay Media and Chairman (formerly founder, CEO, and Creative Director at Iron Realms Entertainment. He is known for pioneering the sale of virtual assets in online games.


One thought on “The Genesis of the Virtual Goods Model

  1. the most important thing you can do is ensure that you’ve got a range of desirable things

    You’ve said it.. the dual currency model and other models of organising the virtual economy are interesting, but the real question is what makes the items desirale.

    Nice article anyway.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s