The Evolution of Distribution

Introversion Software was set up in 2001 by three university students with nothing more than £600 and a strong belief that word of mouth alone could sell their first game.

That game was Uplink, a hacking sim which went on to become a cult classic. It was followed three years later by Darwinia - a project which, as financial director Tom Arundel explains in this guest editorial, almost left the company in ruins.

In September of this year, Introversion released its third title - Defcon - offering the game via both download and mail order. Here, Arundel explains the benefits and drawbacks of Introversion's retail model, and looks back over the lessons the company has learned.

Introversion is a small independent development studio based in the UK. By small, I mean seven full-time staff, with a couple of freelancers. We strive to release a completely new and unique game every 18 months although that hasn't always worked out - Darwinia was a three year development folly that nearly ruined us.

But with five years now behind us, we're getting a grip on managing that, as well as understanding how our business model compares to others within our industry.

Actually, thinking about it, we've come quite a long way. Uplink, our first game, was written by Chris at university whilst Mark and I were otherwise engaged in trying to catch the attention of the only two girls at our engineering university. When we finally got to see Chris's game we realised it was good enough to sell; we gave up university beer for a couple of weeks, threw in £200 each and set up a store online.

Perhaps surprisingly, five years on, we still own all the IP to all our games, and we've managed to gather world-wide critical acclaim, a couple of awards and an entire smoking room's worth of wacky anecdotes along the way.

Meanwhile (and skipping over the fact that Chris and I had to sign onto benefits for 10 months, move back in with Mum and Dad, and sell everything we've ever owned on eBay whilst Darwinia slipped and slipped), we've been really quite commercially successful too.

Despite reckless attempts to justify to our accountants that our newfound riches would be best spent on company Maseratis - or at the very least a second hand Scorpion tank - they are nonetheless gushing every time we meet about the strengths of our model. That is, an absolute and stubborn refusal to give up the direct relationship we have with our customers.

Indeed, it's a principle that seems intoxicating enough to light up the eyes of fellow developers when we talk of 95 per cent gross margins, 20k strong communities and projects that give a five-fold return on sunken development costs... In the first year. Okay, it's not millions, but then we've only got seven mouths to feed.

Right now what's making this possible is the ease with which we can distribute a 30MB demo, receive wide exposure and huge demo download figures (due to the enthusiastic reporting and word-of-mouth phenomenon more usually associated with our off-the-wall type games) and then convert those players into fully paid up customers.

It's certainly easier now and perhaps this is why digital distribution is so fashionable as a conference topic at the moment. However, what rarely gets reported is the responsibility and headaches that occur when a lucky team of seven has to deal with 15,000 new customers per month.

In all walks of life, newer generations reinvent the wheel whilst trying to claim genius. What we're doing is really no different, in essence, to what we would have done 20 years ago, had we been writing games instead of watching Postman Pat in Spider-Man pyjamas.

Back then, small teams working out of their bedrooms were knocking out games in a week and selling them via mail order to the burgeoning group of home computer owners.

Of course larger corporations had already mapped the trend and started to develop a retail distribution network. As the install base of personal computers grew and with it the potential volume for mass-market consumer software, like games, retail became the dominant distribution channel and one where most gamers expected and went to buy games.

Mail order declined and it wasn't until 10 years later that Shareware re-kindled developers' long held desire to cut out the middle men, and more importantly content filters (i.e. buyers who controlled the retail channel) whilst enabling them to put out truly innovative games.

The popularity of Shareware grew with the emergence of cheap modems, BBS, and the ability to distribute the game on floppy discs which could be quickly copied amongst friends. The problem with Shareware was that it relied heavily on honesty, and the distribution mechanisms that favoured the initial distribution of the trial version were even more favourable to the unwanted distribution of the full version.

When we launched Uplink in October 2001, selling physical copies on the internet and distributing them via mail order made a lot of sense. We'd get 90 per cent of the revenue back after card processing charges and shipping, and we could be trading within a couple of weeks.

Uplink itself was only 15MB, but to try and prevent piracy we bulked up the game to 60MB, by including a reversed MP3 of Chris cranking out some riffs on his Strat in one of the data files. Since Broadband penetration was relatively low at the time, we could harness the vastly cheaper bandwidth of a moving truck and keep illegal downloads to a minimum.

Not sure it worked as well as we had hoped - when we released the first patch we'd sold around 10,000 copies. Within a week, the patch download counter was at 89,000!

But what about customer expectations? Actually, because of the unique gaming experience and niche content of Uplink, Darwinia and to a lesser extent DEFCON, people were happy to wait a couple of days, or indeed weeks to get their copies via the post.

We did manage to release Uplink into UK retail, but in contrast to the two weeks it took to get Uplink for sale on our website it took us four months and a leaked preview of the 80 per cent PC Gamer review to persuade retail buyers to stock even 20 copies of Uplink in their flagship stores.

That said, Uplink did well in retail and it was worth while commercially releasing it this way. Darwinia did less well in retail; we made the grave mistake of mis-pricing it - in this case too high, at £30 and even in the three years since Uplink, the retail space had changed to the extent that PC shelf-space for anything else other than AAA titles was almost non-existent.

Even so, we still believe that retail presents a very worthwhile opportunity even for small companies like us. With a modest commitment to in-store marketing we can expect a good return, though in the future we'll be signing the rights to a third party who has a specialisation in this area. Managing retail relationships is a lot harder than managing end users.

Which brings us nicely up to present day, where we're making the most of high broadband penetration and an older demographic (certainly in our area of retro-looking strategy games) who can all use credit cards.

As a small studio the benefits are large. We can set our price point and change it quickly if necessary. We can spend 12 months developing a product and market test it in the real world to see if the IP is worth developing further, and we can access cutting edge console hardware and customers without cashflow crippling licensing fees - and still get back most of the revenue pie.

More importantly, we can revisit the Shareware model and take the benefits without the risk. Namely, its propensity to get people talking about and trying out the game, but also an ability to control piracy, particularly in the case of multiplayer games, by using keys checked against a game matching server.

We can also get instant and cheap access to millions of hardcore gamers via services like Steam and still get the lion's share of the revenue back. We can track customer behaviour - where they heard about us, what areas of our website are most important to the purchase decision, and the conditions in which people most effectively persuade their friends to download and buy the game in order to edge conversions up.

Of course, we also get the benefit of impulse purchases, as people pay and download the key in a fit of post-demo enthusiasm.

Surely that bloke from the Fast Show should drop in shouting, âAin't digital distribution brilliant, right?'. Well, 72.353 per cent right. Tying oneself up to some gaming portals could leave you with as little as a 30 per cent cut, and you may even have to give up your right to distribute the game by yourself.

If you develop your own key / distribution system, you have to maintain it and that alone is a full time job. What about the 250 emails a day from customers asking for misplaced keys?

Also, if you license the distribution rights to someone else, you have little control on the price. That's particularly true if you're receiving royalties as a percentage of sales; if you sign up to lots of smaller portals, you could wind up sparking a price war - with you as the biggest loser.

Even so, digital distribution is helping level the playing field towards the smaller guys like us. Admittedly it's of less help if you're an incumbent with 50 mouths to feed and a burn rate of 250k per month, but in that case you're revenue model is most likely to be work for hire.

But don't forget other distribution methods; there's still good money to be made in retail and if you can distribute the game via cover discs or bundled with OEM hardware then for those sales, all you actually have to do is distribute a 50 byte key.

At Introversion we would argue that digital distribution is an enabler, allowing us to follow our true development panacea - that of being truly independent, allowing us to develop, create, license and sell our own original IP. But the realities are quite a bit different to what most small teams we've met consider ideal. After all, out of the seven full time members of our company, only two have anything to do with actually making games.