Designing the User Experience Curve | October 30, 2007

I’ve been interested in how the lessons learned from game design can be used to improve online experiences for a while now. I guess this interest started when I started learning about the concept of a “flow states”.

Flow is the state of being where you lose all perception of time and you flow from one successful task to another with seeming ease. It’s great if you can get into this state at work as you feel “in the zone” and can get a lot done in a short space of time. Sadly the number of distraction in the modern work place, combined with the fact that we’re perpetual multi-taskers, makes entering into the flow state at work a rare occurrence.

The place where we’re most likely to experience flow is when we’re playing a new computer game. As you start playing a new game you start encountering a multitude of small challenges and rewards. The high from each reward spurs you on the the next challenge, creating a cyclic effect.

If each challenge was as simple as the last you’d soon get bored, so computer games create a user experience curve. As the game progresses, challenges get incrementally harder, as do the rewards. However if the challenges get too hard too quickly, people give up. So the skill is in getting the curve just right.

Creating the perfect game curve used to be the preserve of level designers and was more art than science. However as game development budgets start outpacing those of the average Hollywood movie, the market needs to expand. Blockbuster titles are no longer the preserve of the hardcore gamer, demonstrating their l33t fragging skills. Instead these games need to appeal to a much broader audience. People who want to pick up and play, without having to learn overly complicated controls.

As an increasing number of these titles are sequels, the game designers also want to make sure that people can complete the games and are primed for a sequel. Because of this, games have become less about skills and more about creating an experience, a narrative and a sense of momentum. It would be no good if your newbie gamer gets stuck on level one and gives up.

I bought a copy of Wired magazine recently, and was fascinated by an article on the user experience design of Halo 3. The developers set up a fully featured testing lab and recorded more than 3,000 hours of play from 600 gamers. Normally game testing is restricted to bug testing, but what Bungie did was usability and user experience testing in it’s purest form.

Though usability testing the developers were able to pinpoint areas of the game where inordinate numbers of players were getting lost of killed. By examining real user interaction they were able to figure out what was going wrong and come up with ways to smooth out the user experience. This involved everything from making ammo more obvious, through to channelling users in the desired direction by stopping them from going backwards.

While this could be seen as an overly prescriptive way of creating games, it’s the essence of good user experience design. By removing usability barriers and helping people achieve their goals in an enjoyable manner, you end up crafting the optimal experience.

I was really excited by this article, and think there are a lot of lessons we can learn here. Not least the importance of usability testing on the user experience. At Clearleft we always budget for at least one round of testing, and even the smallest test produces amazing results. However imagine the improvements we could make if customers really started to value the importance of usability testing and budget accordingly.

Posted at October 30, 2007 5:54 PM


Kars said on October 30, 2007 8:37 PM

I enjoyed that WIRED article as well. It’s a great example of what can be done with a heavily instrumented game and plenty of time & money. I think this approach works well with games in a well-developed genre, like the FPS, since the kind of fun you’re measuring is well-understood.

Daniel Cook recently proposed a formal way of modeling and measuring the fun a game has to offer using so-called skill chains. Highly recommended. His model can be used to create a dashboard of some sorts that gives you insight into wether a game is actually delivering a fun experience as it is being played.

On a different note, and without wanting to be too nit-picky, there is a difference between the flow state and the fun a game offers. For one: Games can induce a flow state, but for a player to have fun it isn’t required. Also, the flow state is not unique to games. I know I’ve had experiences of flow while practicing martial arts, but also while wireframing. Koster puts it this way:

“flow relate[s] to exercising mastery, not learning”

Good to see you’re looking to games for inspiration!

DannyT said on October 30, 2007 11:05 PM

Great post Andy, however I’d be interested in any ideas you came up for applying such a “usability curve” to online experiences. The concept is great, what I’m thinking is where to draw the similarities and differences between the goals of a game and that of an online experience (other than online games of course).

Pebsville said on October 31, 2007 2:16 PM

This type of usability testing is nothing new, it is common practice when designing products. The game experience in this case is viewed as totality, the journey is the product. Products go through many iterations, usability tests, market tests, redeisgns, demographic variational designs etc etc.. So ok, the game was tested for areas of ‘fall-out’ or stickiness, what can we take from this insight? The creators ‘thought’ about what they were doing by testing it on a number of people who would inform on its use…hmmmm that’s a leap forward in innovation.

Mike Stenhouse said on November 1, 2007 1:58 PM

Read Raph Koster’s A Theory of Fun, if you haven’t already. He describes flow state as being when the challenge of the lessons being taught by the game and the pace with which they’re being fed match perfectly. You can get through the book in an afternoon… I didn’t think that much of it immediately but it’s actually given me quite a different way of looking at problems. Interesting!

Rob said on November 5, 2007 8:10 AM

Shame Bungie didn’t put any of that testing onto the user interface! I spent about 10 minutes trying to find out how to change my armor, and it took me ages to discover how to play with friends in match making.
Great to know testing is becoming less of a luxury and more normal.

Ben Hayes said on November 6, 2007 4:34 PM


Web apps have their own ‘curve’ of course, as people become more experienced users and uncover more features. In a similar way it seems a good idea if we let a novice user accomplish a basic task, but design little pointers that lead him on to extra, more sophisticated features over time.

Matt McVickar said on December 2, 2007 12:02 PM

I find it very interesting that the idea of a “challenge” is a design consideration itself — something that is the result of achieving the perfect balance between frustration and boredom.

After the first two paragraphs I was expecting you to put the idea of game design to use in attempting to create the “flow state” while working. Perhaps it’s impossible or at least a bad idea (the loss of productivity, the unpredictable nature of work), but wouldn’t it be neat to structure one’s workday (-week, -month, career, even?) around the sort of curve you mentioned in the article? It may not always work, but if you tried to plan your scheduled tasks and brainstorming to create an increasing level of challenge and output and return, it may encourage flow.