Performance (P) vs cost (C)
Let’s go back to our equation. On one side, we have the added value of the system: its ability to help the operator reach a higher level of performance.
Performance plays out along several axes. It covers, for instance, what the system is expected to do: an email client is judged on whether it gives users a better grip on their inbox, or hooks up cleanly with other tools (instant messaging, and so on).
Candy Crush owes its success on smartphones to a high level of performance. The kind it’s after here is sheer entertainment value, and several things make that possible. The first, obviously, is the fun factor: a proven recipe whose appeal needs no demonstrating. The second is the social factor: by going all in on social networks, the game fosters interaction and competition with the people around you. It isn’t just about having fun, then, but about having fun together and against one another. A social hook done right.
The cost
On the other side comes the cost of running it: the learning curve, the friction in the experience, the need to double-check settings on screen, anything that takes effort to use, whether mental, physical or social. The higher that cost climbs, the more performance the product has to deliver in return, or users will go looking elsewhere.
Here again, Candy Crush is on home turf. The interactions are simple (swiping everywhere, the drag-and-drop gesture that smartphones made second nature). The game and competition mechanics ramp up gently and stay easy to grasp (a learning curve kept firmly in check). And on top of that, quick loading times and a free app. By leaning hard on how accessible the product is, the game keeps the cost of using it down to a minimum.
You can also buy yourself extra tries to clear a level, or pay for certain bonuses. Higher (financial) cost, higher performance: the balance holds.
In the end, it comes down to a balance between the performance you’re chasing and the cost of running it. The Puzzle Games segment is a saturated market, and the idea is hardly new. Yet pair a powerful social hook with a proven dose of fun and a very low cost of use, and the whole thing clicks.
The question of safety and risk (R)
A long way from a round of Candy Crush on the metro or the sofa, in high-risk environments you have to add serious safety concerns to the picture, and risk is the notion that completes the triad. For the operators, it becomes a matter of keeping the level of performance, the cost of running the system, and the risk of using it (or not using it) in balance.
Take GPS apps on a phone. They’re a very concrete example of consumer software where the performance/cost/risk triad comes into its own:
[list-ul type=“square”][li-row]guidance performance: getting you from point A to point B in the least possible time (or trouble… (opens in new window)) ;[/li-row] [li-row]cost of operation: clear, unambiguous directions, information shown plainly without demanding constant attention ;[/li-row] [li-row]risk: not sending you into a U-turn in the middle of the motorway, up an exit ramp the wrong way, the wrong way down a no-entry street downtown, or even across a landing runway on the way to the airport while following Apple Maps (opens in new window)…[/li-row][/list-ul]
Get the balance between those three components right, and users stay loyal to their GPS.
It’s worth noting the flip side, too: when switching to a rival comes up, the cost of moving over feeds into the equation as well, with a new interface to learn, questions about whether the data will carry across, and so on.
Improving performance
In high-risk environments, not every situation carries critical safety stakes. When they don’t, people will accept a new system that’s a bit riskier (though not critically so) as long as it improves performance at a lower, or at least equal, cost. Among air traffic controllers, for example, studies (Wickens, 2000) showed that displaying predicted aircraft trajectories, even when partly wrong (significant risk), still helped operators sharpen their picture of the situation (better performance), or gave them a good-enough estimate so they didn’t have to work it out in their heads (lower cost).
Better performance, lower cost, risk kept under control: the system gets used. But if the risk of using it went up, it would most likely be dropped.
Coming back to Candy Crush, the same triad applies, except the risk is contained enough that only performance and cost really matter (though a degree of addiction is probably a risk worth taking seriously (opens in new window), it’s clearly offset by a very real payoff).
So how does this triad show up in your own design projects?
References
Wickens, C. D., Gempler, K., & Morphew, M. E. (2000). Workload and reliability of predictor displays in aircraft traffic avoidance. Transportation Human Factors,2(2), 99-126.
