Loser-Centered Design or let it be: the Frankenstein Persona's

This first entry is an editorial that covers the phenomenon of creating persona's without user research, which made quite a buzz at CHI2007, even though it first appeared in interaction in 2006. Frankenstein persona's, a particular egregious example of user-uncentered design, is something Elizabeth and I have witnessed not only in practice but also as reviewers for many conferences and journals (all of which incidentally have no name).

All too often, terms are misappropriated, if not outright hijacked, and their meaning becomes either obfuscated or totally changed. Have you noticed that user-centered design is quickly becoming such a victim? We feel the only way to protect this saintly term is to fight fire with fire and provide a counter weapon whenever the term user-centered design is misused: Enter the term loser-centered design.

Loser-centered design focuses on the wrong things, and worse, ignores the user-yet it still uses the term user-centered or user-focused; either to justify some Machiavellian ends or to cover up poor process. Unfortunately, loser-centered design, loser-focused design or loser-centric development (the terms are sadly synonymous) take many, many different forms. We want to give you some examples here; surely you have your own. (Feel free to add them as comments to this entry and you will have the immense gratification of commiserating other designers worldwide.)

Frankenstein Personas
One obvious example of loser-centered design is pandering to a marketing department's market research instead of conducting and using real user research (e.g., talk to someone who isn't in your company, on your team, or co-habitating with one of the aforementioned). You can smell these projects a mile away—the task analysis or information architecture just mimics the research, and neither questions basic assumptions nor adds anything that doesn't fit the pre-conceived marketing model. Loser-centered design in these projects also takes the form of user segmentation based on personas that were defined largely by internal focus groups and a few friends of the marketing department (or anyone else who has read too many in-flight magazines). All too often these personas are created from magazine clippings illustrating brand desires and wants based entirely on marketing mumbo-jumbo. Yet they become so tangible and visually vivid that designers rush in and design for these modern persona Frankensteins as if they really existed. Tell the truth, are you guilty?

The Loser Researcher The loser researcher is a charlatan or inexperienced user researcher, with just enough knowledge to be dangerous, who conducts user research for a product and instructs designers on what problems to solve. Instead of observing what users actually do, they ask users for instructions on what they should design. Loser research has some tell-tale signs: out-of-context telephone interviews, open-ended survey questions, and site visits conducted only in the on-site lunch room (assuming the software has nothing to with employee lunches).

One example of such folly follows. A designer was asked to look at improving communications since something was obviously wrong with the intranet. Indeed, upon being interviewed, senior management offered up numerous complaints that the junior members of the staff could not find the right information on their intranet. These complaints were addressed at great cost with much changing around of data layout and information architecture, but the results did not improve the situation.

A user (not loser) research firm came along and observed what the users actually did. The real problem, as became apparent by observation of the office, had nothing to do with the intranet and everything to do with a recent reorganization and the physical traffic through the work area. During the reorganization the company decided to reward high performers with window cubes, centralizing all the junior and new employees several walls away from the more senior employees. Not wanting to appear as uninformed as they felt, junior folks stopped asking for advice, since they had to expose their ignorance by making a public trek to a senior level staff cube to seek advice. A room redesign solved the problem.

Designing the Emperor's New Clothes
Reducing call-center-response time forms the crux of another loser research opportunity. Convinced that shaving seconds off of each support call would add up to entire individuals that could be dismissed, the user researcher was called in to analyze the business process and observe why results were still below expectations. Through interviewing the call-center staff, the user researcher discovered a huge number of problems—none of which management expected. Told that mistrust, disrespect, lack of information, and unfair treatment was the root of the problem, management reacted with outraged disbelief. "We already solved that problem!" Evidently not. The user research results were uncomfortably human. Subsequent loser research focused on technical issues; while the results showed no promise of improvement, focusing on machines rather than social practices "felt" better to management. After all, one should only have to "fix" people once. When loser research results were available, management breathed a sigh of relief and felt that the problem was well in control—and that its own management style was no longer to blame.

So the next time you hear the word user-centered design applied incorrectly, jump on the offensive and whip out the loser-centered design retort: "user or loser?" There is, indeed, a difference.—eic - Jonathan Arnowitz & Elizabeth Dykstra-Erickson


This is a draft version (preprint) of material which was later published in substantially the same form in my Rant column in ACM’s magazine. The published version is a copyrighted feature of Communications of the ACM. All requests for reproduction and/or distribution of the published version should be directed to the ACM.