3 ms·

Generally, using yourself as a user proxy for most applications will be worse than using personas, unless you’re part of the core user base. Doing user-focused

Generally, using yourself as a user proxy for most applications will be worse than using personas, unless you’re part of the core user base. Doing user-focused research and road-testing prototyping against users is critical, but impractical during the detailed define and design phase. So personas are an imperfect compromise, but a compromise that works well as long as you keep its pitfalls in mind. Our practice is to define a taxonomy of user types that map to quantitatively-defined archetypes, with personas developed qualitatively to draw out specific motivations and viewpoints uncovered during ethnography. It sounds like the MSFT team got to archetypes and then dropped stock photos and a description on top of them, which isn’t unusual but is definitely not good practice. You can’t just create an “average user” and call it a day, because you need to encompass a pretty broad set of behaviors, intents, capabilities, and needs.

erikpukinskis 8y ago
> Generally, using yourself as a user proxy for most applications will be worse than using personas, unless you’re part of the core user base I disagree. Part of training to be a designer is developing the skill of empathy. You learn to live with users in your body. Arguably, this is the most fundamental skill of design and the primary way the profession functions at all. I think your suggestion that you can somehow model users in your frontal lobe as “the other” and get something more accurate is not correct.
- HillRat 8y ago
  I get where you’re coming from, but I’d argue that design requires both formal and embodied empathy — personas let you manage the complexity around a multiplicity of viewpoints, whereas embodied empathy (informed by ethnography) gives you the ability to understand those viewpoints at a visceral level. Every design choice means tradeoffs between user desires, so personas are useful for “lensing” those decisions and ensuring that everyone gets an equal (albeit virtual) seat at the table during description and definition phases (just as actual users get their seats during prototyping and UAT). Having said that, I know some designers are downplaying personas or even omit them entirely (at the risk of Alan Cooper rising from the sea to take his vengeance on them), so I’d be curious to know your experience in working without an archetype net for clients, especially with journey maps and similar high-level deliverables. There’s nothing sacrosanct in the discipline, and if we can find better ways to get to the end state faster, so much the better for all of us!
  - erikpukinskis 8y ago
    Can you explain more what you mean by lensing and the equal seats at the table?
    - HillRat 8y ago
      Sure! “Lensing” is just analyzing the effects of a design decision from the viewpoint of each identified type/archetype/persona. You can do it informally through think-aloud protocols, or through more formal tools like interaction tables or multifocal journey maps (which get hard to manage!). For example, if you’re designing a healthcare experience, you might stop at a critical journey point and fan out a set of archetypes —- say, physician, patient, nurse, and back-office employee. You walk through how this decision affects them functionally, then dive into the personas to explore emotional and PoV effects, using standard tools like affinity diagrams, roleplaying, emotion mapping, and so on. Once you’ve made a choice, or deadlocked on a set of choices, you can formalize the effects with an interaction table or narrative descriptions, and now you have an artifact that you can refer to elsewhere in the define/design phases, and that forms a basis for user stories during delivery. All of this is basically to help create a kind of virtual roundtable that represents different users’ POVs when it’s not practical to involve them face to face. It also, I’ve found, sometimes lets us intercept usability and design problems that the users themselves might not be able to articulate or, for personal or political reasons, be unwilling to —- which is where your point about “living with the users in your body” is so critical! I completely agree that if you can’t embody a persona, it’s just a useless (or even harmful!) poster on the wall.

erikpukinskis 8y ago

> Generally, using yourself as a user proxy for most applications will be worse than using personas, unless you’re part of the core user base I disagree. Part of training to be a designer is developing the skill of empathy. You learn to live with users in your body. Arguably, this is the most fundamental skill of design and the primary way the profession functions at all. I think your suggestion that you can somehow model users in your frontal lobe as “the other” and get something more accurate is not correct.

HillRat 8y ago
I get where you’re coming from, but I’d argue that design requires both formal and embodied empathy — personas let you manage the complexity around a multiplicity of viewpoints, whereas embodied empathy (informed by ethnography) gives you the ability to understand those viewpoints at a visceral level. Every design choice means tradeoffs between user desires, so personas are useful for “lensing” those decisions and ensuring that everyone gets an equal (albeit virtual) seat at the table during description and definition phases (just as actual users get their seats during prototyping and UAT). Having said that, I know some designers are downplaying personas or even omit them entirely (at the risk of Alan Cooper rising from the sea to take his vengeance on them), so I’d be curious to know your experience in working without an archetype net for clients, especially with journey maps and similar high-level deliverables. There’s nothing sacrosanct in the discipline, and if we can find better ways to get to the end state faster, so much the better for all of us!
- erikpukinskis 8y ago
  Can you explain more what you mean by lensing and the equal seats at the table?
  - HillRat 8y ago
    Sure! “Lensing” is just analyzing the effects of a design decision from the viewpoint of each identified type/archetype/persona. You can do it informally through think-aloud protocols, or through more formal tools like interaction tables or multifocal journey maps (which get hard to manage!). For example, if you’re designing a healthcare experience, you might stop at a critical journey point and fan out a set of archetypes —- say, physician, patient, nurse, and back-office employee. You walk through how this decision affects them functionally, then dive into the personas to explore emotional and PoV effects, using standard tools like affinity diagrams, roleplaying, emotion mapping, and so on. Once you’ve made a choice, or deadlocked on a set of choices, you can formalize the effects with an interaction table or narrative descriptions, and now you have an artifact that you can refer to elsewhere in the define/design phases, and that forms a basis for user stories during delivery. All of this is basically to help create a kind of virtual roundtable that represents different users’ POVs when it’s not practical to involve them face to face. It also, I’ve found, sometimes lets us intercept usability and design problems that the users themselves might not be able to articulate or, for personal or political reasons, be unwilling to —- which is where your point about “living with the users in your body” is so critical! I completely agree that if you can’t embody a persona, it’s just a useless (or even harmful!) poster on the wall.