Skip to main content

Lesson 9 - The Importance of User Validation

Lesson 9 from: Introduction to UX Design

Joy Liu

Lesson 9 - The Importance of User Validation

Lesson 9 from: Introduction to UX Design

Joy Liu

buy this class

$00

$00
Sale Ends Soon!

starting under

$13/month*

Unlock this classplus 2200+ more >

Lesson Info

9. Lesson 9 - The Importance of User Validation

Lesson Info

Lesson 9 - The Importance of User Validation

I feel like for traditional graphic design a lot of these air just like gut reaction. So when you're designing something, you're when your current ing um, moving the letters around on a logo time. You're kind of just getting a feeling of what? It's right. There is no concrete pixel where you're moving, right. And then when you are designing a movie poster, you're kind of just shifting elements and finding the right place where it's like, OK, this is good. The composition look size, the symmetry looks good. Um, color balance is a little bit green was changed, the color alot about ready to read. OK, great. Lets print it, um, for digital products. I feel like we have concrete evidence here. So, for example, um, let's say the first grader math app rate. Um, we were suggesting Okay, so there is Gamification. There is sound, and, um, there's probably characters and their some live footage integrate into the up. Cool. These are all assumptions. Now let's go test right? If the test come back, ...

comes back to negative, um, or not so great. Maybe the kids are like being all scared wife by like the live action character that we provided were like, Oh, OK, that wasn't so good. Right now it's time to make. Maybe we can cheese I go dog character instead of like a human talking to you. Like they These are the stuff that we can swap out the variables and just started testing. Um, or, for example, we might be talking about the smart toilet, how we are targeting babies and elderly's. And then maybe the elderly market did a fly so well, right, and then that were like, Okay, great. We don't spend that much time into. That is just an assumption. Now we tested and it's been invalidated. Great. Let's just focus on the other markets and it's we can always some build upon none. So there are concrete evidence based on your validation. Um, it is driven by data. So these are these are your, um, these are sort of the tools that you can use when you're negotiating between business goals and user goals. So let's say if you are the user advocate, right? And then you kept on say, you know what? The user really want this, And if you're talking to a client and the clients like. No, no, no. I know what my user wants. You there wants a B and C. Then you can use data to prove that. Hey, you know what the user really wants? Feature deed? I don't really care about a B and C so we can go your way where it might not be effective, but or we can go our way where I did do some user interview. I did this time use a research. It's showing in the data. Right. So I think that's diocese importance of validation. Um, you can also set up validations to compete, compete against yourself. So when you do, maybe you built a nap and its version one then you would go out, do around a validation then that becomes the baseline for you to compete when u boat version to race so you can build off something and say, OK, well, did this perform better or worse than view the the want? Very. If it's war, so Okay, well, how can we improve upon it if it's better than V two is? But if it's better than V one, then say OK, well, now this becomes a new baseline that we can measure against the three. Um, and the importance of validation is also giving you feedback before it's too late. Um, it's definitely if you feedback before you ship out everything into the APP store. You spent all these this money in in marketing, right? And then the product. I don't work as well as you expected. It was like, OK, well, now we just spent all of all the money on the marketing budget, and now we run out of money to do more development groups. Um, so I think there is a lot of, um is very flexible in terms of designing a product and is OK, but you don't have the final answer. Um, a lot of times. Well, how I feel is sometimes I'm just like I'm acting esta proxy for the users like I don't really like. What I'm doing is I'm I'm taking the design from the product team and then I'm showing into the user. I'm gathering the user feedback and I'm bringing it back to the product team a gun and it's OK if I don't have the answer because I don't. Maybe I am not the target users race, so anything that I come up with. They are assumptions. Maybe they are 90% accurate. Maybe they're 10% accurate if there are 10% accurate. I'm probably not doing a good enough job, but maybe most of times, like the difference between 90% to 95% and that kind of means a lot. So then what you do is you can do, um there are different types of tests that you can do, right? So if I if I come up with two designs, I can say, OK, um, here is a method to test us so we can do a B test, right? We can set up lay out number one li L number number to show it to people to see which one they like more. If, um if I have If let's say the technology is so far out of my mind, for example, let's say we're designing for NASA now. We're designing, like the next rover like to oh, no Jupiter or something where there is no concrete ground, right concrete surface to land on. And you're like, OK, I don't know what I'm doing. Right. Then I think you should It's okay, though. right. It's like start making assumptions. And if your assumptions are wrong, just start crossing things off. And then I think as, um as you spend more time in U ex's I like Euro. Some show will become more accurate time after time, then like towards the end. You kind of have this, Um, you kind of have this backlog of all the best practices up your sleeves that you can just pull it out. And you kind of know already that this is going to work. Um, when you were starting out, obviously, you probably don't have that big of a sleeve for that. As you go, you're experience will only help you further enhance yourself so again is OK if you don't have the final answer qualifying. Um, it's OK to just built something out tested, and then you should always embrace surprise. Um, yeah, there comes a lot of time. One. I just learned something, and I think, Oh, my God, the user's gonna love this. And the turnout they hated. Um, don't take it personally. Think, um, it's always interesting. And I think it also makes you humble that used you don't know your users if you're not using. If you are not the target users, you don't know them for sure for sure. And the only thing you news keep asking the right and then if If they do have surprises for you, let's call it keeps your job exciting. And then you should always anticipate change because digital product is so fast, Um that, like anything is possible. So, like, always think outside of box, right? Like the wearable device we don't have wearable to rise a couple years ago and now, like everyone's worrying Fitbit and all that And yes, oh, that's all on empathy in design. And I think if you survive, be exercise on empathy and then if you think you have, ah, awareness on people's frustration and I think, yeah, maybe you access the right feel for you. And, um, we'll take some question now to see you guys have any questions. Yes, um, what validation points to use. So it's validation for you. If a customer buys the products and pays you upfront, are what do you use for validation? Validation for you? Yes, so obviously depends on, like what type of product you building if you're building a marketing page. Obviously, the sign ups it's the most important. Ah, validation point, right. If we are building a, um, if we're building an e commerce page, um, it doesn't matter if we have 20,000 visits. If there's only purchases, right, so then purchase at that point becomes the bigger point. Like like how many people do purchase? And if a lot of people purchased, then what's the volume of purchasing? So there's some a lot of there's definitely a lot of metrics, so you can measure depending on your site. And then if you are building an internal tool, then obviously maybe the completion rate or the speed to complete something, the task completion and the speed to complete a task becomes the most important thing. Also, these days you have great tools like crowdsourcing Kickstarter to validate before you build something as well. It's a physical product. Yeah, yeah, Um, a lot of times people think, oh, user testing or validation means you have to, like, get a focus group. You have to set up everything that's going to cost $4000 to run. We don't have a budget right now. It doesn't have to be. It can be just as you can do something called hallway testing. Right? Just walk down the hallway, grab someone and say, Hey, I have a quick question for you. Come here. Take a look. Which one? Which design? Do you think it's better? Right? And they were like, I think that was, like, Great, right? And then you move on to the next person. It's free. Um, so that it's called Holly testing, right? Just got to go down the hall grass on one really quick and do something. Do some improv with, um um, another one is called, um, guerilla testing. Red girl user testing. Um, so it's just very is like guerrilla tactic, right? You go down the street or you just go outside the building. You just grab someone and Assam or do some quick research. Obviously, for those, I think you have to grab the right people. You have to kind of find the right. You can just go up to some random stranger. If this is like a very specific product you're building, um, if that's the case, maybe there's some other. You can also recruit people from Craigslist. You can also, um ask other people to help you find the right people to perform these quick usability tests, right? And then there's a lot of tools out there to that helps you through these tests like, um, user testing dot com can be one. Um, you I dot com can be another one where they do the eye tracking. Um, you can even do on thinks he's ability help dot com That also does. He's like, Quick A B tests or five second impression test. Um, what else dismissed the survey? Monkey can be one, too, if you just want to throw out some surveys to kind of get a sense of trends. Yeah, cool. Just adding on to that I'm curious about is when you are user testing, do you have a checklist that you go down if you want to see, you know somebody's going to be interacting with the app and, um, you know, apart from them actually physically using the app and you might be collecting data through the app, Do you have a checklist that you go through? Are there certain things that you ask as somebody's interacting with the app to communicate your data? What's that process. Like for you. Yeah, I think, Um no matter what type of test you are doing, you should always have an objective going through that, right? It's not just I'm going to do the test. This you have to really think about, OK, so what are you testing? Um, if you are testing, like, what? Stored of metrics? Are you measuring right? What? What data you collecting from this test? Is it the results from a, um, layout? A. If, if layout A's more successful, only I'll be right. Like which one do people point more to two. And, um, for checklist wise, there is a lot of resource is on here. Stick evaluation. So, basically, is it is this, um, big check. Let's stop. Were they written by on Jacob Nelson? So he has this very formal checklist, and then I think people can just changing a pillow bed to make it less formal because that it is super scientific. And you're just like, on a scale of 1 to 5, like house itself, so that the user complete this task gray and then a lot of times is just impression Test, um, on user testing dot com when you're setting up a test plan. Um, they also have a question bank for you already easily, even just kind of up. Drop in some pre populated questions as gonna help you gauge how all your website is performing. Yeah, So there's a lot of pre populated or prove return questions that you can use, and you can also write your own. And, um actually, that is also being covered in the other course. Um, yeah. So while or more if we do that, Um, but basically, to sum it up, yes, When you are doing testing, don't do it on a would. Definitely. If you have existing data, use it to your advantage. If you are collecting new data, sure, let's create a baseline, right. And then let's do another test one. You make some changes and then measure against

Class Materials

bonus material with purchase

Joy Liu - Events and Networks List.pdf
Joy Liu - Prototype Softwares and App List.pdf
Joy Liu - Reading List.pdf

Ratings and Reviews

Sasha Pax Malich
 

It's a solid introduction to someone who has never come across the field, as it introduces some basic terminology used in the industry.

Student Work

RELATED ARTICLES

RELATED ARTICLES