 So I'm sorry I lied to you. I have no idea how to talk to girls at parties But I can talk to you about how to talk to your users in your app I'm Alex, also known as Avsa, if you have seen it from my head, from my pin and from my badge here So I have an interesting curious story on Ethereum because you could say that My resumé is a long list of flopped projects And a long list of dead and failed projects But I'm very proud of each one of them because from From each one of them they were built with bricks that have outgrown and outlived The original projects that they were like and and I think you don't actually see that in other industries It's this this wonderful thing a thing about working on open source But most of all I'm a designer and I think The role of the designers and the space is you need to be a translator, right? You're connecting the users You need to both understand the project and good enough that you can translate it into simplest Words for the users, but you also have to talk to the user and figure out what's really missing What's what what is the real hurt? What is the real problems and you need to take back that to the engineers who might often not see that Because they are not typical users. They are power users and often they will tell you things like oh this We can solve that this isn't possible and it's your job as a designer to understand What they're saying understand what the users they are saying to figure out a way that you can talk back to the engineers In a way that transmits the needs of the user users And if needed you should also be able to talk to the CEO and that's where user research is very important Because you can go there to the test and say you know what we are taking this approach But it tested this other approach and people were a lot less lost by doing that and That's why I think it's so important And I think what if it creates sort of what I call this design gear in which you have user experience Which is this idea of how we want things to work and then you use that to build your UI Which mostly are what the pixels are you're pushing pixels around and painting stuff, right? And then use that UI to test with real users to figure out, okay how much the experience I wanted to give is experience that they had and That's you keep turning on that gear and at some point you should be connected to the dev gear And like sometimes you'll be pushing that gear and helping them get along by just Doing a lot of tests before they build it sometimes they will be coming back to you with with things So it's a little a two-way gear If your dev team is just coming to you saying we've built this product make it pretty Then something is going very wrong in that company and and I think anyone can do it And I mean if you have a startup with three people, you're just the CEO We don't even have a designer. There's a CEO the CTO and like a dev guy You can still do it with like online tools and I hope to give them to you and the first thing I need to you need to remind is whenever you're talking to a user This is not teaching your users how to do your app. This is not a tutorial There's not a moment where you're trying to do a marketing on board people. You're listening to them speak You're not explaining your app and you need to remember that your app is being tested not you So it's not. Oh, do you think you can do this in this app? Do you think you can click this button is really it's about do you think this app will allow you to do this? Do you think that can you put money in this thing right use generic words like that and Classic user research is basically there's like one main way one way mirror Rooms with big prototypes or even like paper prototypes as you build around and just get to random people in the office But if you don't have an office it becomes a lot harder and I had to learn that the hard way when I spend like mostly a year working Balancer doing you the research and like there are great tools online I want to give them a few for you usability hub is a great place where you can learn those things for instance five-second tests Just show your app for five seconds and ask questions. You'd be surprised how much of Before they can read your copy before they can even understand your app How much you already form an impression by just looking at it and knowing okay? You are like an investment thing you're like a game You're like a social media by just taking a look at five seconds Then you can go one one one thing further and just ask the user for a first click and just and by this point It's just a plain image like this can be like that image You're just showing it to your investors before you build that product because they want to test it And you can figure out that oh if everyone is clicking on your hero image instead of the bottom that you want them to click Maybe you shouldn't have a hero image Maybe you should just show the information they want and then from that you can do a navigation test Which it's just slides that you can try to see where people are going and one of the nice things with navigation tests Is that you can use them to benchmark your competitors? You can just screenshot the flow a typical flow of your competitors and then screenshots The flow that you want your your your app to do to test and just see is it any better? Is it faster? Are people more confused sometimes? You don't even want to you just want to see how much people are clicking on the wrong links That's more important than the right links But mostly a lot of user research is actually talking directly to people with online meetings online meetings Of course you schedule meeting with Calendly get like them on the zoom or Google Hangouts, but there's also great website called user brain Which allows you to give them a link give them a script and then like about a few minutes They will give you a video of someone using your app and looking at it and describing and it allows you again This is a great feedback cycle. There were days in which you wake up Design a little bit put a prototype in throw it to the user Throw it to use a brain go back to like check my emails come back half an hour later, and they were like three or four Videos of people struggling with the prototype. They just built and honestly just need often three to five users It's not a number of numbers game you get most insights with like three or five numbers and then you can just go there fix it and Creates like a new user user thing and like get get your coffee and then when you come back It's like it's still early in the morning. You've already like run that prototype a few times and It's not The reason that often you get those fast videos on user brains because they are like you have to pay them You're almost in almost case you have to pay user brain may may like may be expensive But you know what what's more expensive is Spending weeks of engineering building something that it shouldn't have been built in the first place Sometimes when you're just having Google Hangout calls you can like be creative I've I've tried giving NFTs to people that I created they gave them I promised them to pay gas fees and which like when gas fees were expensive. That was a very nice thing and I just was offering people. Hey, I want to give you money I want to give you money to watch you do stuff Which is weird Right and that's that's where you your users might feel a little bit Uncomfortable with that because there's a lot of things that scammers and user researchers Who have in common like both of them are promising you easy money to do something like in 10 minutes They are asking you to click on shady links like figma links or whatever that is They're asking you to download some browser thing that you allows me to show your app and see you while watch your wallet Right, so your users might be like the only difference is one of them wants to see your type or password The other doesn't how do you overcome that? first of all try to use your Yourself as Has a leverage like if you have a reputation just tell them look I am a real person. I'm not a scanner You know me from whatever or sometimes I would start a Meeting by telling them how to be safe on the internet like how do like how do how can you do a Google hangout and share? Just one screen instead of the whole just one window instead of the whole screen or just tell them I don't want to see your never show your passphrase get your metamask window in somewhere else and You can also be creating in many ways. So there was a point in which I wanted to know Is there a way that we can like a pool? The balancer is a if I happen had this pools and we had like there's a lot of information that goes in a pool And I was wondering can you like translate that into a tiny icon? I wanted to know how small the icon can be and I created a little game as a prototype game Which I just ask users can you find the die the fastest person to find all the dies in like this Google capture thing We win and of course I didn't care for people who find it to find the dies what I cared is that as Each step was progressing. I was making the icon tinier tinier until at some point How they had to work with was the two color gradient and they still found it And it's a great way for you to understand. Oh my god like you you can reduce it a lot by just making it small, right and Another super important thing is that options are never equal We will often happen that you have a system in which there are two modes, right? Balancer had two modes in which you had deposit money Doesn't actually matter what the modes are because for the user is just wow, wow, wow, wow, right? It's just random noise for them And the wrong way to do it is just to ask the user to like if we have two modes on the system We have two modes on the UI and that's often the worst way to do it. First of all, you can Do a little a better job by just selecting a default just it forces you to think about the default Forces you to think about which one is better often I've seen people would think that mode a would be more important because everyone on the team was using mode a But turns out mode B was the most common because most users don't have all the tokens or have connected their wallets I have a done all the things that most people in your like death department have But a better way is don't make it a choice thing Just make a checkbox and then you have to explain it in a tooltip. No Never try to like your users will not read and there's this often there's a mistake of thinking that your users They must understand your app when the opposite is true You must understand your users and try to reduce whatever explanation is there into like a five word explanation But a better yet is instead of having your users pick a choice Just allow your users to do whatever they want and just add later a little thing saying hey Click this button and we'll like I mean I see you're trying to do X click this button You save money or something like that Often it gets even worse when you have like multiple options there So try to make them in different ways. Maybe one is a checkbox and the other is an action You have like one big submit button. Maybe you have to submit bottoms notice that the submit I also picked a I also picked the defaults in both of them because like one is big one is the normal Something it's also great if you're a designer who can code a little bit You should try Framer, which is another wonderful thing because often in a nap you are thinking about some sort of behavior and Is there a way they can like make the video go Okay, can you yeah, just click there often you are thinking about something and then in the end It's going you're thinking about a behavior that you can only understand when you see it So this way I was testing a thing where I realized oh you can put money in many ways You can she select it by the prior amount of tokens you have them Percentage of tokens the dollar value. Maybe you can click on the on the on the on the total and you can type it and you can Scroll it and he was like a wonderful idea on the design space But of course when I tested with users, they were all confused on weight I clicked this and I clicked click that and it's all confusing, right? But it's it's the kind of behavior you only understand when you actually make it work with real. Oh, yeah This can you click the video there? Thank you and and then Hello, yeah, thank you So and then I realized that there's a lot of things you can do but just Testing the behavior of people just allow people to click around and do whatever they want and just allow them to Click around and check. Hey, if you click this button suddenly your numbers will change here because Numbers are also you why if there's a button there and they realize oh if I click this then this goes up This goes this goes down it really simply works And you can do a lot of things like that in which you can like and those are the sort of behaviors You only realize When you are like just running a live app And that's something that you can do as a designer I'll alone in your office and then keep testing it keep testing it keep testing it and it's it's it's really important Don't overdo it because I've seen cases where everything is clickable Everything is great and the user is just clicking things around He has no understanding of what they're doing But they're just glad that they like they disable it becomes a game of can I disable all the warnings? Right, I'm gonna click around until all the warnings are gone other things There's always a like a Facebook moment whenever you're talking to user and I think this is one of them where we had like a little box just Setting all the actions that have happened and then I realized that users are confused about that And then I realized that they don't know that this is a feed of actions They think that this is an I mean do of things that they should be doing and then yeah Maybe you should we can redesign it as a social feed and that's another thing in which Like you come back to your product team and they was gonna say it's sort of impossible because not everyone has the NS name Not every NS name has photos. That's okay You can solve that just give the nickname nicknames just call them whales give them emojis, right? The problem is just make it look like it's something more that they expect it like activity And finally there are a lot of ways and a lot of moments in which you're struggling with words struggling is trying to understand and explain a concept to your users or Even to yourself That you actually don't need to then you need to take a different approach a lot of my time I spend in balance. I was thinking about what is impairment loss? Every defy product has this issue They all call it divergence loss in which you put money But if like one token goes up by 10% and the other goes up by 20% in the end You lose money by putting it in the pool instead of just keeping it with yours and then when they try to explain it It feels weird. They put it like graphs like this But in the end like what the user wants to know it's not whatever is the impairment loss They want to know something very simple. I have one token a I have one token b And that's the value that they have together that portfolio over like a month or like in dollars eaters Whatever right and if I had invested in your app That's how much money I would have and of course you can see that the user is better off by not investing in the app And that's true for most defy apps And I think that's a great pattern because I think most apps don't actually try to solve that problem because they Realize that by solving that by a visualization the user will realize that oh my god They're losing money. I should not be putting my money on on on this uniswap pool, right? And that's the wrong way a wrong way to think it you need to think in a different way and realize that's not a whole story There's there's not only thing they've divergent loss thing There's also like fees you collect some fees as things are traded Sometime like that can bake up for the divergent loss thing some other protocols will give out Liquidity incentives right just by the fact that they are using it They're gonna give you like other tokens and some others will like put your tokens to work putting on staking platforms on leading platforms and It's all sort of complicated and it's all sort of complicated to explain often They will tell you look this is the API. This is the APB. This is the impairment loss This is the amount of tokens you get in terms of measuring tokens the amount of thing you have measuring dollars But really what the user just wants to know it's what I called is like rainbow short rate develop which is just the dotted line is what would happen with my tokens if I did nothing and The color bars or everything that this app does for you an interesting thing about this is that that little difference I'm not sure you can see it before that the gray the bottom gray bottom green and the dotted line is Impairment loss. That's how you visualize it. That's the best way I've learned to visualize it But the funny thing is that by finding the best way to visualize it I learned that you don't actually need to visualize it You don't actually need to explain it to the users like all we need all they care about is not the bottom Of the bar is the top of the bar is and that helps you explain what the product is Your product is just taking that and doing all that stuff to make your money work for you, right? And then you can take all the other information like the volume trading etc. And put it in a you know in an order And I think that was one of the That was one of the best ways I could explain to someone how your API works in 25 protocol And I haven't seen anyone do like a solution like this if you are defy guy Just feel free to use it right but my point here is that in order to get there You need both to understand deeply the product by just understanding. Okay. This is this is what we do This is what this divergent like what is divergent loss, but you also need to understand the user What does the user really want to know right what do they really care about they just care about how much money they made And I think you were in the middle trying to connect something that is complicated with something that the user wants into something that is roughly easy to look and other grading sites I've had is Often never treat an unknown variable as zero Very very often you're usually just the fact that you don't know how many tokens you have in your user wallet Doesn't mean that they have zero tokens Maybe just because that information hasn't loaded yet That doesn't mean it's zero and often I've seen users fall into errors because they were like they would open the app There are a bunch of errors and those errors were not like actual errors It's just that because the users were had not collected or something had not loaded a lot of issues like that Don't tell users show them things as I said before is like often instead of trying to explain your user How to do it just allow them to click around and like if you want to look if you click this button This thing will happen just allow them to click the button and they will see the thing it will happen because Numbers are also you why right and I think that it's go goes for both of them Often I would ask a user saying look what is this number represents and it would think a little bit and think I think this number represents That number multiply that that number and they would like do it on your head And if it if it made sense they would be yeah, that's that's it I found that they would be gratified so often just showing numbers and allowing people to think with numbers I was just saying look this this way is cheaper than that other way That's the best we why you have right if you just like you can put like a red button But if they tell you this way is five dollars cheaper They will find and click that that little like that little icon there And users love to see other users like just allow them to see action in your app Like they always would love when you say oh, I had a user would would say that you liked Watching the activity on a pool because they he knew is that every time a user traded He was going to get a few cents on that and he was like yay someone traded I just got a little more more money on that and if you do it live they will love it and Finally avoid great patterns often not solving an issue is an issue And I think a lot of DeFi protocols do this because like they are not sure how to solve it or they are not sure Like or as I say like by solving it showing them the information the user might realize it's not a good idea Just think about it. You can certainly solve that That's the main message I want to tell you is that like as a designer your goal is to connect both sides and as a Developer or as a CEO or anyone in your company you can be that designer You can be someone who can try to always connect and talk directly with users because You have to know your non-customers also Just the people in your discord server. They are not your market fit, right? They're often people who like are already most knowledgeable about your project What he really wants is people who don't understand the product that you need to go there and grab them and make them understand the product and That's it. Thank you You can find me on Twitter at after and I really like thanks to the whole balancer team that helped me during the year and The whole in ask you made it. I've been working on the past year. Thank you. I Think I have time for questions Yeah, it's great talk We will try to utilize the stuff that you said in road key, but I'm really curious How can I get girls at parties like that? That's what I came to know I'm sorry. I lied about talking to girls at parties. I only know how to talk to users I have no idea all the other. Okay. I'll keep searching. Thanks a lot other question Hey, great talk Can you tell about good ways to reach users online? That don't seem like a scam Yes, so I often had this problem because like every week I was having to reach out to users not some point I sort of like reaching out to our discord community and reach out to Twitter There's a point in which people are just get like you just can't kept keep asking people to click and click Hey, can you click this thing? Can you do me a favor? Can you click this? Can you participate in this test? You have to be creative There are some apps that you just give you out like that that like user brain or usability hub They you can just pay and they will get users for you often you have to like I think incentivization is a good thing there Just tell look at whoever clicks this link and completes this form You will get a A B and C like schedule a meeting with me You get $10 like complete this form and you get like one die Like talk to me with one hour and I would like There was a moment I one day one time I promised everyone I would pay every gas fee or whatever they did While they were I was watching and one guy literally said oh my god I'm like it on the video. He could say okay. That's great I'm gonna I'm gonna be withdrawing and depositing a ton of stuff because the gas fees are expensive And I know you're you're buying it you're paying for me But it eventually it was a great video because he was like oh damn I he needed to follow those instructions I just can't do whatever I want and then just by seeing him like going and withdrawing his money and etc It was super useful information. And yes, I paid his gas fees and he was everyone was happy So you just need to be creative I guess Hi Alex Yeah Nice to meet you Do you think we've web 3 user research should be or is any different from web 2 user research? I Think one of the main things I've seen it different is that There are a lot of apps that you you allow you to like record someone like just clicking around the website, right? But in order to that you you only get people who don't have like a wallet ready, right? And often if you want to see someone actually using web website with a wallet You need to be recording them on a Google Hangouts and then as I say they will be Uneasy about that because you're like why do you want me watch like two transactions? And it's complicated because they might be worried about you like getting some information I think that was the main difference for me. It was first of all making sure that I was reaching those the people who had Who had like metamask had everything installed, but those were precisely the people who would not be like very likely to just Be thought like be using their wallet while some way weird guy was watching I Wanted to ask you I mean you just mentioned the tools that I mean you can use to record the users and so on Do you think I mean we also have the team from rotki here that are focusing a lot on privacy Do you think especially when we talk about defi and about users interacting with their own funds? Do you think that's a that that's something that users might feel? I know that their privacy is bridged or just following them and recording their actions on the website Yes And like their privacy is being bridged because you're recording so I Think you're not talking necessarily about User research you're talking about more analytics and I think I actually think that user research can help solve a lot of the Problems with analytics right if you just create a website with a bunch of trackers everywhere You get basically the same information that you get if you just watch five users Use your website with the difference is that those five users have consented to be in your research fame While when you do analytics you're making your websites lower and you're like collecting a bunch of information Information that can be used against every single user in a future leak or something like that Right, so that connects like their their IP address to their to their to their transaction things And I think that's a lot more dangerous. So I think just Like if you're trying to analyze an app just a blockchain app You can just do on-channel analytics that is already public and do use the research that you could you probably have almost the same insights as like Super Google analytics thing Thank you. Thank you. Thanks everyone. Applauso