← Insights Unlocked
Insights Unlocked
UX for startups: MVP tips and common design mistakes
00:37:51
UX expert Oksana Kovalchuk shares startup UX mistakes, MVP tips, and how to design products users actually want—not just what founders imagine.
Loading summary
Transcript36 lines
- [00:02]
Nathan Isaacs
Welcome back to the Insights Unlocked podcast. In this episode, we're chatting with Oksana Kovalchek, UX expert, startup advisor and CEO of Onada, about why so many early stage founders struggle with the user's experience of their products. She shares practical advice for building MVPs, avoiding costly UX mistakes, and making user experience a team sport from day one. Lots of quotable moments in this one. Enjoy the show.
- [00:32]
Mike McDowell
Welcome to Insights Unlocked, an original podcast from User Testing where we bring you candid conversations and stories with the thinkers, doers and builders behind some of the most successful digital products and experiences in the world, from concepts to execution.
- [00:51]
Nathan Isaacs
Welcome to the Insights Unlocked podcast. I'm Nathan Isaacs is principal Content marketing manager at UserTesting. And joining us today as host is UserTesting's Mike McDowell, our principal solutions Marketing manager. Welcome back to the show, Mike.
- [01:06]
Oksana Kovalchek
Hey everybody, it's great to be back.
- [01:09]
Nathan Isaacs
And today's guest is Oksana Kovalchek, a seasoned UX expert and startup advisor who's helped countless early stage founders shape better digital experiences. She's the CEO and co founder of Onada, a design driven development agency that's worked with clients around the world. Welcome to the show, Oksana.
- [01:29]
Oksana Kovalchek
Thanks for inviting me. Hi, guys.
- [01:31]
All right. All right, let's get started. So I love your background, Oksana. So we're going to get into a lot of stuff today. Before we start talking about UX for startups, just can you tell us a little about your background and sort of the work you're doing at Anoda?
- [01:45]
Okay, Mike. So overall, like I have around 20 years of software development experience, so I still started doing this when I was 14 and I was obsessed since 5 years old with development technology. It was totally, absolutely, barely nothing that time. But it's some like, inner feeling. I don't know how to call it properly, but yep. And right now I'm driving in order to help our clients to help people who struggling with bad user experience. For 13 years. I'm also doing consulting for startups in Techstars, in United Nations, German Council, in Venture out and other different startup accelerators who usually invite me.
- [02:35]
Well, people struggling with bad UX experiences, they must be tough to find. Yeah, I totally am right there with you in terms of like having that calling to. To come to do something. Yeah, I've been around for quite a long time as well. I happened to be in the right place at the right time. My boss was like, hey, do you want to learn the Internet? Back in the 90s so really wild. But yeah, it's such a wide open area that people don't realize that there's just so many problems with user experience. We think, oh, people have been doing this for 30 years now. It should all be perfect, right? No, because every time someone new comes in and starts their own company, there's more problems. So talk about some of the most common UX challenges you see with these early stage startups. What are they facing and how do you and Anoda, how do you guide them through some of those problems?
- [03:29]
There is like two big areas. The very first, when people start building the startup, the biggest problem they want this application love themselves like they building it for themselves. They're not thinking about target audience or anything else. For example, today like I'm fighting with my clients so I'm not letting them, I don't know, burn it with fire. And the man, like 45 years old man, he wants to build AI application for women with some beauty products, with beauty consultant, etc. He wants like everything logical. Like usually men like, but not about the women. And we said like hey, no, just stop, please. No, it wouldn't work like the blue colors, you need to provide feminine design or if you want to be like more in logic, that's not a problem. But different user groups, different segments, they have different buying intent in any ways. Anyone who building the software, they won't earn money. And this money background, you need to keep this. Okay, if I need to skip my ideas for example with colors or anything else to earn more money, then it's like decision making. But when clients, when like startup founders, they want to build application, they, they will love the second one, the focus. To be honest, startup founders, they want to spend something between $5,000 up to $20,000 including design and development. And my favorite phrase, what I say in this case is guys, you always will have ideas more than money to be honest. Keep it lean, keep it small, keep it accountable. Because on some stage they just forgot that adding one single button or one single menu it affect dozens of screens and it will be significant change. So it's like complexity will added, added, added. And in the end they see like the total price, they scare it, they worried like what about the deadlines, how it will go to the market, etc. And I always say okay, fail fast, move fast, you failed okay, grab the feedback, do something with it iterate, keep the money for this iteration if you will. Put everything like you're like you're not a God, you're not a very Spiritual person to guess for 100% and be sure with this guessing that yes, it will work. I gonna make a lot of money on that.
- [06:24]
I think that you just said one of the most poignant things I've ever heard when it comes to a piece of startup advice. If you're ever gonna write a book, this should definitely be a chapter. You will always have more ideas than money and that is, that could not be more true. You have to pick and choose what you're going to do. And unfortunately to the first part of what you said there, if you are relying on your own opinion and the opinions of your friends, that is going to probably be sending you down the wrong road. Because this is an argument that I used to have with fights with development teams and design. You are not your customer. Like that is, that is as long as we've been developing apps, people have had to know that. Because what is good for you? Who has all this foreknowledge about your business, your idea, your brand? The customer doesn't have that. They just popped into here from some Google link and they don't know what you know. So. And then the, the other thing about just, you know, a guy like me designing stuff, you know, for women. Yeah. Clearly I don't have the life experience or the audacity to think that I could design something for, for a group like that who is nothing like me. But taking this one step further, we're talking about, you know, we don't have a lot of money, so we kind of have to go fast. We want to go fast, but how do we balance that speed, the desire to go fast with true UX quality. When you're trying to get your product to market and you don't have unlimited funds. So how do you help them balance that out?
- [07:54]
The very first we putting their focus into one single direction. So you don't, when you like lean startup, you want to move fast. You don't need $2,000 logo, you don't need $10,000 branding book, you just don't need it. Just no, no, no, no. Like if I can scream this, I will do it. And like don't spend money on something that you not really need right now. They thinking that, oh, if it's not perfectly done, nobody gonna love it. Hey, even like eight star software that we all using every single day has problems, has bugs. It's not the best. Like money transferring for example, like Neo bank accounts, Neo bank applications, whatever you like, you infinitely pissed off with their experience, but you still continue use it. It's not such a big problem. Second, where we can speed up using existing design systems, like not try to build, like, reinvigorate the wheel, but somehow startup founders want to do this. And I usually stop them, say, guys, guys, no, no, no, no, no. Let's use design systems. Yeah, it has its own limitations. You can bring like additional colors, the fonts or icons. It will be like really very hard task. But hey, you still can explore something like Untitled UI or material, more modern material, UI kits. It looks fine. It's enough for you to solve the problem. And like, stop arguing about shadows, about shapes. It's like it's not the focus right now. And of course the feature set. It's like the most important.
- [09:51]
Yeah, I was going to say exactly. It's kind of like before. It's beautiful, gorgeous, perfect. Do people even want this thing? Like, are they willing to try to even use it? And one of the best pieces of advice I ever got from an old boss of mine was, hey, there's a reason every ATM works the same. Because it's a design everybody understands. So when a new bank puts out their atm, they don't try to redesign it from the ground up. They just go, yeah, this is what people understand. You put the card in, you put your password. Yeah, really, really interesting stuff. And my dad was a mechanic, so I was thinking about him building, building my own car or something when I was younger, shifting a little bit now to those design systems and trying to, you know, get your MVP to market first. You mentioned in the, in the beginning, or maybe Nathan mentioned it, but your work with tech stars, and funny enough, I actually did some work with them, like five years when some of the other players in that space. I was doing a lot of hackathons back then, but there's a lot of mentoring that comes into play there. And so when you're not just through your formal business, but when you're doing mentoring and working with other startups in that way, how do you encourage them to use feedback or user testing? Not, not user testing, like where I work, but just user testing in general. Getting feedback from users to build, to build those things into their process to sort of improve on their mvp. Or figure out what is the mvp.
- [11:13]
Yeah, for figure out what is mvp, we usually working on the backlog, grooming, because the backlog is usually over 1000 user stories. And we say, okay, oh my gosh, no, no, it's like 10 years roadmap. Jesus, guys, stop. Just stop doing this. And when I do the mentoring, I have these Two cards, yes and no. I always come in a card, no. And when they asking me something, I just put the no card like with very, very sad face and said okay, maybe we need to explore this feature. For example, to see the user traction. Your software is not alive even yet. You have zero users. Do not explore additional features. Understand. Just feel these words. Understand that you wouldn't get anyone's attention in a second. Even if you publish your product. You published it in your social networks like links. You upload it to the app store or Google Play or both. There wouldn't appear like 1,000 users in a second. No, maybe you will get two users. Like two people who get lost and download that application. You have a very long way to put it in front of your user eyes. Concentrate on that. Check your competitors. Like how to focus what needs to be mvp. Check your competitors, what users? The real users writes about this like it's pretty nice idea, but it missed this, this, this feature, three stars. Then. Oh, I love it so much. Especially this and this feature. Five stars. Aha. Then you go in total disaster. It takes so much time to load whatever one star, okay, you, you downloading the software or you grab the trial and checking how long it really takes to load the page or make some action. And if you can see for example the really slow, low speed, okay, maybe it's like not very relevant feedback, but in your software you will see, okay, if you see these buzzwords like speed, reliability, bugs, can't find the function. That's what's your users. It's your competitor, it's really your users. That's what they say, steal, steal this feedback and the same like you can put in front of their eyes your software and redirecting them in your software. Just selling your unique value proposition. Hey Gu, we have this function, this function you mentioned that we bring it live and that's enough. Like 2, 3 features is overall enough to build an MVP. Just give it to users. They will say what they really need. But still you will need to filter that feedback. Because sometimes user complains, okay, about really ridiculous stuff like we don't have the Pink theme in DevOps software. What like, yeah, that's.
- [14:37]
It's really funny. Taking things with a grain of salt sometimes is really important. You have to know what to pay attention to, what to not pay attention to. It's one of the reasons why. And I'm sure you know, in your career you've seen this. People go, wait a minute, you're getting feedback from five people, seven people, you know shouldn't you be doing a lot more? And depending on the situation, yeah, you should. But when it comes to your ability to respond to the feedback, you can't have a backlog a thousand things long because you can only build so much. So why waste time continuing to try to find more and more and more things when you've already got those five key things. I actually did a webinar yesterday talking about validating and vetting new ideas for businesses and concepts. So really super important stuff in terms of that like that exact framework. Is there any story that you can share about maybe where you had to advise a UX pivot like you need to change this to. And it completely changed the startup's product or their trajectory in terms of hey, they were trying to do 20 functions but you had them narrow it down through the UX to just like hey, just focus on these three things and let's see if it, if it blows up.
- [15:44]
Oh, I have so many stories. One my favorite one how like the bad user experience and when the client like the founder don't want to meet expectations of users. It's like a sad story. So we receive website inquiry from some startup and they attached the funnel of the users. I never saw anything like this in my life. Just bear with me in a second I will tell you amazing thing. They have 98% drop off on the user onboarding in the mobile application. Wow, like what? I just. The next my question was what happening on that screen? Like, like what? Okay, I become like very interested. Like I was very confused what's going on? We booked a call, I asked this guy to share screen and show me the application. Your guess? Just try to guess what was there. It was a fintech application for personal finances. Okay, so any guesses?
- [17:01]
It was probably right in the very first field or the second field. They were asking for a very, very specific personal information. I had a similar story to this as well.
- [17:13]
So yes, they want access to credit cards and bank accounts even before you saw real working application. And he wanted at first screen he wanted user to make a purchase entering their credit cards data in plain text fields. So they're looking on this like it looks like if it looks like scam, it's working like scam probably that's a scam like a duck duck problem. So all of them just close the application. I'm worried about that two persons of users who passed that screen probably with a fake data or whatever and like the cost of application was $500. So you just download something, it looks it Was looking like real mvp. It's glitch has glitches. Not optimal way, not modern. There is no big name around this, this product. So there is no trust. So in this story like do not ask, ask anything from your users while you didn't build trust. Ask any required information step by step. Like do not overload users with huge form special mobile because like if application gonna crash, user will open this again and the form will be empty and user needs to start from the scratch.
- [18:49]
Oh my gosh. Yeah, we had this. Okay, so now I'm just going to tell you this one story before we move on. So I used to work at Hertz, the cardinal company and you know I built the original website basically in the 90s and did a lot of these kind of things. So one of the, one of the big forms like you're just talking about was how to join the loyalty program. It was massive long form. We built it out, you know, just like the paper form. But this was like 2002 so it was like really early on. We didn't know any better and eventually the decision was made. Let's break this up into multiple five different steps. And when I was doing analytics at that time and someone said to me, hey, can you tell us how it's performing the new page? I said, well you have a massive drop off on page one and then massive drop off on page four. And I'm looking at the pages, I go, well page four is the credit card page. So that's obvious. Maybe people don't get their credit card, but what's going on with page one? So I went and looked at it. The second field was the driver's license number which is not often given online. And I'm like, well you're putting this in before it even invested any effort. If you require that, put it in on page five where they've already committed to giving all this information and then they'll give it. But you can't do it right up front, people. It's too much of a shock. So that's like very similar stories. And yeah, that one is, that's a hard lesson learned. Unfortunately, when people think you're a scam, if that gets out and they rate your app and they say this is a scam or whatever, that's. You don't want that. All right, so to do that obviously to identify these things really UX needs to be more of the culture of the organization. We really need to be thinking about the user experience from the get go, not just, just founders opinion you know, or founders friends opinion. So thinking about user experience, how do you kind of help and mentor these early stage founders, build that kind of culture. Especially when they're saying no, my resources are limited, my time is limited, how do you, how do you sort of help them with that?
- [20:42]
Anyways, whatever you're doing with your application, with your software, it just needs make sense. For example, if you're using font, it shouldn't be less than 14 points on the screen because nobody gonna see it. If you build an app for example for elderly people, the contrast and everything, what's happening on the screen should be bold and very, very contrast. Then if you having like a few blocks and action items related to this block, these action items like button for example should be closer to this block rather than to another. So it's about like paddings and merchants and that's this laws, they just make sense. It's obvious we just not seeing it in like we're not paying attention in our life. And then when you have a very limited budget on the ux, start with user flow. It's the crucial point is the cheapest, to be honest, is it the cheapest stage of user experience like engineering itself, do the user fl. So you will ensure that if user landed inside the application, he can go and reach every screen. Because previously in my agency when we had only development, that was a real disaster. Why startups came to us and say okay, I have small application, 12 screens, can you build it for me? Okay, I started to review the requirements and understand this 12 screens is, let's say it's 20 persons of total number of screens. And that guy, that client, he already paid for this design, he paid for the user experience, barely nothing. And right now he expects like not a big budget, it's only 12 screens. But no, that screens are not connected together. They're missing like huge flows. For example, like in profile there is no payment or subscription management, invoice downloads, anything.
- [22:48]
Yeah, stacking standing up those verticals before you make all those bad decisions and build a disjointed application. So critical I now I gotta throw in the obligatory question here. AI, right? Is there any way that you are using AI to help with any of this stuff? To help people figure out better UX practices and culture?
- [23:13]
Let's say before yesterday, I will say yep, AI helping us on every single second. And the biggest problem with clients with founders, they see their idea in a hat in their heads, but they can't transfer it into requirements. So we need some translator from the visionary Language to technical and business stuff and then from business stuff to technical stuff. Because developers and designers, they expect some kind of user stories, like what do you need to do? Like what this application needs to do. And like clients, founders, they want you need to fill that application. So AI really helps to build in the requirements list. It helps to outline features and sometimes helps as a source of truth for our clients. And the biggest downsides what I see, it's a creative limitations. People starting stopping to think out of the box. Like I was a guest around a week ago on the podcast and we speak a lot about AI. Like can. Can AI replace the designers, for example? Whatever you need to have, like, you need to ask very deep questions to yourself. You need to have very creative mind to put everything together and you need to filter all that information that ChatGPT gives. And yesterday I had a call with one of my potential clients and she validate everything with a chatgpt. We provided her a test task, we we pitching it, etc. And she reads whole call. She just read me like sounding everything that ChatGPT gives her an input. It can't understand what exactly the wipe was on the screen, the colors it gives. Like we need to add more creativity to this application. Mike, it's a funeral application. What creativity illustrations. Like, wow.
- [25:35]
Yeah, I could definitely see where it's not giving you the best. I was hoping we're going to come back to what happened yesterday, because you started with that. That's wild. Yeah. Whenever I talk to researchers, especially who are UX researchers about using AI, it's always trust but verify. You know, there's a lot of hallucinations out there. And yes, it does get trained on a lot of people that maybe don't have the best opinions or the most training on certain things themselves. So yeah, you've given me a couple of really good quotes during this, during this podcast, which I love. All right, so now that we've gotten through a lot of this, the design, we're doing stuff, we're using AI a little bit. Let's just talk like as we round this up. You know, a lot of people, they, they think about metrics only, like AI and, and research a lot of qualitative stuff. But what metrics do you bring in to sort of track UX success? Is it engagement, conversion? Is it just dependent on what the application is for, like a funeral app? You know, certainly not how many people died, but that's not a UX measure. But yeah.
- [26:42]
It depends. For example, if we're doing B2B software usually like we're mostly doing the B2B. There is number of tickets to the support the sales cycle. Because B2B software usually like you need to go come on the demo call. They will be pitched and then maybe some deal gonna happen. And if software like just ugly to be honest. If it's like doesn't feel usable. The sales cycle and the salesperson need to negotiate like weeks, months, all the time. And then when we do the updates we see like real shrink of this sales cycle. And we usually like measure our UX with these metrics. Then if we're doing even like B2B or B2C like client facing application in two weeks. In first two weeks especially we check in volume of support tickets that comes to support of this application. And we check in what questions happen in there. If there is like oh, something doesn't work, not our problem like development team, it's on you. If something. If user can't find the function, if user can't understand how to make something happen, then we okay, we have a problem. Let's sit down see how many requests like this, how many users affected we need to fix. Here is our fail additionally. Yes NPS if that possible. What else? Time to value. It's also very good UX metrics where how long user needs to spend time like inside the application to start seeing the value. Yeah, we have if we're Talking about the B2B application, something for analytics. It's huge onboarding. You need to enter thousands of data connect connectors and then you want to see like finally when and if like for the users we talking like what not hours even like in this case but days. Oh no, we totally screw that up.
- [29:05]
I think that's fantastic that way that you just said that. Because I don't think a lot of people think about time to value for the customer. They think about UX purely as being a service to other business metrics that value that are important to the business. Like you said, like you know, conversion rate, sales cycles, shortening amount of effort, number of touch points between a sale. But the customer also has to get value. And measuring that time to value for them actually is as important if not more important than the company metrics. That's actually really great. If everyone who's listening to this takes one thing away aside from the, you know, you'll have more ideas than you have money line. It's. It's understand time to value in your application. How long does it take for people to get true value using it because they're going to abandon if they don't get value. So that's really important. All right, I'm going to just ask one last, one last one here. And I want everyone to understand this is important for not just startups and early stage companies, but even big, big Fortune 50 companies have these problems and this is it. This is going to be your chance, Oksana, to weigh in and solve the world's problems.
- [30:17]
So.
- [30:19]
The dynamic of founder, designer, engineer or sort of PM designer engineer, what's your take on how that's shaping ux? The synergies or the conflicts between those three groups? And how do you coach teams to sort of avoid or break down those silos?
- [30:39]
They were refers like UX is not only one single responsibility of the designers. Like hey, the project manager should also add value to the user experience. Developers, they hate it. Like if it's not about the code, like don't even call me. But developers, they need to be connected to this process. Start asking questions. Yeah, developers QAs, they hate all these meetings, especially when there is no software involved yet. But it's very important even for the QAs. Connect on the user flow creation and start asking questions. For business analytics. I see like less and less business analytics involvement right now in the project. But everyone needs to talk with a product team and like not say okay, you are UX engineer, do whatever you want. This person not specially like focused for example on future development and developer can say whoa, whoa, whoa, whoa, whoa. No no no. This flows too. It is good but we need to split it because from the software development perspective it's like it takes unrealistic, it even doesn't make sense to spend such amount of money or time to do how you designed it. Like just know the same for project managers they usually can like is anybody on fire? No goodbye. Because the project manager has like thousands of fires out of this door. But here is the place where you can stop creating like the fires on your project lately. Like because I usually tell hey guys, when we shaping the requirements, cost of mistake, cost of error, less than $0.01. You can rewrote, regroup the text user stories, it's like it's nothing, it's pixels on the screen. Then when we do in the user flows it's like around 1 $2 per mistake. When we do in wireframe, 10 $20 up to $50. When we're creating the UI we're talking about hundreds. When we on the development we're talking about thousands. And then when it's DevOps stuff and we Releasing our software, it's like dozens, dozens of thousand dollars. So connect the team for the user experience, engineers for the designers. It's really important to start thinking as a developers because right now I'm seeing the huge gap. So what my company right now really does all my team leads designers. They are switches from the front end development. So we never will design something that developer will come and say oh my God, where is my beer? Like no, because we need design not for a dribble. We design it for the users. And we in the with developers. We in one team when in one board and we need count on them. And developers need counsel on us on designers. And when I see like designers, they avoid the grid totally misaligned screen. They like skipping the part where it should be with the components and consistency. The developer will just implement everything. How he gonna like this day? He wouldn't pay attention to your figma file. He was go away like you have no idea what the real software is and your design wouldn't be implemented. Client wouldn't be satisfied. Developer will point on you all the time and will say that you an idiot. And yes, that happens daily, every time. What I see like in the teams. So to avoid this like Silas everything just start thinking about application on the UX stage, not when okay, we did the requirements then team do whatever you think is the best. No, let's connect, let's grab the feedback like from client. Let's grab feedback from product manager. Let's connect the developer or at least QA manager, QA tester to provide us technical feedback. Then you would avoid like all this back and forth, back and forth and reduce.
- [35:28]
You are just spitting facts today, which is so great. I love the way you broke down the cost of the bug fixes or pivots as you move further down the development cycle, which is amazing. And then the idea of for you guys at Anona that you are having developers that become designers and things like that. So people truly they have experience and empathy for the rest of the team, which is so key. A lot of companies will try to create these little development pods and things, but having people move through the different roles is super great. So again, another great thing to take away for people outside of the at least eight T shirt slogans that you have said on this call, which I think are maybe gonna start a business with these. Is anyone on fire? No. Goodbye. No. I like that. That's really good. This has been such a phenomenal conversation. I want to thank you Oksana for being on the show how does someone, if they want to find out more about you or Enoda, like how would they learn more about your thought leadership and the work that you're doing?
- [36:28]
Do it. Oh, of course. You always can reach me on LinkedIn. I've started my Twitter to share more like thoughts what I'm thinking about the UX and overall the business. Sometimes I'm just struggling there. And of course my email and my website where we post cases and like interesting stories about UX World.
- [36:49]
Excellent. And if I'm not mistaken, you're hiring right now, right? You've got a couple of openings that people, if they want to work with you, it looks like there's a couple of jobs that out of no doubt.
- [36:56]
Yeah.
- [36:57]
Yeah. So go invest in those. Yeah. Oksana, thanks so much. I mean this was so much fun. I learned a lot actually, and I was laughing a lot too, which is great. That's the best kind of podcast I want to host is the ones that are fun and informative. So just thank you so much for being here.
- [37:13]
Thank you for inviting me. Have a nice day, you guys.
- [37:16]
All right, take care. Goodbye. See you all next time.
- [37:19]
Mike McDowell
Want to keep the conversation going? You can find the show notes@usertesting.com podcast if you haven't already, don't forget to follow us on Apple Podcast, Spotify, Overcast or Google Play, so you never miss an episode. And if you enjoyed today's show, please share it with a friend or leave us a rating and review on Apple Podcasts. And until next time, this is Insights Unlocked, an original podcast from User Testing.