Brand Design Tips From Linear Founder Karri Saarinen

0:00Building a website for your startup is easier than it's ever been. The hard part is creating a unique brand that really stands out. So today, we'll be joined by Kari Saarinen, co-founder and CEO of Linear, to review your websites with an eye for what makes a memorable brand. Welcome to another episode of Design Review.

0:27Kari, thank you so much for joining us. Well, great to be here. So you were the first designer at Coinbase and you were a lead designer at Airbnb before ultimately starting Linear as co-founder and CEO. I'm curious, what have you learned about how to build a strong brand when you're looking at your website? One of the first things I would say that it's something that it should be authentic to your company. I think these days maybe a lot of startups look at I don't know Stripe or some lower linear or some some other company out there and they they think like okay this is the way to build sites or this is the way to build your brand none of those companies started that way like Stripe first website was very different than the current website as well as their product and similar with linear it's like the first website was very different than the current website so I think like you should understand like what stage of a company you are and like what kind of people are you trying to find.

1:20These days there's maybe this trend of and like because it's easy to build websites you might try to like build very polished or fancy websites very early on but what that kind of like tells you is like it's maybe sets the wrong expectation for the user that if your product is just that I don't know you build it last month and like you don't have everything figured out yet the product might not be there yet. It's not that polished. It might be set this wrong expectation that this is a very, I don't know, like polished product already and it does all the things that you promise it does. So I think like there's the danger that you go too far like trying to look like a large company or like a mature company too early.

2:02So I thought like maybe it would be interesting just look at what it was the linear website initially and what it is today to kind of highlight this difference. Love that. Let's take a look. I pulled the first linear website from the Wayback Machine. And in some ways, I would say, I don't know, the website is fairly simple. It doesn't look horrible. But also, I think it's something we created in basically one day. We didn't want to spend weeks or months trying to create a website where we think actually the effort should be put into the product itself. In some ways, some of the things with the brand or with even the copy is that we wanted to use this issue tracking term because like if you go to any large company website like you go to Slack or you go some other productivity app website it often says something like this is the work platform or this is the where the work happens or something very vague yeah and I think like that comes later when you're trying to build this narrative but in the very beginning what you're trying to sell is you try to find someone to actually use your product and like just understand what it is.

3:08So what we did here was that we wanted to put the issue tracking. We knew that that wasn't the long-term vision for the product or the company, but we knew that that's like something that the people we are looking for, it's just early stage companies and engineers will understand like this is issue tracking is a technical project management in a way. It doesn't attract people that don't know what issue tracking is. So it sounds like with the copy and specifically your headline, you honed in on like the specific phrase that your ideal user would resonate with and be like, yes, I'm looking for an issue tracking tool rather than we're building a work platform, blah, blah, blah, where people would say, I'm not looking for a work platform.

3:48I'm looking for an issue tracker. Yeah. And I think it's, it's the way to think about it. It's like, you are trying to maybe filter out people. You should be very specific initially and maybe like try to like, you can talk to investors in very ambitious ways. You can say like, yeah, this is going to be the work platform of the future, but like on the website you can just put like this is a chat app yeah and like this is a nice chat app yeah like just do that that's such an interesting point which is it sounds like what you're saying is like own being small own being a startup and there's a lot of advantages that come with that rather than trying to pretend that you're some big company and that will set better expectations with users and more resonate with them and it's also interesting that a lot of the times during yc we spend a lot of time working with companies on their two-sentence description how they describe themselves.

4:34And oftentimes we say that there's usually two different two-sentence descriptions. There's the one for investors, which could be the big ambitious sounding thing. And then there's the one for your users where you want to use specific language that will resonate with them, maybe jargon that investors wouldn't understand. And so it's really interesting. You're saying focus on that customer two-sentence description, that customer pitch, and really put that right front and center. Yeah. We don't have to go too much in the design itself. I kind of used this screenshot to show, hey, there is some kind of product, even though at this time we didn't really have it ready, but we were working on it.

5:10So there is some kind of product. It kind of looks nice. You can a little bit see what the features might be, but it's a little bit faded. So it's a little bit mysterious. So it's kind of curious because in the end, what we want people to do is put their email here and sign that join on the wait list. That's so interesting because a lot of times I think people put a screenshot on and they just put the literal screenshot. And a lot of times it doesn't work because people are not familiar with the app. And you're not trying to communicate all the different features and everything like that through this.

5:37You're just trying to kind of build mystery and be like, this looks like a modern tool. You should check it out more. Yeah. Yeah. So I think like it's it's very like simple message in the end. And I think like we also like went very simple with the list of the site. It's like, OK, it's like fast because that's one thing we heard about a lot. And after Superhuman, we were one of the products to add this like common menu that like also makes things easier. and we showed that it connects with some of your tooling that you already use. And I'm like, here's the team. We have some, this is not some random people, but we have built other things in the past.

6:08So it's very simple. It's one page. There's no other pages. We don't have an about page and all kinds of... We don't have a blog and we don't have anything really. I think for us it was just like, let's just put something out there to capture people that are potentially interested, that show that we're pulling something new, but not exactly kind of explain too much what it is. And so this was like a week of work or something like that? Yeah, like a couple days. Couple days, yeah. This was how many years ago? This is 2019. Okay, so six years ago. And so now it's evolved a bunch because the product has evolved and what you do has evolved and you have a lot more customers and connections and integrations and the product and everything.

6:49So I'm curious to see the latest version now. I think this may be like a third evolution of the site. and the second one was the famous purple website that was replicated a lot across the companies. And I think with this, you can see that the title is different because the product is actually doing more. So it's not just issue tracking, but we think there's this building part that has multiple different kinds of features or workflows you can use. And then there's the planning part, which is also the planning, like the roadmaps, the projects that you can write project plans. but we still like to keep it pretty clear like here we say like issues, projects and product roadmaps so it's not like again it's the language that technical teams would use and yeah like I think visually it's like we went more this kind of black and white we try to make it a little bit mature in a way like from the purple from the past like we still have like nice effects and like we like doing this that they kind of signal that we care about the experience and that's kind of like why we do this.

7:55But there's now more information. We try to explain some of what you can do with it. And we're still like pretty direct that these are like, yeah, you can do analytics and you can do like different, we have different kinds of features. So in the end, like I think it's something that you should think that the website, that the brand always evolves with your company and your customer base. So even now, we have more larger companies, enterprises. I don't want to turn into a fully enterprise Oracle company, like a brand or something. But I do still have to think about it. It's like, is this something that fits to the current customers?

8:36I think one of the interesting things to me comparing both the original one from six years ago and this one now is how like you've resisted the urge to make it overwhelming people with features. And, you know, obviously the product has grown a lot. Your customer base has grown a lot. The company has grown a lot. And I think a lot of times the temptation is to want to throw more and more stuff, you know, on the page to communicate that to anybody who's coming here. and it seems like you've actively fought against that to try to keep it that same simple kind of ethos that you had when you launched six years ago.

9:10The way I think about the landing page, the first, the home page is kind of like the front page. So you can't explain everything. You can't explain all the features. You can't explain all the customer stories and everything. That's why there is other pages in the website. So we now have specific sections for specific pages and sections for the areas of the product. So you can go deeper in it. So I always think like the homepage should be for the people who don't really know about you. So it should give you some kind of enough information that you know that is this interesting or not. But not too much so that you get overwhelmed and not too little that you don't understand what it is.

9:47So you have to find that kind of balance. Well, it's awesome to get a sense for how you think about this and building your own product. Should we take a look at some YC submitted websites and give some feedback? Yeah, let's do it. Okay, so first up we have Sprites AI. Let's take a look. Build custom AI workflows and streamline growth. Very interesting. What's your initial impression? These graphics certainly stand out in their style, right? My initial reaction is like, I'm not just looking at this. I'm not exactly sure who is this for and what can I do with it? So it says about like build custom AI workflows and streamline curls.

10:26So in some ways, like, okay, like I can understand there's some kind of workflow, but for what like is it for designers is it for yeah um engineers is it for like youtubers there's something there's a there's a card about youtube there so and streamline growth is well there's all kinds of growth it's like revenue growth and user growth there's marketing and so i i would like probably like go a little more specific on on the on the copy here that there is some kind of templates i guess for different kinds of workflows it's kind of like a random list of different kinds of workflows so it would if it would be like meaningful to like useful to group them somehow that hey there's workflows for i don't know like youtubers or there's workflows for business people or some kind of analysts or something so there's a little bit more like oh okay i'm not a youtuber but like i'm an analyst so i could look into this yeah it seems like the who it's for is missing here the interesting thing for me is i paid so much attention to the graphics because they're so unique that I didn't even notice that it was like brand mood board generator, interview questions predictor, social media ads generator.

11:35Like I think they're trying to tell us some of the use cases that you could use it for, but my brain just didn't even notice those because the graphics just took all of my attention. Yeah, and I think like, well, there's this movement in the background and the movement on the cards, which I think can be nice, but it might not be the best thing to have this looping movement in the landing on the first screen because maybe what happened to you is like you're kind of kind of focused on the moment and like what's coming up next and like I think like the website should be kind of like this almost like this pyramid of like you start with this is what it is and then it's like it's very like as simple as possible and then you keep going down it's like expands and like okay well there's this thing and that thing and there's more and more and more things so like oh okay like I can that's interesting I will take a look but the first thing should be like very like kind of pointed or clear.

12:27Yeah. Yeah. There's like good animation, which draws your attention to important things that you want users to pay attention to and notice. And then there's distracting animation that maybe takes away attention from whatever you're trying to focus on. There's also this interesting element here, which is the prompt kind of like it's some kind of prompter, which is interesting, but like, I just wonder it's like kind of floating here and you might think it's, this looks like a cookie panner in a way, because it just kind of floats there. it's not it's it's like you kind of often often ignore that so I just wonder if if you just would put this prompter on the page here at the top and and it it could kind of like tell people yeah just like try it out and that way like you could communicate it better yeah it's so interesting like you're not designing in a vacuum of your page you're designing in the context of what users are used to and if users are used to ignoring cookie banners and things like that that tend to pop up at the bottom, then like maybe they're just ignore your thing there too.

13:25I mean, those kind of banners, like the UI itself, like this could work in the product itself. Like when you're, when you're looking at the, like a page or some kind of document, like it could be nice to have this, like a floating thing, but in the website context, it feels like it's some kind of banner that I don't have to pay attention to. Okay. So we scroll down a little bit, start building from a template. So these look like, I think the same kind of templates that we're scrolling across, maybe slightly different. And here we've got some more categories of types of things, business, marketing, AI generation, image generation.

13:58Okay, so that's interesting. I think one of the challenges when you have like a generic workflow builder or something that can do anything, it's difficult to communicate the specific things that people should use this for. And this seems like an interesting way to do that, to jog people's memory of like, what workflows do I need to solve? people probably have a hard time thinking about that, but just like, oh, do you need to create an infographic? Then, oh yeah, that's something I would use this for. Maybe my biggest question here is like, what is this workflow? Like, so it's not, so there's a lot of this, there's this promotion on the templates that you can have different kinds of templates, but I don't think the page exactly describes how does it work.

14:40Like it might be worthwhile trying to add some kind of almost like a flow chart, not really, but explain that what it, like, take one example and then, like, kind of, like, blow it up a little bit. Like, this is what a workflow could look like. Hey, there's, like, other workflows we created, but, like, I guess maybe you can create your own too. And so I think it's kind of highlighting the use case. Awesome. Well, thank you, Sprites AI. This was really interesting. Really appreciate you submitting your site. Okay, next up, we have Giga ML and they came to us with the question, our website conversion rate is low.

15:16How can we improve that? Well, I guess the first question is like, what are they trying to convert people to? And I see two buttons that say book a demo. So my hunch is that's the goal here. They're trying to get funneling people to the website. Maybe they're doing cold outreach. Maybe they're doing other types of marketing or something. And then they're trying to get people to book that demo. says AI agents for enterprise support trusted by some of the most admired companies in the world. And then there's a glowing blue orb here. I don't know, what's your first impression? In some ways, I think it's quite clean and I think it fits well for the market, which is an enterprise.

15:56And I would just say that the reason the conversion is slow because you are going after enterprises and it's like there's probably no way to like increase that conversion flow like the conversion rate really because enterprises are not necessarily coming to websites and clicking buttons that much I do think like if you do want to have more users or more people coming in you probably should have a way for them to try it out themselves and and like kind of drop the demo if you think that you still want to go full enterprise I think this is fine but then you I think you just have to find those enterprises somewhere else.

16:30Like you have to do sales or some kind of events or some kind of like dinners or some kind of ways to connect with the enterprise people because they are not often evaluating tools randomly. Maybe sometimes they hear about it. Like maybe you do some press or you maybe do something else. The conversion rate doesn't really matter because if you sell, if you do one deal, that's probably going to be like a huge deal. It is interesting. They do have a demo here. So this is a little small and it's hard to read this thing, but the animation did draw my attention to it. So let's see what happens when we click on this.

17:07Hey there, I'm Giga ML's voice AI agent ready to chat about anything. Do you have something in mind? Tell me about enterprise support. Our enterprise support is all about making life easier for big businesses. We offer seamless integrations with platforms like Amazon Connect, Genesis, and Salesforce. So everything just works together. Our voice AI handles massive call volumes without breaking a sweat. Okay. So a couple of interesting things here. First of all, when she started talking, she basically just said, what do you want to talk about? And I was like, I don't know. So it might be better to start off with a prompt that could lead people to what they want them to ask about, which could be features of the product, things like that.

17:56Another thing that I noticed is when she was describing the features of the product, she described it as SOC2 type 2, which most people know it's like SOC2 type 2 is how it's described. So just making sure that that's actually communicating the right thing. If I'm an enterprise user and, you know, one of the first things I do is talk to this thing to see how well it works and it gets things wrong or something, then I'm nervous that like, oh, maybe I don't want to use this product. So I think steering people towards the conversation you want them to have with it that's most going to upsell them would probably make them more likely to want to book a demo and use it in their own company.

18:33Yeah, I would say like on the Orb, it's hard to see the icon in it. It's kind of gray. So I kind of knew something was happening, but from the side here, I wasn't quite seeing the icon. So that might be one thing if people are not... That's something you could track if people actually click that and interact with it. I don't know if you need to worry about it too much, but just to see if it could be useful to improve it. And maybe, I don't know if this element is the best. Maybe there could be more like a box or something. is like maybe the box could have some prompts or something to like show the use cases.

19:09Prompts, yeah. Like kind of, well, enterprise customer support or enterprise. Like I think that the agent kind of described what maybe a lot of things they do. And I think, again, it goes to like, what are these people looking for? So if I'm an enterprise buyer and I don't know, the CEO tells me like, We need to reduce the cost and we want AI support. What do I type in? What do I support? AI phone support or something. And so kind of trying to think what is the language these people are using and then try to put that language here so you are aligning with what they're looking for. Yeah, I totally agree.

19:50The headline AI agents for enterprise support is very generic. I think the word enterprise is important. We clued in on that. But the sub headline there, trusted by some of the most admired companies in the world, is a missed opportunity to literally describe what the product does. And more about specifically enterprises, the big thing, like specifically who within enterprise should be the person that cares about this. And then, you know, they say trusted by the most admired companies in the world. And then they go right down and say trusted by these great companies. It's like you don't need both of these.

20:20So there's a missed opportunity here to communicate more. I mean, I would say like, I don't know if it's like super honest to say like these are the most admired companies in the world. Like I think it's good, but they obviously seems like they're big companies, but I would say like in my mind, there's other like, I don't know, Apple is very admired company or something. It's like, I don't know if that statement is exactly true. But this statement is true that these are probably like great companies. So I think that's fine. Yeah. And then coming down here, voice AI that is human. So it is voice tune model.

20:51you know it's all about voice stuff but like nothing up here says voice so that's probably a missed opportunity to really target who your early adopters are going to be my understanding with enterprise buyers is that they they need a lot of almost like i think you need need a lot of materials you need a lot of text you need to check a lot of boxes yeah you need a lot of customer stories like there's a lot of things they they want to see and maybe you should put those things in the front page, but maybe there's a specific category. Maybe you have a specific page about security. Maybe you have a specific page about something else.

21:26You could still, when you email them, you can link to those materials. And then once they're on the website, they can travel around. And so I think there is some benefit of adding things on the website. But I wouldn't consider this as a consumer startup where you are trying to optimize for the conversion rate. It's a very different kind of business. Awesome. Awesome. Thank you, GigaML. YC's next batch is now taking applications. Got a startup in you? Apply at ycombinator.com slash apply. It's never too early and filling out the app will level up your idea. Okay, back to the video. Okay, next up we have Unreal Milk.

22:04So this one seems very opinionated. We've got a cow that's chilling on a chair on my forever vacation. We got some hand-drawn clouds in the sky. It's interesting when I look at the cow, it's not immediately obvious what it's communicating. But when I think about it for a second, I'm like, okay, so the cow doesn't have to produce the milk all day, every day. And so we have milk that is not real milk, which is what the product is here, I'm assuming. It's definitely a different kind of category than I'm used to. I would say like from this screen, yeah, it's hard to tell like, well, what is it? Like, I mean, it's unreal milk.

22:41So that's the only information so far we have. so it's not milk but like what is it made of it's not real milk yeah so maybe we need to maybe we need to okay like so it has a side growing okay so then we move into to some some kind of use cases so you can make cheese and ice cream so we're still not sure what it is yeah but it's all good things kind of like they're pitching us on like why it's good yeah why it's better uh okay all right a lot of personality here mama we need you and the cow's crying and then our way yeah our way there's some kind of container where the milk comes from um yeah okay so there's some actual pictures okay we've got some scientists in a lab so yeah this kind of reads like a nice story and finally they did it they they created the whole cow milk in lab so okay so it's kind of like grown yeah grown milk yeah okay yeah there's a lot of personality even the copy like down here we're moving towards a more sustainable future this definitely has like a fun playful brand um seems like they don't take themselves too seriously it is interesting that you talked about um on store shelves because i feel like on store shelves you would expect a bottle that's more like these bright colors the bright yellows bright blues that kind of stuff but it actually feels very black and white if that's the actual container right there um i wonder if their website which honestly i think is pretty memorable um i think i would remember this later on the side scrolling is unique and memorable the graphics and illustrations are unique and memorable.

24:24This doesn't feel like it's copying anything that I've seen before. It feels like they, from first principles, came up with the concept of this, which is something that I think helps make it feel more memorable. A lot of times, a lot of websites are kind of cookie cutter, and it's almost a meme, you know, in the design community around, like, all websites look the same. And I don't know, got to give them a lot of credit for trying something different here to try to stand out and be. Yeah, and obviously, it's like, it's not a product you could sign up to. So it's something you have to buy. I wonder, okay, well, there's navigations.

24:57I wasn't expecting. I would not have clicked on. I would not have known that was a thing. I was like, well, what do I do now? So I was definitely looking for what is the next step? Or actually, okay, it doesn't really do anything. It just like scrolls you. Scrolls you to the thing, yeah. So what I would have wanted is that, okay, this is, in a way, it's an announcement. But, well, if I want to buy this, is that possible? Right. Right. Is it in stores? Where can I find it? Yeah. So I think it's like a nice announcement. I would consider how effective this will be. Maybe later they will have to think like how to actually explain this or like a little bit explain more.

25:36Right. And go deeper into some of these things. Right. Yeah, it's very interesting. There's no call to action that we could find. This is almost like an old school, like the early days of the Internet, people took the medium that already existed, like books and magazines, and just tried to like make that a website. and obviously websites have become way more interactive now but this feels like an old school like it's almost like a magazine you know display or ad or something like that yeah no maybe the last thing I would say like maybe just add some kind of call to action maybe just add like enter your email to to join like some kind of mailing list yeah just like people who like if people land here like even if that's not that many like people sign up you could start some kind of community maybe you have a lot of people like students signing up or you maybe have someone else signing up and you can start seeing this patterns of well who who are who is signing up and then you can ask them like why did you sign up like why is it interesting and then you can kind of like start the conversation yeah um i think like with these kind of things they seem to have this like a like a more like a broader social mission or kind of like a just cause that we should live in a world where we don't have to like use the animals this way right which is i think something like you could do more here is that, hey, join this cause.

26:49We could change the world. And then that's the call to action. Join the mailing list. We can, I don't know, send you some updates. We can talk about different things. And so people who are interested in that cause could join it. The font choice feels very opinionated too. I'm curious your thoughts on that and how you think about choosing fonts to communicate part of the brand. Yeah, I think a font or the typeface is a very big thing on any website because websites in the end are textual so there's text on websites right this this website doesn't have a lot of text i don't think like their font choice is good here given that the style of it it's more like a hand-drawn kind of environment in other websites i would just think that again like what is the what what is the world the brand is living in and like what kind of signaling you want to do sometimes you want to do the serifs for some of the like more like retro or older kind of feeling to it or sometimes like a lot of more modern kind of geometric typefaces it's more like the kind of like the standard of the day and that's like considered modern then there can be like more futuristic typefaces and it's not like you should just use one typeface everything because in small sizes a lot of typefaces don't work you you probably have to use something fairly standard so people can read it.

28:08But for headlines and special points, you can pair it with something more unique. Very cool. All right. So from the founders, they have a question. Does the website leave a positive impression and recall for the brand? Well, yes, I would say so. It felt very organic, I would say, which is interesting because their product is not literally organic. It's created in a lab. But it seems like what they're trying to communicate is that this is just as organic as real milk and not filled with a bunch of chemicals and it's good for you and stuff like that. And I feel like this font choice feels very organic, like hand-drawn, handwritten.

28:44The illustrations and the graphics feel that way. The colors are very colorful. So yeah, this feels very memorable. Like I feel like I will remember this site a week from now. I think the organic thing is interesting because technically this is a very like a scientific thing where you're creating something with with science yeah which is not organic per se um it's it's like you could go a direction of jurassic park or something and like you make it like super like scientific but they they they did like what i think is the right way is to to try to like battle that feeling i think people have a little bit uneasiness of manufactured foods.