Stitch Fix: Katrina Lake In 2010, Katrina Lake recruited 20 friends for an experiment: she wanted to see if she could choose clothes for them that accurately matched their style and personality. That idea sparked Stitch Fix, an online personal shopping service that aims to take the guesswork out of shopping. Today, it has over two million customers and brings in nearly a billion dollars in annual revenue. Plus, for our postscript "How You Built That", how Brian Sonia-Wallace built "Rent Poet" — a poem-on-demand service for weddings, corporate gatherings, and other events.
NPR logo

Stitch Fix: Katrina Lake

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Stitch Fix: Katrina Lake

KATRINA LAKE: At that seed stage, like, it was 100 percent just an idea. I would come in, and I would say, like, you know, here's the beta that I did with, like, 20 people, and here's how I think the business model will shape up and then probably talked to 30 angel investors who said no. And when you're doing something that nobody else is doing, you are either the smartest or the stupidest person in the room.



From NPR, it's HOW I BUILT THIS, a show about innovators, entrepreneurs, idealists and the stories behind the movements they built.

I'm Guy Raz. And on today's show, how Katrina Lake defied the investment world to build a tech company that transformed clothes shopping online.


RAZ: So every year, it seems there's always that one buzzword or idea that everyone in Silicon Valley seems to be talking about. A couple years ago, it was cryptocurrency. Recently, it's been blockchain. But there's also a widely held idea in the tech world that the days of retail stores are numbered. Now, of course, this is a combination of wishful thinking and hyperbole, but there is something to it, especially with big retailers. Macy's is struggling. Bon-Ton and Toys R Us have filed for bankruptcy.

And in October of 2017, Lord & Taylor sold its flagship building in Manhattan to WeWork. That amazing Italian Renaissance building on 5th Avenue is now worth more as a WeWork than as a department store. And the main reason why, of course, is that people just aren't going into big department stores. They're buying a lot more stuff online. Now, convenience is a major reason, of course, but so is time.

Time is increasingly limited. And that's really where Katrina Lake's idea begins because as she got older and busier with work, she found that she had less time to go shopping for clothes. And only rich people could afford personal shoppers, so she thought, hey, why not figure out a way to make that more affordable and less time-consuming? And that was the genesis of Stitch Fix, an online personal shopping service.

Now today, Stitch Fix has 2 million customers with annual revenue close to a billion dollars. And in 2017, Katrina became the youngest female CEO ever to take a company public, but it wasn't necessarily going to be a slam-dunk business. And in fact, Katrina had a hard time raising money in the beginning, but we'll get there.

She actually grew up in San Francisco in a bicultural home. Mom is Japanese. Dad is American. And at the age of 15, her parents decided to move the family from the Bay Area to Minnesota, which, if you've seen "Inside Out," is the exact opposite of what happens in that movie.

LAKE: Like, to go from - in San Francisco, it's a more diverse city. And, like, the high school that I went to was probably 70 percent Asian. And then to go to a school in Minneapolis where I think there were three Asian people - and 3 1/2 if you include me - in my class. And we were a very bilingual, bicultural household. And so we - my mom...

RAZ: You speak Japanese?

LAKE: Mhm.

RAZ: Oh, wow.

LAKE: My mom spoke only Japanese to me - still does - and my dad obviously spoke only English. And we celebrated a lot of Japanese holidays, and we would bring lunch to school. I think it was - we had kind of like a schedule. And three days a week, it would be onigiri, which is rice balls that have something in the middle. And then two days a week, it was sandwiches. So it was kind of (laughter) a blend.

RAZ: Yeah, yeah. Right.

Katrina went to Stanford for college and initially thought she was going to be a doctor like her dad, but she ended up majoring in economics. And after she graduated, she ended up working at a consulting firm, advising other businesses. She was 22 years old and still not really sure what she wanted to do.

LAKE: You know, honestly, at that point, I wasn't thinking that I wanted to do something entrepreneurial, but, like, I had all these ideas on what these businesses could do better. You know, this should be personalized. And, like, this is just a transaction. It should be an experience and likewise...

RAZ: Like, ideas in your head or ideas that you actually told people about?

LAKE: Both.

RAZ: Yeah.

LAKE: Yeah. Like, there was an example - we were working with a large retailer, and they actually said, like, that they wanted ideas of what's the future going to be like? And this was in 2006, and it was literally before the iPhone. And, like, I had this idea where I was like, why - you know, in a store, why do you have to have every single size and SKU on the floor? It seems silly that as - a person has to walk in and, like, find the size and SKU and then take it themselves to a fitting room and, like - I was like, how about this idea where you, like, have half of the store is a warehouse, a distribution center, and the other half is like a museum?

You walk around, and you see all these beautifully merchandised things, and you log in when you come in, and you get a wand, and then you can just put the wand in front of the outfits that you think you like. And then you walk into the back, and it's like hi, Guy, like, here's your fitting room. And all of the things that you wanded (ph) or whatever would be stacked up there. And then there would also be recommendations. And, you know, you could have one size up and one size down and, you know, it could just be, like, a much better experience than, like, going and weeding through racks to try to find your thing. And, like, people looked at me like I had seven heads.

RAZ: Yeah.

LAKE: And so, like, you know, in retrospect, I'm like, that was crazy. Like, if I was the CEO of that retailer, I'd be like, interesting idea, totally impractical. I mean...

RAZ: And who is this, like, 23-year-old telling me about this idea?

LAKE: Exactly (laughter).

RAZ: Right, yeah.

LAKE: And so, you know, I mean, I didn't share all of the ideas, and I wouldn't say that all of the ideas that I shared were well-received, but, like, I had tons of ideas about how things could be done better.

RAZ: So you worked for this consulting firm, I guess, for, like, two years, and then you decided to go work for a venture capital company. Why?

LAKE: You know, I look back on that, and I really left because I was just, like, really...

RAZ: Yeah.

LAKE: ...Unhappy with the amount of work. And so I took this job at a VC fund, which - like, I didn't actually want to be a venture capitalist. And I - you know, I didn't - it wasn't the industry that was appealing to me, but I was like, if I'm going to work at a place that's going to be the crazy retail idea of the future, like, this is a way I can meet the next thing.

RAZ: Oh, wow. So you actually thought, if I'm working here, I'm going to hear all of these pitches, and maybe I'll find a cool company and join it.

LAKE: Exactly...

RAZ: Oh, cool. That's smart.

LAKE: That is exactly what I thought, yes.

RAZ: Wow, that's smart. Yeah.

LAKE: And so - and I did, and I met hundreds of entrepreneurs and, like...

RAZ: And just to interrupt, were you a - like, a money person? Were you managing money, or were you on the analysis side? Were you in a room where these people would come and pitch, and you would throw, like, "Shark Tank" questions at them?

LAKE: I was in that room, but I was more of, like, the note taker (laughter). Like, I took the notes, and I would create investment memos that were, like, a summary of the opportunity. And I could have my opinion on those things for sure, but I wasn't an investor.

RAZ: Yeah.

LAKE: It was also - it was a weird time to be in venture because I was there from '07 to '09. You know, we had a...

RAZ: The crash.

LAKE: ...Little bit of a recession.

RAZ: Yeah.

LAKE: It was a weird time to be in venture. And I didn't ultimately find the company.

RAZ: You must have heard tons of ideas, though.

LAKE: Totally. I probably met over a hundred entrepreneurs in the time that I was there. And the most important learning, actually, was that all of these people were just, like, super-unqualified, normal people with lots of ideas just like I was (laughter).

RAZ: Yeah. They weren't, like, these superheroes with, like - they were just really good at, I guess, kind of like, you know, making the case for their idea.

LAKE: Yeah. And so I think it was - actually, the more powerful thing was that I realized that if I have all these ideas about what should be happening, like, I don't need to be in the peanut gallery, lobbing my ideas at people. Like, I should just do it myself - and that I could do it myself. But it still wasn't my, like, first choice. Like, my first choice would've still been, like, if I had met an entrepreneur who was amazing and compelling and creating the retail of the future, like, it was probably still my first choice to join somebody than to start it myself. But now, all of the sudden, I had this other plan, which was like, well, I guess if I don't find that, maybe there's a way that I could do it.

RAZ: So when you come to that conclusion that maybe you can do this yourself, what did you do?

LAKE: I don't know that it was, like, a moment. I think it just entered my mind as, like, this is another possibility. And I actually ultimately applied to business school, which...

RAZ: You decided to do that instead of just leaping into starting a business?

LAKE: Yeah.

RAZ: Why?

LAKE: And that wasn't even a choice. Like, it wasn't like I could start a business or go to business school.

RAZ: What do you mean it wasn't a choice?

LAKE: I was - like, my risk profile - I was never going to, like, quit my job and...

RAZ: Go away. Right.

LAKE: ...Take all this risk and have a gap on my resume. Like, that was just - that was never a choice in my mind. And so I was like, I can take these two years. I can be a mediocre student. And, you know, my end goal by the end of the two years is if I can have a company funded, if I can be paying myself a salary, if I can be paying back my student loans the day I graduate, then the risk profile of entrepreneurship was tenable to me. And then my plan B would be, well, if I can't do that, I'll have this MBA from Harvard, and I'll be able to join a great company and a great founder and do something interesting.

RAZ: So you get to Harvard as a business student in 2009. And did you like it?

LAKE: I did. It was...

RAZ: Because it's kind of an intense place, right?

LAKE: It was very different from maybe all of the experiences I had had.

RAZ: I mean, we've heard these stories from - especially from professors there, particularly women, who've talked about how in the classroom, the discussions tend to be - you know, the hands that go up are men or the voices that you hear mostly are men. Did you find that? And was it a place where you felt comfortable raising your hand and talking?

LAKE: I did - I mean, I did. You know, I think that stereotype is probably - probably was true. But I did feel comfortable talking. I think - you know, and I had a hard time disentangling, you know, is it gender? Is it being from California and saying like? Like, what were the things that created this to be true? But, like, I did feel a little bit underestimated, and I really tried to actively strip like out of my vocabulary. You know, people say that even, like, a southern accent makes people think a certain thing. And I felt sounding like I was from California also did the same thing. And so, you know, I didn't feel like I was the - I don't know - most likely to succeed in the classroom.

RAZ: Yeah. Tell me how the idea for what would become Stitch Fix - which we'll talk about obviously - but what's the early idea that came to you? How did it even begin? Were you given an assignment? Were you told to analyze a problem? Like, what's the beginning of this?

LAKE: So I had a bunch of - I kind of collected a bunch of theses and ideas in my head. If you look at industries - and, you know, I probably had, like, a little bit of an academic approach in how I looked at these things. But, like, the hunting and fishing industry - it's - I know it's crazy - it's a huge industry. It's massive. People are super passionate about it. It's really fragmented but also concentrated, where it's, like, really mom and pops, or else there's, like, Orvis and - anyway, it's a super interesting industry because it's massive. And there's elements of fashion, actually, where it's like - whether it's technology or actual fashion, things go out of style. And so I rented a car, and I drove to New Hampshire and, like, visited, like, two gun shops. And, like, the second - by the second one, I was like I just - you know...

RAZ: You were going to disrupt the whole gun industry.

LAKE: I just was like, I don't care how big of an industry this is, like, I'm just not super passionate about it. And, also, like, I am not the person to, like, disrupt this industry. And so...

RAZ: But who knew? There must still be an opportunity to, you know, disrupt the hunting and fishing industry.

LAKE: I think there probably is. But this is all to say, like, I was actively exploring a couple of ideas, that I'm like, this is a real thing, and it has a reason to belong in the world. And the idea of Stitch Fix was one that I also tested and explored. And the way I did that was...

RAZ: What was your idea?

LAKE: The idea was, how can you - I mean, there are couple of different theses. So one thesis, overall, was just, like, more and more dollars are moving to e-commerce stores. Like, this was the math that I found was super compelling for my time in consulting - was dollars are leaving stores and going to e-commerce. As dollars gradually leave, every one of those dollars is hugely hurtful to a store as it loses it because, all of a sudden, you can't pay your rent. You can't pay your people. And so, you know, like, a 10 percent, 15 percent decline in your store volume can kill a store. And 10 or 15 percent doesn't seem like all that much. And so there's this kind of, like, death spiral that I kind of anticipated would happen with stores.

And Netflix and Blockbuster is kind of like the best example of it, of like...

RAZ: Right. Yeah.

LAKE: Even if 30 percent of the people in your town are using Netflix, and 70 percent of the people in your town are going to Blockbuster, like, that Blockbuster probably can't make money.

RAZ: I mean, the writing's on the wall at that point.

LAKE: Right. And then the death spiral component of it is, like, that one Blockbuster closes, and now the 70 percent of the people who lived in that town have to decide, do I drive 10 miles to another Blockbuster, or do I try this Netflix thing that 30 percent - the minority of people in my town are doing?

RAZ: That was a - probably a big conversation at that time because Netflix was killing Blockbusters at the time.

LAKE: Exactly. Yes, it's fascinating. So I expected that would happen in retail apparel stores. And then on top of that, I felt like there was this massive depersonalization of retail that was happening, which, when you're buying toilet paper, iPhone chargers or diapers is fine. You don't need that to be personal. But that totally ignored this category of apparel, which is, like, deeply personal and deeply nuanced and are product categories that are super emotional. Like, you're not kind of emotionally buying a box of Kleenex, but, like, you are very emotionally buying that blazer.

RAZ: Right.

LAKE: And so I felt like all of the emotion and nuance had been stripped out in e-commerce and that that wasn't the right approach.

RAZ: Yeah.

LAKE: So the idea was, how can you deliver a really - actually, a personal experience in apparel and use data and technology to make that scalable and to make that better?

RAZ: But why - I mean, why wouldn't you just, you know, offer up a bunch of options and just let people kind of scroll through - I mean, I guess the beauty of online shopping, certainly at that time, was, you know, just infinite choices, right? You could just scroll through and find the thing you like.

LAKE: Yeah. And, you know, I think for some customers, you know, that's a way that they want to shop. But, like, I'm 5-foot-2, and so there's a whole world of apparel that's, like, literally not relevant to me just simply because they're not going to fit me because I'm the wrong height. And there are things that - you know, if you're living in California, you can go on a home page, and they're trying to sell you cashmere mittens. And it's like, that's totally irrelevant to me.

So if you peel back, like, what Stitch Fix does, it's like - yes, like, if you could, instead of going to a mall all day or instead of spending, you know, your entire evening with 30 browser tabs open to try to compare and contrast different jeans you want to buy, if you could get a box sent to you that had two pairs of jeans, or even better, one pair of jeans, and they fit you great, that is an infinitely better experience than the other alternatives out there.

RAZ: OK. So you have this idea about, like, a business where, I guess, you would send a subscription box to - of clothes to people, but how did you - like, how did you even begin to think of - about how it was going to work?

LAKE: So the way it worked was, basically, I would have friends of friends fill out a survey, and you would let me know what size and style, what things you liked, what brands you like, all that kind of stuff...

RAZ: I mean, you're just experimenting at this point, or had you already formed a company and named it?

LAKE: No. This was in 2010. I was experimenting.

RAZ: As a student, essentially, writing to other friends and students in...

LAKE: In Boston.

RAZ: ...In Boston.

LAKE: And that - and that was the benefit of being in Boston, too. It was like, I didn't really have that many friends in Boston. And so it was easy to meet strangers. And so one of the things that I needed to prove to myself was, like, it wasn't going to be scalable for you to tell me, OK, these are the seven things I want and then for me to go buy those things. Like, the math doesn't work for that to be scalable. And so you need to have a set of inventory you're working from.

And so I was buying inventory at retail from boutiques in Boston. And I would keep track of, like, do they have a 14-day return policy or a 30-day return policy? And then I would use that - I wasn't making any money doing this, but I would use that as my inventory so that when somebody would fill out their profile, I would then put together a box of things that were relevant based on what they shared in their profile. And then they would try things on. And then if they wanted to buy things, they would write me a personal check, and I would sell it to them. And if they didn't want to keep it, and if I ultimately didn't sell it within my 14-day return window, I'd go and return it to the boutique.

RAZ: By the way, how did you buy all that stuff?

LAKE: I had a credit card with a $6,000 limit on it.

RAZ: Oh, nice. And you would buy them and then sell them at a markup.

LAKE: No, I was just selling them at the price that I bought them.

RAZ: At the price. OK.

LAKE: I wasn't making any money. I just wanted to, like...

RAZ: Just to test this out.

LAKE: ...See if it would work. Yeah.

RAZ: Wow. And how many friends were in on this at the beginning?

LAKE: I mean, not - I don't know. Twenty, maybe.

RAZ: Twenty. Oh, wow.

LAKE: Ten or 20.

RAZ: Wow. What did you find from this group of 20 people? What did you discover?

LAKE: I discovered that you actually could learn a lot from somebody. I think there is just this thesis around, like, do I have to, like, be with you in person to understand your style and who you are, or can I do that through a survey? And, like, I felt pretty confident, like, you could do it through a survey.

There was a number-of-items thing. I started out with 10 items. And the reality was, like, people weren't going to try on 10 items. It's a lot. It's a lot of work. And you just look at something, and you're like, no. There's no way I'm going to like that. And you just don't try everything on. And so 10 was too many.

What else did I learn? I think I learned the brand thing of like it didn't - you know, people were actually surprisingly very open to buying from brands that they weren't aware of and, in fact, found some value in finding brands that they wouldn't have normally have found on their own.

RAZ: So when did you pull the trigger and say all right, let's - let me form a business? Let's get an LLC going? Let's, like, make this real?

LAKE: It was - we didn't actually - the company wasn't formed until February of 2011, but I kind of became all in on this and not other concepts, I would say, like, late 2010.

RAZ: So when you - when you really kind of, like, got this going - and, at that time, it was not just you, right? It wasn't - you were not the only one involved in the company.

LAKE: Yeah, I had a co-founder at the time. She was the wife of my friend from college.

RAZ: And so how are you running this business and also being a student?

LAKE: I scheduled my classes so that they were a lot - they were all lumped together in the middle - either in the middle or the end of the week. Or I did something with my scheduling that enabled me to be able to spend almost a week in California. And so in January and, then again in February, I spent almost a full week out here. And what I would do out here was meet with angel investors. And on Valentine's Day of 2011, we got a term sheet from Steve Anderson.

RAZ: Steve Anderson is an investor that you met through other people. And he gave you - how much money did he give you?

LAKE: He gave a term sheet for a half-million dollars...

RAZ: Wow.

LAKE: ...And then said now go talk to other angel investors and try to find others to join the round. And then probably talked to 30 other people who said no, ultimately, even though he had already said yes.

RAZ: Yeah. I mean, in that - in the sort of winter and spring of 2011 when you - you took the research that you'd gathered from these 20 people, and you went to meet with investors in the Silicon Valley. Did you just - you showed them this data, and you said, see? I tried this on 20 friends, and this is what we found, right? I mean, you can understand - you could understand, I guess, why they might have been skeptical at that time...

LAKE: Oh, totally.

RAZ: ...Right?

LAKE: Yeah. No. I would come in, and I would say like, you know, here's the beta that I did with, like, 20 people. And here's how I think the business model will shape up. And, like, I would bring a box in, and I'd say, like, here's an example of a box with, like, laminated cards. And, you know, the box wasn't, like, all that nice looking. Like, it's not like I had spent any money on branding or anything. And at that seed stage, like, it was 100 percent just an idea.

RAZ: And explain how you explained what the business model was at that time because I - just to clarify, Stitch Fix today - you sign up. You pay 20 bucks a month. You get five items sent to you every month depending on your subscription. And if you buy them all, you get a discount. And if you return them all, that's fine. It's free, but it's $20 a month, essentially, for a personal stylist, right?

LAKE: Yeah. And it's $20 - and, to be very clear, like, you don't have to do it on a monthly basis. You could do it quarterly. You can do it a la carte. Like, you don't - you know, you don't need to commit to any timeframe. But it is every time you get a fix, you pay $20 styling fee. The $20 is credit towards anything that you choose to buy. But the heart of it was really this, like, we're going to buy inventory. We're going to get to know you. We're going to become experts on you, clients, and we're going to become experts on the products. And then we're going to generate really great recommendations...

RAZ: Yeah.

LAKE: ...At the intersection of those two things.

RAZ: But buying inventory - that's a big investment, right?

LAKE: Right. At Stitch Fix, like, what we did, and what we did then and what we do now is like - we are like a department store. We buy...

RAZ: You bought a bunch of stuff.

LAKE: ...Wholesale.

RAZ: And you keep it in warehouses.

LAKE: And we have the - yeah, exactly. And so when I would go and raise a half-million dollars from Steve Anderson, like, half of that was literally going to buy...

RAZ: Clothes.

LAKE: ...Dresses and clothes. And venture investors hated that.

RAZ: And not only would they have hated that, but you - if, in your business model, you said we're going to personalize this. We're going to have human beings involved in picking stuff - I mean, venture people - they don't like to pay people. Like, they like companies that are like - they like user-generated content, or they like...

LAKE: Yeah, totally.

RAZ: ...They like stuff that's free. They don't like to pay for human resources, right?

LAKE: Right. Yeah. People said, like, this is going to be an inventory nightmare. Like - and it was good feedback because, ultimately, inventory - like, we are very good at managing inventory, and we have to be in our model. And, you know, people saw that. And people were skeptical that you'd be able to execute.

RAZ: Over that period of time, when you're trying to raise money, and all - everybody, essentially, except for Steve, said no, was that demoralizing? Was it disheartening? Or were you just like, yeah, whatever, I'm - you know, water off a duck's back?

LAKE: I don't know. It was both. Like, you just - you had to be able to have reason for confidence in what you are doing, regardless of what venture investors thought at the moment. And so, like, you did have to have thick skin about it. And you did have to let it be water on a duck's back. But at the same time, what we were doing - like, nobody else was really doing it for, like, a very long time. And when you're doing something that nobody else is doing, you are either the smartest or the stupidest person in the room.


RAZ: When we come back, why, despite early growth, the money men in Silicon Valley didn't want to give Stitch Fix the valuation Katrina thought it deserved and how she fought back. Stay with us. I'm Guy Raz. And you're listening to HOW I BUILT THIS from NPR.


RAZ: Hey. Welcome back to HOW I BUILT THIS from NPR. I'm Guy Raz. So it's the spring of 2011, and with the seed money she got from just one single investor, Katrina Lake got a small office in San Francisco and got to work.

LAKE: Yeah. In February, we got the term sheet. In April, it funded. In May, I graduated. And then in June, moved out to San Francisco in 1,100 square feet.

RAZ: And San Francisco was important to do it because this is where you wanted to live or because of the tech industry or...

LAKE: Honestly, you know, the smart way to talk about it is the tech industry. But honestly, like, when I first was starting the company, I was like, OK, we have to be in New York since this is where fashion is.

RAZ: Right.

LAKE: And then, like, a couple of months in, I realized there's no reason we have to be in New York. Lots of brands are in LA. Lots of - you know, lots of companies are here. Like, I think it would have been really hard to do what we did anywhere else because we have many, many data scientists and many, many engineers. And, you know, in retrospect, it was not just about being comfortable. But if we wanted to fulfill on the vision of using technology to deliver our service, like, it would have been very difficult to do it elsewhere.

RAZ: Were you able to build a website with that money?

LAKE: No, not yet. Yeah. So the first seven months of Stitch Fix - like, we didn't have a website. Actually, the first name was Rack Habit, which lasted for about three months before we got - it was very clear that Nordstrom would have sued us for it...

RAZ: Because of Nordstrom Rack, right?

LAKE: ...If we used that name. Yes. Exactly. The lawyer took, like, 5 seconds of looking at it - was like, this - like, you got to change your name, which was fine. I hated the name...

RAZ: Not a good name. Not a good name.

LAKE: ...It was a terrible name.

RAZ: Very smart, yes.

OK. So you start to call it Stitch Fix. You don't have a fully formed website yet. And what did your first page look like? Like, how were they interacting...

LAKE: So you would go to And it said coming soon - do you want to be part of the beta?

RAZ: I see.

LAKE: And then it said - if you clicked on yes, I want you part of the beta, it would click you over to - I forget if it was a Google Doc or a SurveyMonkey - but a something - and then you fill out all your information, and you say you're interested. And we actually had a waitlist for a long time because we didn't have clothes to send to you that were relevant, and demand outpaced kind of what we had for a long time.

And so once I was ready for you, I would send you a link that said, like, you know, Stitch Fix is ready for you. And then I would send you a PayPal request for $20. And so this history of styling fee that we have was that I was worried that you could be a bot or somebody who's going to steal money or steal my clothes. And so I felt like if you were willing to PayPal me $20 before I've even sent you anything, you are probably a legitimate human being who wasn't going to steal my clothes.

RAZ: You could take the risk.

LAKE: Right. And so I would PayPal request you for $20. You'd give me $20. And then I would send you a shipment of five items. And inside, there was, like, a paper return survey because we still didn't have a website. And so it was a piece of paper that was, like, an Excel thing that had all the items. And then you would fill out - with words - what you were going to keep, what you weren't going to keep and explaining why. And then we would have TaskRabbits who would type that into Excel so that we had the data...

RAZ: You just hire people of TaskRabbit, like, on an hourly basis.

LAKE: ...Yep - who would, like, type in that information so that we had it. And then based on what you were keeping and not keeping, a net of your $20 styling fee, I would send you a PayPal invoice for the net. And that was kind of how it worked in the beginning.

RAZ: Where were you getting the clothes from?

LAKE: We were buying it wholesale. And so we were buying six items at a time.

RAZ: How did you find the wholesalers?

LAKE: We would go to, like - there are these shows you can go to and you can go introduce yourself to brands. And, you know, most of the brands weren't that interested in working with us. But, you know, some would give us a shot and say like, I don't know, that's an interesting thing. Sure. You can buy six items from us. And literally, we would buy six of one thing at a time.

RAZ: So initially, your brands were not well-known.

LAKE: Yeah. Definitely.

RAZ: And who was picking the clothes? How did you - I mean, obviously, you know how to dress well. You are well-dressed right here. But I mean, that's a big responsibility to, like, be the personal stylist of, you know, even 100 people, even 30 people.

LAKE: Yeah. And, I mean, it was - we hired our first stylist in - I'm trying to think - like, February of 2012.

RAZ: So the first, like, year, it's just - it's like...

LAKE: It was probably me. Yeah...

RAZ: And who was packing...

LAKE: ...It was me or my co-founder at the time. And we were the ones packing the boxes, too.

RAZ: You were packing the boxes.

LAKE: Yeah.

RAZ: So you are in San Francisco. And it's - you know, you're essentially doing everything. And you've got this website. And how are you getting the word out?

LAKE: It was literally just people telling other people. And so it was 29. Then it was 35. Then it was 110. Then it was 235. And then it just, like...

RAZ: Over that summer of 2011?

LAKE: Yeah. Yeah. Those are the actual cohort numbers for each month. And then I think after that, it started getting the attention of what I would call, like, small- to medium-sized influencers.

RAZ: It wasn't in Vogue yet.

LAKE: Right. It certainly wasn't in Vogue. And it wasn't, like, the bloggers that have, like, millions of followers. But it was people who have, like, 50,000 followers or 100,000 followers.

RAZ: Would you email them about it?

LAKE: Yeah. I mean, if they - I would. And that was a really big driver for us for a long time.

RAZ: So now you're starting to get some momentum. And obviously, you're growing. You need more money. You need millions of dollars. When you would show up seeking money, was it challenging? I mean, let's be honest. I mean, the vast majority of people in tech are men. And in this space, you know, going to VCs - they're probably used to seeing mostly men. Was a gender dynamic ever something that was present, that you felt, that you felt worked against you, that made it harder to raise money or generate interest?

LAKE: Raising money was really hard. And I think there were a lot of factors that contributed to it. I mean, certainly, the inventory side - like, that's real. Like, Silicon Valley venture investors want to buy software engineers, or they want to buy apps. Like, they don't want to buy clothes. The gender dynamic was real, also. You know, 94 percent of venture investors are men. And probably, the majority of men don't feel passionate about dressing women.

RAZ: I'm curious, you know, with respect to your experience with venture capitalists because there's a mythology about them and the genius of venture capitalists and, you know, how they take great risks. But, you know, they can invest in hundreds or thousands of failures, and they just need one or two to, like, come out and look like geniuses.

LAKE: Yeah. That's definitely true.

RAZ: There's this like mythology of the oracle, you know, like, this oracle venture capitalist who's just, you know, gifted and can just pick winners. But I don't think that really matches with reality.

LAKE: Yeah. And it is - like, it creates another dynamic, too, where - and I don't think I realized this at the time - but you're right, like, they want billion-dollar businesses. And they would either rather - they want you to be a zero or a billion. And everything in the middle is not that attractive. And so people were like, this is a real business. And it's really working. But, you know, I don't know if it can be more than a $30 million revenue or more than, like, $100 million in revenue. Like, I just don't think it's that big. And, like, to a normal person...

RAZ: That's insane.

LAKE: Thirty million dollars or $100 million is insane.

RAZ: Insane, yeah.

LAKE: But, like - but from a venture capitalist perspective, like, they make the most money when it's a billion or zero. And so, you know, there are a lot of people who are like, hey, what you're doing is traction. And it's real. And it's working. But like, I don't think it can be big enough for me to be interested.

RAZ: Yeah. But, I mean, you still raised a fair amount of money - I mean, like, $42 million in, you know, in the beginning, which sounds pretty good. But I guess you still needed to keep at it?

LAKE: Yeah. I mean, once we had a little bit of traction, in some weird ways, it made it harder because now all of the sudden, people are, like, you know, trying to put that billion-dollar price tag on you and trying to figure out, like, you know, is this big enough? And so in the middle, it was really hard. And...

RAZ: In the middle to raise money?

LAKE: ...In the middle, yeah, to raise money. It was really hard...

RAZ: Even though you were profitable or on the way to profitability?

LAKE: Yeah. And I hated it. Like, I hated raising money. I didn't feel good at it. Like, it wasn't my intention when I started the company, let me do this with as little capital as I can. But after a couple of experiences with fundraising, I was like, I want to do this as little as I possibly can.

RAZ: But meanwhile, the company is growing. And I guess we should mention here that you guys employ a lot of data scientists, like, a huge number.

LAKE: Yeah.

RAZ: So at what point did you just sort of become like a data company, a data tech company, and not just a fashion company?

LAKE: I mean, honestly, from day zero, like, we were always a data company. Like, even - like those return surveys that I described - like even when we were physically shipping the fixes out of our office, and we didn't even have a website, we were collecting the data about the product, the attributes of the product. And we were collecting, physically with pen and paper, your feedback about what you liked and what you didn't like and what you thought of the price and how it fit. And so the data element was always super important. But, you know, I don't think we were employing true data science, I would say, until - I mean, I don't know if that's fair, though, because Eric - Eric joined us in 2012...

RAZ: Eric?

LAKE: ...Eric Colson, who...

RAZ: He was at Netflix before.

LAKE: He was at Netflix.

RAZ: And he was a data guy at Netflix. And you're thinking, wow, I mean, that doesn't get better than Netflix. They know what you want to watch...

LAKE: Right.

RAZ: ...Before you even grab the remote control.

LAKE: Right. And especially at the time - this was before they were doing original content - like, their whole thing was algorithms and recommendations. And he was like, at Netflix, we show you, like, a hundred recommendations at once. And he was like, we were just having this conversation where, like, if we were really bold, we would only show you five or three. Or if we were really bold, when you opened Netflix, we would just start playing the content that you'd want to play. And he was like, that's kind of what you're doing here with clothes. And so he was very attracted to the fact that recommendations were, like, actually at the core of what we are doing.

RAZ: Were you able to describe your company in shorthand, or did you ever describe it as sort of like a Netflix for clothes?

LAKE: I don't know that I really used that, honestly. But, you know, I think there are elements. Like, we were - 100 percent of our revenue is tied to recommendations. Like, that's a very compelling thing to a data scientist who loves recommendations, feeling like that's going to be at the core of what we're doing.

RAZ: At what point did you become profitable?

LAKE: We turned profitable - we had a profitable quarter, I think, in 2014.

RAZ: So it took you three years to actually become profitable, which is super fast.

LAKE: Yeah.

RAZ: How much do you think about competition - right? - because there - I mean, there are all kinds of potential competitors out there, the obvious one being Amazon. Just it's massive, right? And they lay their fingers in every pie, you know? What's the sort of like the killer app or like the thing that you guys have that makes you sustainable and resistant to big challenges?

LAKE: Yeah. I mean, we - I would say that we pay attention to but don't obsess over competition. And I think when we think about what's special about Stitch Fix, it's the one-to-one human connections and the recommendations at the heart of it. And so, you know, some of the things that are really hard in today's e-commerce are, like, if you want to find a pair of jeans that fit you, like, it's not a great user experience to have literally a million pairs of jeans to choose from...

RAZ: Yeah.

LAKE: ...And try to filter and sort through those. And, you know, that's an example of a place where, like, the human element of like, you know, these jeans have stretch or these jeans don't have stretch. And these are great for people who are your height. And we know that with data. Like, all of those attributes are hugely important in helping you find which one of those millions of pairs of jeans are the right ones for you. And so, you know, there's a transacting part of e-commerce. But before the transaction, there's the discovery of, like, what is the thing that you're going to love? And that is a really hard problem to solve in today's e-commerce and one that I think we are uniquely good at solving.

RAZ: How many people, by the way, work for Stitch Fix today?

LAKE: We have over 5,000 employees.

RAZ: Wow. Wow. And how do the stylists work? Because you've got this sort of hybrid model - right? - where all these humans are kind of collaborating with algorithms, right?

LAKE: Yeah, we have over 3,000 stylists. And the way Stitch Fix works with the integration of humans and data is that they're both very kind of richly combined. And we see it as, like, they both are uniquely good at different things. And so, like, a stylist should never have to worry about, like, do you hate yellow or love yellow? Like, the data could take care of that. And so the...

RAZ: I answered I hate yellow, by the way, in my Stitch Fix. Yellow and salmon.

LAKE: ...And salmon (laughter).

RAZ: I'm not going to wear salmon. Yeah.

LAKE: The - I mean, the stylist can see that, but the data is going to exclude that from being able to be even available to your stylist. And so the stylist will layer on his or her own judgment over those recommendations to choose the right things for you. And so in some cases, the data science is going to be really accurate. And, like, trying to figure out which jeans are best for you, for example, is a place where our data science is enormously helpful. And then there are other times when the stylist is layering on human judgment.

RAZ: You took this company public last November.

LAKE: Yeah.

RAZ: And, again, like, everyone - you know, people are talking about the first - right? - youngest female CEO to ever take a company public. You know, probably this year, you will appear on Forbes list of wealthiest self-made women. As you became more and more prominent, and the company became more profitable, and you started to get more attention, all of a sudden, you were sort of a successful woman and tech kind of symbol. Did that irritate you, or did you embrace it initially? Or did - how did you feel about that?

LAKE: You know, that's an interesting one where I think my mind has changed, maybe even in the last year or two years. At first, I was really resistant to it. Like, I really didn't like being a successful female CEO. Like, I was like I just want to be a successful CEO. And, you know, I think what's changed about it now is that when I wrote my application essay to HBS, like, I wrote about Meg Whitman, a female CEO.

RAZ: Right.

LAKE: And, like, there were so few people. Like, I still very much admire her. And I did admire her then. But there were very few people for me to choose from in saying, like, who I want to be like. And now the - being able to be part of what opens people's lens of possibility more, I think, is really powerful. And so I think now, like, I'm - I don't know - kind of proud to be able to be part of that. And, like, I'm very happy to wear that label. And I'm very - you know, I think it's good for people to be able to see more examples of what success can look like that maybe are different from what it's looked like in the past.

RAZ: There was a story - I don't know when it came out, maybe a year or so ago - that at a certain point in your journey to raise money, you experienced harassment from a prominent partner at a prominent venture capital firm. They released a statement eventually saying something like, you know, we should have done better or something like that. Was that a common experience for you?

LAKE: Well, I can't speak to that specific example, but, you know, I think that there were real challenges with the power dynamic. I think of, you know, who has the power and who has the money and who doesn't.

RAZ: Was it mostly you standing in front of a room of men?

LAKE: Yeah, that's mostly true. Yeah. And I think that there are elements of it that just feel structurally unfair that I don't think get solved unless you change the proportion of men and women that are in leadership roles in venture capital. And so, you know - I mean, there are examples where, like, I was at a conference, and Chris Sacca was on stage. Like, there are hundreds of people in the room. Like, this isn't a secret. He was saying this publicly.

RAZ: And Chris Sacca - he's a venture capitalist, right?

LAKE: Yeah, and, like, one of the things he was saying was he was like look, like, I'm, you know, I'm not a typical venture capitalist. Like, I don't have an office in San Francisco. Like, I just - I live up in Tahoe. And, like, you know entrepreneurs - they come to me. And like, they come and hang out with me in Tahoe. And they, like, drink beers in my hot tub. And, like, I get to know them and their businesses. And, like, you know, there's part of that you're like, oh, cool. Like, that's different. But, like, if you think about it, like, I'm sitting in the audience.

And, like, I think I was pregnant at the time. And I'm, like, sitting in the audience. I'm pregnant. I'm like, I'm not going to come meet you in a bikini. And I can't even come into your hot tub because I'm pregnant, and I can't drink your beers because I'm pregnant. And, like, you know, there's just all of these structural elements that, like, just make it harder and more challenging. And, like, you know, you just realize, like, well, I guess I'm never going to be able to be in his pipeline. And, like, that's OK. But, like, it's just their fewer - it's just challenging.

RAZ: You know - and there's like a tone deafness, right? And I'm wondering if since, you know, more and more of this stuff is coming out and more people are talking about it - and we've had Jenn Hyman on the show, and she's talked about her experiences with Rent The Runway - whether some of that is trickling down in a way that's starting to change behavior. I don't know. Do you - are you hearing that in any way?

LAKE: I certainly hope it is. And I don't want to make it seem like everything is super unfair and that everybody's a creep. Like, that's really not the case. I've had great partnerships with venture investors. And by and large, I think people have good intentions. I do genuinely hope that it's getting better and that people are at least more aware. And so, you know, my hope is at least being able to see the success stories creates another pattern recognition for venture investors.

RAZ: I just want to get back to something that you alluded to earlier, which is that in the midst of all this - like, in the midst of building this company - you got married and had a child, too.

LAKE: Yeah.

RAZ: You became a mom. And how old is your kid now?

LAKE: He is 17 months.

RAZ: I mean, that's a lot to balance.

LAKE: Yeah, it's a lot. And it is - I don't know that I ever actively was thinking, like, I have this company and, like, I also need to have my life. And, you know, I - but it was. Like, I - in the time of the company, I met my husband. I got married. I had a baby. Like, you know, I've gone through a lot of life transitions through the company. And I think it's really healthy because I think, you know, in a lot of ways, your company feels like your kind of baby and your only thing. And, you know, I think I was - I don't know - lucky, fortunate. Like, you know whatever it is to be able to have had the space in my life to also be able to do - you know, to have those milestones.

RAZ: To be a mom, yeah. So I should admit that I get knocked sometimes from some of our listeners, particularly women, who listen who say, you know, you don't ask men about their children as much as you ask women. And so I've been more mindful about doing that. I think the one difference, though, is that, you know, as a woman, you are carrying the child for nine months and running the company. I mean, that's very different. That is a - that's a really - that's a challenge. I mean, physically, it's challenging. And then you also have to put your emotional energy into your business.

LAKE: Yeah.

RAZ: Right?

LAKE: You know, it's - and it's different. You know, like you show up at meetings, and you're, like, eight months pregnant.

RAZ: Yeah.

LAKE: And, like, you know people didn't really know how to react to that...

RAZ: Yeah.

LAKE: ...And, like, had a hard time navigating it. And I don't blame them because it's, like - it's different.

RAZ: Yeah.

LAKE: And, like, I took a 16-week maternity leave. I was out of the office for 16 weeks.

RAZ: Wow.

LAKE: And my...

RAZ: You were really unplugged?

LAKE: Yeah. I mean, I - there are a couple of things that I did. I did some recruiting things. Like, there are some things that as CEO, like, only I could do. But, like, I pretty much didn't set foot in the office for 16 weeks. And, you know, that's the example that - I have a very female employee base. Like, you know, if I'm not honoring the maternity leave policy that we have, like, how are people going to feel comfortable? Like, you know, feeling like it's the right thing to do...

RAZ: Totally. Yeah.

LAKE: ...To take 16 weeks.

RAZ: You know, there's something that I want to ask you about, which is that pretty early on in building this company, you parted ways with your co-founder. What's the story? I mean, did you guys just have a different sense of what you wanted to do with the company?

LAKE: Yeah, so my co-founder and I - we parted ways in the summer of 2012. And it was - I mean, I - you know, know there's probably not a lot of specifics I can share on it. But, like, it was really hard. And it - like, in a divorce you talk about, like, irreconcilable differences.

RAZ: Yeah.

LAKE: And, like, I feel like through that, I really understood what that meant. Like, if you can be in a place where you really deeply care for somebody, and at the same time you see things so differently, and you can't - you just can't see the future in the same way.

RAZ: Do you think that people who start up businesses are better off doing it on their own?

LAKE: No? I don't know. I mean, starting a company is such a team sport. And, you know, I look back, and I could not have done it without the many employees who believed in it, the many clients who believed in it, the - like, my friends who - I didn't go to their wedding because I was like too poor and too busy at the time. And like...

RAZ: Yeah.

LAKE: I mean, there's just like - it's such a team sport, and you have to surround yourself with people who help you to achieve those goals. And, you know, does that mean you have to have a co-founder or not? Like, I don't know.

RAZ: Yeah.

LAKE: But, like, I couldn't imagine having done it without having a network of people who believe.

RAZ: Katrina Lake - founder of Stitch Fix, the company which, by the way, is now valued at over $2 billion. And on the day the company actually went public, Katrina had some help from her 1-year-old. He was ringing the NASDAQ bell while she was holding him.

LAKE: He was holding this stuffed animal, and I was holding him during the dress rehearsal - like practicing the - you know, going through the motions. And he's like really chill and happy to be up there. And I was kind of like, all right. We're going do it this way. He's going to be up here, and this is how we are going to do it and....

RAZ: I thought you were going to say he spit up all over me in the dress rehearsal.

LAKE: (Laughter) I mean, that's the funny thing - is, like, if I had planned to have him up there, there's no way it would have worked. He's like - he's a year old or whatever.

RAZ: Yeah.

LAKE: Like, he wasn't going to cooperate. But the stars aligned, and it worked.


RAZ: And stick around because in just a moment, we're going to hear from you about the things you're building.


RAZ: Hey. Thanks for sticking around because it's time now for How You Built That.


RAZ: This is Brian Sonia-Wallace tapping away on his vintage Smith Corona typewriter in his apartment in Los Angeles. He uses it for his business, which is writing poetry on demand.

BRIAN SONIA-WALLACE: Poetry is like the ultimate noncommercial art form. You know, unless you're a professor or an Instagram celebrity poet, you're really not usually making a living doing this.

RAZ: But about four years ago, Brian sort of stumbled into the poetry business when he was laid off from his job writing grants for a nonprofit. He'd always liked writing poetry. So he thought, hey, you know, maybe I should just set up a typewriter in public and sell my poems to people passing by.

SONIA-WALLACE: I thought, OK, let me see if I can pay my rent for one month. It was this experiment in whether you could make any money off of this poetry thing.

RAZ: So Brian started to show up at arts festivals and other events around town with his typewriter and a small table.

SONIA-WALLACE: I originally had a sign that said, give me a topic. I'll write you a poem. Pay me what you think it's worth.

RAZ: And it turns out people would pay anywhere between $5 and $40 for a custom poem about their boyfriends or their girlfriends - even about their dogs. Anyway, by the end of that month, Brian had earned enough money to pay his rent and his utilities.

SONIA-WALLACE: And it was funny because after the month, I was like, OK, well, now I'm going to hang my typewriter up and focus on other things. And I kept getting calls from either people who I had met on the street or people who had seen something about it asking me to do private events.

RAZ: But when it came to private events, Brian had no idea what to charge.

SONIA-WALLACE: And I remember there were a couple of early ones. One in particular, I remember I did, like, an investment banking conference. And I didn't really understand that you could set a baseline price and that people had budgets and would pay for event entertainment.

RAZ: Brian even scored a gig at The Mall of America, where people lined up for custom poems, including one particularly interesting customer.

SONIA-WALLACE: I did my usual open. Hi, how are you? What do you need a poem about? She thought for a minute. She said, you know, I've actually just come from a week-long silent meditation retreat. It was really intense and emotional. And so I just came to get some Dippin' Dots as a reward.

RAZ: So Brian got to work.


RAZ: And a few minutes later, he presented her with the poem.

SONIA-WALLACE: (Reading) It's scary how hectic a week of silence can be, left alone with our thoughts and no release fouled (ph). Do we implode?

RAZ: That woman ended up hugging Brian after he read her that poem. Apparently, it happens to him a lot.

SONIA-WALLACE: (Reading) Finding a lightness that starts deep in the core of my belly and ends with Dippin' Dots at the mall.


RAZ: Brian Sonia-Wallace's business is called Rent Poet. This year, he expects to make about $30,000. And he no longer does this solo. He works with a whole team of freelancers who make money with poetry. If you want to hear Brian's entire poem, you can visit our Facebook page. And, of course, if you want to tell us your story, go to

And thanks for listening to our show this week. If you want to find out more or hear previous episodes, you can go to Please also subscribe to our podcast at Apple Podcasts or however you get your podcasts. You can also write us - Our Twitter address is @HowIBuiltThis.

Our show was produced this week by Rund Abdelfatah. Ramtin Arablouei composed the music. Thanks also to Neva Grant, Sanaz Meshkinpour, Thomas Lu and Jeff Rogers. Our intern is Nour Cudsi. I'm Guy Raz, and you've been listening to HOW I BUILT THIS from NPR.


Copyright © 2018 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.