How to Build a Product for Everyone

at UXDX EMEA. Video: https://www.youtube.com/watch?v=LCRiidGZ-lo

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

An experiment in focus

[00:00:00] Hi everyone, how's everyone doing? Everyone's good? If you were here this morning, you know it's going to be interactive, so get ready to raise your hands and use your voices. But first I'm going to do a little experiment. This experiment is from a man named Olafur Eliasson, and it's going to be about what you see in your mind and what you see in front of you. So let's take a look at the screen first. The topic today is going to be about focus. Nope, okay. Right, cool. Oops. We're going to talk about focus, but not bad focus. Just a small comment. Great.

[00:00:46] Cool. We're going to do a little experiment about what you see on the screen here, and I'm going to ask you to focus really, really closely on how it makes you feel and what you think about when your first thought crosses your mind. First we're going to start with a little color. Now it's just a big black screen. It could be anything, and it could be whatever you want it to be, and this is essentially how our jobs work: we create something out of nothing.

[00:01:10] Then imagine this: a beautiful white canvas. Imagine it is the mobile app you're designing. It could be the landing page you're working on, it could be the email campaign, it could be anything. How does that make you feel? What is the first thing that crosses your mind when you see a big open canvas? Imagine the screen is like a lamp. And now it's yellow. Enjoy the yellow, take a deep breath, and let's change the color to red.

[00:01:40] Do you feel anything special now? Do you feel anything different? We're using CMYK, yeah; it's a very fun version of red. I like having my hook. Red light makes us more calm, and pink light does a very similar thing. Meanwhile, blue light keeps us awake, right? We all now have the story about blue light filter glasses and all these things. And it also influences the way our customers react when they see our products, our marketing, etc.

[00:02:14] This is a little experiment done by Olafur Eliasson, and this is like art, so what you see is quite up to you. Some of you are like, "I don't see anything." Some of you are like, "Oh, I really feel like I've been moved and transported to a new world." It's really just about grabbing your attention and getting you to focus, even if it's forming the opinion that this is BS. Now I have your focus, and now we can start talking about the real topic.

[00:02:38] This is the mess of being in UX, right? As creatives, as designers, the whole world's full of mess. The user needs something, because we have to get some insight, and this is how we prove our value. This is what we're constantly pushing for: fresh yet quick ideas. We prepare, prepare, prepare, but we lack focus. And focus on what? What I always say is, you're focusing on outcomes, on metrics, you're trying to hit a certain point or cost, or you're building a case. But everything we do should be about users. And I hate this term, because it feels like drugs, and it shouldn't be an addictive kind of thing. It should be about humans. It should be about people.

Who I am, and Glovo

[00:03:20] Before I go into my meat, I'd like to tell you who the heck this guy on stage is. Let's talk about some people. This is my mom. Say hi, Mom. Cool. And this is very, very young me and my even younger brother and my dad. I hail, from my accent, from Washington, DC. I live in Spain, in Barcelona, working for Glovo, and my family is from both West Africa, as you see here, Liberia in the green, as well as the Philippines, all the way on the other side of the map. That's one way you can think of me.

[00:03:56] The other way you can think of me is maybe my professional background. I have worked for a ton of companies, because I like to leave companies quickly, AKA consulting. So it's a lot of embedded work, a lot of consulting work, a lot of in-house, a lot of day one, a lot of day ten thousand four hundred. I've worked with really, really small companies as employee number one. I've worked for NGOs, I've worked for banks, I've worked for governments, I've worked for big corporations, and now I'm back in the middle at Glovo, at around 5,000 people.

[00:04:31] The vision of Glovo is to give everyone easy access to anything in their city. Sometimes you think that's a crock: "Oh, anyone, anything, easy. These are really great fluff words that we're never going to achieve." So we started putting real meaning to these things. For everyone, we particularly target the average consumer. We went from food delivery in 2014 to groceries, to laundry, to pharmaceuticals, to COVID tests, to handymen, to plumbers who unclog your drains. Because if we could deliver the goods, we could also deliver the services. It's a time-scheduled delivery: when you open your door, whatever walks through your door can technically be considered a delivery, as long as we're the ones dropping it off.

[00:05:14] We decided to do this, and it made the UX team a very critical part of the organization from a very, very early stage. We hired 2,000 people in the company last year, but even before that massive jump we already had 100 people in UX, and now we're growing really rapidly. We just joined the Delivery Hero group, so now we're in 75 countries with a total reach of 2.2 billion customers, 800 million of which I am personally responsible for.

[00:05:43] All right. I'm wearing a little jacket from Barcelona to bring a little Spain here with you. Unfortunately, the warmth didn't quite make it, so it's a little cold. But let's talk properly now about focus.

Hygge, and going home at 4:30

[00:05:57] In Denmark, how many of you know how to pronounce this word properly? Okay, let's just hear it. Yeah, everyone was like "higgy," and that's what I wish the word was like. It sounds cute that way, like a little Zen vibe: you feel it, it's higgy. I'm like, no, no. It's like that silent hum from the fridge at night. Yeah, it's like "eat a cookie." No. Okay, so hygge is like focusing on the moment. It's making the ordinary extraordinary, making the normal special. It's what happens when you live life well, and this is a guiding principle in Denmark.

[00:06:44] It's also something that I tried to learn very quickly, because I've been out of the States now for nine years. If you don't know about the States, we love to work a lot; we work too much. Then I moved to Amsterdam, and I was at Booking.com, and there was a guy on payroll who went around at 4:30 and he'd be like, "You should go home." I was like, "Who are you?" He's like, "I'm the guy that tells you to go home." I was like, this guy actually gets paid for this? He's like, "I water the plants too." And I'm like, okay, cool, it's a slightly two-factor job. And he's like, "The internet works just as well from your house as it does in the office. You should go home. We want work-life balance. The sun will be gone for about a third of the year; you should really go out and get some now." And I was like, I like that. I like this.

Everyone: no one is left behind

[00:07:27] Let's go back to that yellow. What are the words you think about when you see the color yellow? Just shout them out. Yeah, now that I've preempted you with cold, you're like "sun," "warmth," winter came and now it's leaving. Okay. Any other words you associate with yellow? "Refreshing." "Cheap." Somebody said cheap. It's not. I was going to say joy, but it's all right; whatever you see is whatever you get. What does it do to you and your mind?

[00:08:09] Yellow is a really important color for me, because it's the color of Glovo. It's the color of our branding, our bags, our mascot. We have a pineapple as a mascot; don't ask me why. Glovo is actually a play on words, and no one understands it because no one speaks Spanish. Globo is a balloon, and the little balloon that's next to Glovo is meant to accompany that. But if you ever listen to somebody speaking very fluent Spanish, B's sound like V's, so they thought it was a joke, and no one's gotten it for eight years. So, logo.

[00:08:41] Giving everyone easy access to anything in their city: let's break this down. And the yellow we use is more of a mustard; it's more of a food play. So, everyone. Everyone is defined internally as "no one is left behind." We're trying to make sure the product is used by everyone, no matter what income they come from, what language they speak, what nationality they have, whether they're a visitor or they're local. We consider the needs of all the communities. This is where everyone is like, "Aren't you like Uber?" And we're like, no, no, no, we're better.

[00:09:16] We also own this platform, and we own the responsibility for the couriers who enable the platform, so we think about every single one. A third of my team is dedicated just to couriers. I'm not going to do a comparison through the entire talk, but most of the gig economy, most of the industry, focuses 90% on the customer, and there's some small team that does small business sales and cool things for them, and they're like, "We need you to put up content with photos," and that's that part. On our team, we actually spend time delivering food. Every single employee delivers 18 orders a quarter. We put on the bags, we go out in the rain, we come to your front door and we're like, "Hey, sorry, they didn't have mustard, so I got Dijon. Is that close enough? Do you like Dijon mustard?" And then you're like, "No," and I'm like, "I'm so sorry, but okay, here's your food."

[00:10:04] So everyone feels the problem on the courier side, everyone feels the problem on the partner side, and everyone feels the problem on the customer side. I make UX a challenge and a responsibility of every single person. The VPs hate me, because I have a dashboard where I'm like, "Huh, according to the system you haven't delivered food in 14 months, Danny." He's just like, "I was busy." And I was like, "You ordered food today. You could have picked it up yourself and scooted it back to the office." So I just make them put scooter time on their calendar now, which is when they go and pick up their own food. So yeah, everyone, and I mean everyone.

Easy access, and the verticals nobody knows about

[00:10:36] Second, easy access. In order for us to think about everyone, you think: everyone has different needs, different responsibilities, different preferences, different budgets, different limitations. So we define easy as seamless and direct, and we've really streamlined the UX. Most of our industry started out with food delivery. How many of you have ordered food on an app? Right. Keep your hands down, or you're just lying.

[00:11:01] During the pandemic we had triple growth. My funniest moment, because I've only been at this company for about one year: I was doing a product trip in Morocco, and the CTO [?] of Greenpeace happened to be at this event, and she ran up to me. I was like, "Oh, Greenpeace," thinking she was going to complain about our worker conditions or something. She was like, "Oh, you fed me during the pandemic. You were the only company that delivered food for 12 months in Kenya." And I was like, "Oh yes, yes, we did do that. That's amazing, we should do more of this. It's fantastic, we should do a partnership deal: Greenpeace endorses Glovo." They haven't done it yet, so, let's not record that. Sorry.

[00:11:41] So we said easy access has to be really, really simple. If you've seen the majority of other products, you go in and Uber is like, "We assume you want a car," and if you don't want a car, the entire UI changes for food. Instead, our product has six or seven big circles, like you're a seven-year-old. It's like, "Do you want food, or groceries, or pills, or flowers, or anything?" And people are like, "Food." Then it's like when you go to the Coca-Cola machines and you have those rings of other brands, and you go up to three or four or five. It's like a wizard. We're like, "Cool, you want food. Do you want premium food, regular food, or this sponsored food bubble from McDonald's?" No one wants a McDonald's food bubble.

[00:12:22] So they pick regular food, and then they get into the user experience, and they go through it in about a minute and 34 seconds. This has been fantastic. We have the fastest delivery service and the fastest user experience for ordering delivery. What it causes a problem with is new verticals. How many of our customers from food do you think buy groceries on Glovo? Raise your hand if you think it's 10 percent. Raise your hand if you think it's 20 percent. A little more raising. If you think it's 30 percent? You guys think it's a lot higher. It's 13 percent.

[00:12:59] 13% of all customers know. 87% of all customers have no idea we are in multiple verticals. 82% of our customers who pay us 10 bucks a month to be in our loyalty program don't know we have other verticals, and we talk to them every day, because we're like, "Hey, you're missing a rating; you've got 14 points on this order; you have a free voucher for food." So we're trying to change what easy access means. Before, it was just speed: "Oh, speed is ease, ease of use is speed, perfect, good to go." We need to actually make sure we have some loyalty and retention, some kind of trust relationship with the customers.

Anything: hamburgers, flowers and late-night combos

[00:13:35] Then we have anything. I've already gone into all the weird little scenarios we cover. This is my little general chart: we have food, we have shops, supermarket, beauty and care, courier, and anything. How many of you have ever ordered flowers last minute? Yes, we see fewer hands. Just be proud you did it. She has no idea you did it, but you did, and the flowers got there on time, and there was a little note, and she thinks you signed it. No, she does.

[00:14:01] Our number one sales item was hamburgers. Number two was something I can't tell you yet. Number three was flowers and roses, in that combination. And number four is a combo deal that we just made, because four and five were so consistently ordered on Friday nights and Saturday nights that we were like, "We should just sell lube with condoms in one package." Because people are like, "I need them now," and we're like, "We'll be there in 26 minutes." He's like, "Perfect: one shower and a candlelit away from a great night."

[00:14:34] So we literally started optimizing the entire experience around the most used items, but it essentially took us further away from being able to do really cool things that people didn't ask for and didn't naturally do. Imagine this is the user experience: a customer gets their phone, they put an order in, they assume some person somewhere puts that order together, and then some other person goes and grabs it and brings it to you. Do you think this is how it works? Not your own bias; you're like, "I'm not using my hand for anything anymore." Don't be afraid.

[00:15:07] This is closer to how it works. Everyone comes in and uses our little bubbles, and then they use the search bar. We realized that designing for everyone meant we needed to be able to help them find exactly what they want as fast as possible, so again, optimize for speed. We have this insanely specific search that knows the difference between capsicum and green peppers, because some people like calling it capsicum, don't ask me why; I think some of you are here today. We needed to have the search auto-translate into 27 different languages and three different dialects. Even if you're in Spain, the search doesn't have a translate button; it just automatically knows what you want. And this is generally how the experience starts. Then it goes to the partner, and everything's fine.

Photos, and why dumplings are one category

[00:15:54] How many of you like ordering things without seeing them? Who likes blind shopping? Yeah, right. Okay, cool, you like photos, right? How many times do you go to get food and there's no photo of the food item? A lot, right? You're like, "Pad thai, ooh, it's the same thing I got last time, so it should be the same-looking thing." Every restaurant makes the same pad thai, right? There are some noodles, there's a little bit of shrimp, there's a chicken that got in there somehow, I don't know how that got in there. Then you eat your food and you're like, "It's fine." Then you go to another restaurant and you're like, "Why is this one spicy?" And they're like, "This is Thai spice; this is the way it's supposed to be." And you're like, "No, no, no, I want the British version of the pad thai I got last time." And why are there bell peppers in here? No, I'm joking.

[00:16:34] So we started saying we can't do the same thing for everyone. We can't solve all the problems. We tried the Airbnb Brian Chesky model: we'll just hire photographers and send them to the restaurants. How many times have you gone to a restaurant and they're like, "Hey, we have a new special in town"? That's a lot, right? They're always trying some new thing; there's always a two-for-one combo. If you do a photography session for a restaurant, by August of that same year half the menu has no photos again, because they've made 14 new items, and they're not shutting down the restaurant to make the entire menu every six months just to take photos. So they're like, "We're not going to do it."

[00:17:08] So we're like, cool, let's try to do a program, and let's make it more fun. What do you think the number two most ordered food item is at Glovo, or in general? Just shout it out. This is the one I couldn't say earlier. People think it's pizza, which is actually funny. Sushi, okay. Bananas? Bananas, I find this highly offensive. No. Bananas: they don't know we have groceries, in their minds. No bananas. Anything else? Alcohol. Burgers is number one, yes. Fries just come with the burgers; you know that, come on. All right, all right.

[00:17:53] It's not a category that we naturally saw. We had to add up all the data, and then we're like, all of these are the same food item: dumplings. And it's not just one dumpling, right? These are not actually called dumplings; these are called gyoza. Then we have pierogi from Poland, and then you have noodling [?] from Austria, and then you have tortellini from Italy, and then you have empanadas from South America, and then you have bhawan [?] from Taiwan, you have Clipper Lodge [?], you have tamales, you have Festival [?], you have pasties from Cornwall. Any people? Yeah, yeah. It's kind of like a dumpling, right? It's a really big dumpling. It's like a calzone plus, but it's fine.

[00:18:38] We realized that, just like with food, we weren't categorizing things correctly. We were focusing so much on the literal thing that we weren't focusing on what it means to people. When someone's like, "I want a dumpling," they're really saying they want a hot pocket of something good and tasty. They don't want it floating on top of a piece of bread; they want it to be hugged. Okay, so we're like, cool, this is the same category. You want a dumpling, you want a dumpling. You want a Spanish dumpling or a Chinese one? You want a dumpling, right?

Customers aren't just customers

[00:19:04] So we asked, what other categories are we not associating with each other properly? And we're like, oh, our customers, right? What do we all have in common? We come together and we eat food. If this event didn't have food, you guys probably wouldn't be here. We have food at an event and people suddenly show up. I have meetings and I'm like, "Hey, the meeting's at 9 a.m.," and they're like, "Kevin, this is ridiculous." And I'm like, "I'm bringing bagels, empanadas and coffee, good coffee," and suddenly everyone's on time. Then everyone's like, "Where is the coffee? We're out of coffee. Kevin, you promised coffee, and now my donut is dry." And I'm like, "Okay, well, we're here to talk about the budget, so that's why we have no coffee."

[00:19:44] People are not all the same people, right? They have different motivations and different needs; you call them different words. You might use the word friend; that person might call you their acquaintance. Not the same thing. So we did the design for customers, and we're like, but the customers aren't just customers. We have some customers, we have some friends, we have some friends with benefits, you have your family members, you have people you met at a party one time who keep calling you their Uncle Sam.

[00:20:11] So we can't just think about delivery from the one perspective that all the customers just want their stuff. They want it for different reasons. Sometimes it's a secret, sometimes it's a gift: "I need you to secretly give this to me through the back door. My wife doesn't know that I just bought her present for our anniversary two minutes ago on Glovo." We're like, "Cool, meet you near the green door. Is it the right one?" And it happens all the time. Our couriers are like, "Yeah, you guys don't understand. That's not the address I had to go to. He was like, 'Go around the back,' and that's why I was late." And we're like, okay, that seems believable.

[00:20:40] But now we offer much, much more than food. Now, if Tony, your Uncle Sam, is not going to come and make the paella, we will deliver the paella, the spices and the rice at 2:30, and at 2:45 the chef comes with the paella dish. It looks really impressive, cooking in the sun, and everyone's super happy, and now your friends think that you made paella for them. In reality he just came out the back, just in time for you to be like, "Smells good, I'm just checking on it. It looks great." Everyone's happy, right? So, problem solved, party time? No.

[00:21:12] What ends up happening is that as we launched every new vertical, we ended up focusing on more and more and more people, to the point where the product is now so bespoke and so different in every single vertical in every country, and then you have the language, and then you have localization for culture. That is an incredibly hard product to maintain. Again, we're 100 people, 40 of whom are designers. We have 5 million point two six seven eight nine million orders a week. That'd be funnier. No? Cool, all right. We have a ton of orders per week, and it's always going up.

Listening to 50,000 couriers

[00:21:44] So, okay, what do you think we did? Just shout out a verb. How do you think we solved this problem? "Centralized." Any other verbs? "Parameterized." "Tag it up." Okay. "Prioritize." You can deviate from the "-izes," it's okay. All right. So we said, okay, first we're going to listen.

[00:22:14] Here in the middle you have my best friend Danny. Danny's always in a photo; he's always in my speeches, and I'm in his a lot, so now we make fun of each other on public stages. No one gets it, because Danny's not here to laugh at me. But Danny's always in the middle of a photo, by accident, making this face, because he'd be like, "Ha ha, I had this idea two years ago and I sent it to you and you ignored it." And this is Danny's face for that. Sentiments.

[00:22:37] We brought in 50,000 couriers over 12 months, and we said, "Hey, you're the only people in the business who come in contact with every vertical. You deliver food, you deliver groceries, you deliver TVs, you deliver vacuum cleaners, you deliver baskets of laundry, you pick up handymen and drop them off so that Rebecca can get her TV installed. What are the problems you see, because you interact with all the different combinations?"

[00:23:04] The majority of our couriers are not from the country they operate and work in. A lot of our couriers have different reasons for being couriers. In Western Asia, and when I say Western Asia people are like, "What is that?" and I'm like, Kyrgyzstan, Kazakhstan, the Stans; that's not what they're called, just bear with me here. And then you have North Africa and West Africa that we operate in. We pay so well that this is a career. Our average retention for a courier is 12 years [?], and we're an eight-year-old company; it's a joke, but we've had couriers since day one in these countries, and they're like, "We're not going anywhere." It's on their LinkedIn: "Courier at Glovo." People are like, "Ooh, sir, we have a table for you right here at the front of the restaurant, because you have a stable job." We love that. It has a whole different meaning to them than "gig worker" does in Europe, so even designing for them was different.

[00:23:51] So we brought them all in and we just listened. We listened, and we said, "Hey, we have a couple of goals here. We're going to make the working conditions better, not because we have to, but because we want to. We're not an Uber." And they were like, "Huh, okay, well, everyone says that. What are you actually going to commit to?" So we said, first, we are now signed on to the UN SDPs [?], the strategic development program goals, and we're going to commit to them. We're going to make gig work as flexible and reliable as regular work.

[00:24:18] Two, we bought the website thecourierpledge.com. If everyone thinks that we're going to be evil, let's prove them wrong by making it public: what contract you sign, what the onboarding is for couriers, what the pay is per country, what the pay is if it's rainy or it's late, or if someone orders a 10-kilogram pack of water bottles, which, I don't know why, keeps happening. You're like, "You know he doesn't want to carry this. And if I ask you to pay two extra euros, why are you saying no?" And they're just like, "I don't know, you should pay for it." And we're like, that's not how the world works. So we said, let's bring them in and show everyone what we do.

Executives on scooters, couriers on product teams

[00:24:57] That was step one. Step two: again, I like making people do work. They're like, "Kevin, we hired you to solve UX," and I'm like, "Cool, you're going to go do work." They're like, "That's not how it works." I'm like, "That's how I work." So let's go get some scooters. Starting from the executive team, you're going to deliver food 18 times every three months. They're like, "Why?" And I said, to make it more fun, we're going to travel around the world and do this. You're going to do it in Morocco, where there are sandstorms. You're going to do it in Turkey, where it rains so hard that entire cities just become rivers. We're going to do it everywhere. We're going to do it where there are no elevators or lifts. You're going to do it where people live in basements. You're going to do it in Taiwan, where there are 84 floors. You're going to do it in malls where you can't find the restaurant for 15 minutes, because all it says is "You are here."

[00:25:40] Imagine the number of problems you discover when you're the person having the problem, and imagine how many solutions you want to come up with when it's your problem to solve. So we made them do that.

[00:25:49] Second, we voted the men [?]. We actually hired a bunch of couriers. We hired people from FedEx, from logistics centers, from MFCs, micro-fulfillment centers, and we said, "You know so much about the space, you should be on the product team solving this problem." PMs, no matter where they came from, no matter how much experience they have solving in one industry, have never worked the ground in that industry. 18 visits a month isn't going to suddenly make you an expert at delivering packages perfectly, or throwing the mail into the mail spot from the car. You just aren't going to get that same level of expertise. So we actually hired them, brought them in, made a panel, made a committee, and actually put them on product teams.

[00:26:30] Third, we said, let's go do what Google does: launch stuff, and then kill it very quickly if it doesn't work. We're not going to kill courier careers, but we are going to say, you don't want to do delivery in the rain when there are 100-miles-per-hour winds, so let's make this little robot guy do it and see if he complains as much. Guess what? He didn't. So we said, let's do robot deliveries in the scenarios where couriers don't want to be. Awesome, everyone's happy.

[00:26:56] And these are the little bubbles I talked about earlier. You see, it feels like a baby game. Step one: do you want a red book or a blue book? Step two: do you want a long book or a short book? That's how the UI is. It looks nothing like anyone else. Everyone else is like, "Here are 18 ads; McDonald's wants you to buy them." You're like, "I don't ever buy McDonald's; why is this ad still here?" So instead we just said, give us the options you care about, and we'll show you only what matches that.

Values: carbon neutrality, farmers and war zones

[00:27:25] Then we decided we had to do a lot more than that. Focusing on people told us that they have different values, and values are changing. So we're starting to electrify the entire fleet. We are the only carbon-neutral delivery product in the market. We have gotten to the point now where all of our buildings, well, 24 of them, are carbon neutral with solar panels and wind power, and we are now giving couriers financing options to replace their diesel mopeds with electric ones and bikes.

[00:27:52] We said we look beyond. It's not just the end user; it isn't just the customers, the couriers and the partners. It's also the customer service agents, the sales teams and the farmers. So now we go in and help them get seeds. We try to eliminate food loss, so we will also pick up and deliver produce, ingredients, seeds, things that need to go to market, to actually eliminate the waste that happens before we get the food.

[00:28:16] Then we chose to be a little political. We said, hey, between humanity and politics there's a line, and we're going to push that line a little further to the left. We need to actually decide when it's too late for us to care and show care. We have about 500 engineers in Kyiv, and we have about 400 in Belarus. There are no easy decisions here, but we chose to still operate and deliver on behalf of the UN food program. All the food in Kyiv, to this day, all the food in Ukraine to this day, is delivered by us and by Bolt. So we choose to operate in war zones so people have the food they need.

[00:28:53] And last but not least, now that we are the first carbon-neutral program, we said the last thing we need to do is figure out a path to address the biggest complaint we've heard so far: "You're a logistics company; how can you ever be green?" We said that we're going to be the first carbon neutral, and we're also going to be the first company that actually paves a pathway to citizenship for all couriers in our product, so that you go from being an immigrant, to couriering for three or four months, to being a citizen, to being an employee, as easily as it is to hire a PM. So when I say focus on people, I mean really focus on people. All right, that's my talk. Throw me some questions.

Q&A

[00:29:46] Host: Thank you, thank you so much, Kevin, for such a joyful talk, and a little bit of insight into fashion in Barcelona. Thank you so much, lovely. Okay, so do we have any questions from the audience?

[00:30:03] Kevin: My paid heckler.

[00:30:11] Audience: You put out to the audience the problem of having this infinitely complex set of columns and subdivisions, but I don't think you actually said how you solved it. You just gave a set of processes for bringing in couriers and moved on to the great green credentials. But what was the answer to the question?

[00:30:34] Kevin: You're talking about the product being so complicated in all the factions, right? All right. We are in 25 markets, and now, because of the group, we're in 75 markets, and it's impossible to solve this problem from the cushy seat of Barcelona. So what we actually did is we have a learning and development program, and I am one of the faces of it, to train all the sales teams on design thinking and product psychology. We also train the account managers on the same two skills, and we have them have a direct line to myself and the product directors for the space they work in. So anyone who sells a grocery store chain on joining Glovo has a direct line to talk to the product director of cucumbers, retail groceries, so that there is this direct voice of the customer from the NPS marketing side, the internal UX side, and also the front lines of sales.

[00:31:22] Audience: Sorry, but I still haven't understood how you've dealt with the complexity. I can hear how you've done more research.

[00:31:31] Kevin: So let me try this. The result of that turns into a very, very complicated experimentation ML layer. It's like Facebook, right? If you open Facebook or Google right now and search the same thing, you don't get the same results. So we thought that, instead of trying to do a one-size-fits-all solution, going from crazy complicated to crazy simple, we'd come in the middle. We segmented the customers, the partners and the couriers into these groups of eight, three and six.

[00:32:03] Kevin: Quick thing: restaurants aren't the same. So McDonald's and the chains of the road get a certain experience. Then you have what we call Local Heroes, restaurants that are doing really well. We actually have a thing called Glovo Concepts, and we help them launch a second restaurant; we finance a second restaurant or a ghost kitchen or something like this. So we just have dedicated teams now who work with each segment of customer and each segment of partner, and there are eight versions of the product, but you see the version that makes the most sense in the country you're in.

[00:32:32] Audience: Thank you. Okay.

[00:32:43] Audience: Hi. Do you deliver the same product in all the markets, or do you customize the experience based on the market?

[00:32:52] Kevin: We are in 2,400 cities, and I would say it's very bespoke. We launched micro-localization in March, so in Spain we even have Basque and Catalan. If you're in that city, you have the option of an additional two languages that we support just for that region. We have the complexity of translating the UI as well as the menus as well as UGC reviews. So we have a very, very special team that goes through the linguistics and the meaning and intention of words and tries to do a custom machine translation for each language. Then we have a filter on tone. In Portugal we were very, very casual, and we couldn't get older customers, because they thought it was a bit disrespectful, before we talked to them. So we actually changed to the formal tense. In each country there's a flag for local language, tone of voice, customer type, and even minor details, like whether we think you're a household or an individual.

[00:33:48] Audience: Thank God I didn't drop that. Thank you, great talk. You were focusing on the couriers there; I'm just curious about the merchants that you work with as well. There are so many diverse types of merchants in so many different countries. How do you manage the catalog of products that they all have to offer? Do you use a third party, or did you build your own in-house system?

[00:34:07] Kevin: It's very custom for Glovo, but now that we're part of this group, we share learnings and lessons. Some of you may have heard of some of the other companies in the group, because we are marketed less despite being bigger: we own foodpanda, Talabat, Foodora, efood. These are all the same company. Ultimately what we do is take learnings from each other. One of the easiest things is where we do customizations. We have a majority food market in the west of Europe. In Eastern Europe we have 40% margins on groceries; they order everything from us on groceries, and because of the bulk orders there's a lot more margin, and then we use that margin to pay the couriers.

[00:34:52] Kevin: And then there are places where there is a distinct understanding of the problems. In Ukraine right now, people are paying seven euros for couriers, because the courier is literally dodging missiles to get you food, and they're willing to pay. And then in Spain we tried to increase it 30 cents, and they literally had a revolt outside the office. So we are very, very local about these changes and surcharges and things like this.

[00:35:18] Host: Right, thank you so much, guys. Thank you, Kevin. Bye.