View transcript
VideoDays, Webinars5, WebinarSpots, RawMode, Thomas Madsen-Mygdal, Steffen Fagerström, WebinarSpots, RawMode, Thomas Madsen-Mygdal, Let's get this afternoon started. My next guest is Niki Grunfeld. Niki has a super exciting background. Don't be so nervous. I'm only saying nice things about you. He has an exciting background in anthropology, photography. He's a chaos pilot and a lawyer. He was originally or is originally from the Netherlands and is now working in Denmark for Driver. He also runs an open-air cinema in Copenhagen I've never heard of before. It's called Vinmulen in Apel. I have to check that out. Nicky is an experienced leader with talent for analysis and research, concept development, product management, and process design. He likes to take part in the creative process and get his hands dirty doing the UX. He thrives in handling complex problems and has a deep care for other humans. So Nicky told me that this morning he took a cab here, of course, for quality assurance and field research. Once a UXer, always a UXer. Please welcome Nicky Grunfeld. Thank you. Just testing if this is working. The bearded. The beard and the microphone. That's okay. Thank you for the lovely introduction. And thank you, everyone here, for hanging in there. It's been a long day. Are you all ready for the ride? Good. I'm a designer at Driver. And at Driver, we help taxi companies deliver the best possible experience to their customers. Now, there's a whole lot of design thinking that goes. And that's what I would like to talk about today. The hidden design behind a cab ride. And to do that, I'm going to start out with talking a bit about the current ground transportation landscape and about driver's history. Before I do that, I just want to say that you guys rock. It's a kind of humbling experience to share the stage with these people. And with the companies they represent. It's only been two weeks since I last handed my keys over to a total stranger. After babbling for five minutes about where the towels were. And that the door to my balcony doesn't close that nicely. Off I was. And a stranger was living in my apartment. I mean, that's weird. Then we're talking autonomous cars in 2017. Either all of you guys are really good at keeping straight face. Or you're just not freaking out like me about that. That's crazy. Besides that it's really amazing. It's also a huge opportunity and something that everyone involved in the taxi industry obviously has to think hard about. What that means for our future. And last but not least, just to name one more example. I wish that that's Paul's app had come out yesterday. So I could have used a little bit to calm down. But I'm trying to breathe here and get into the presentation. In short, I think it's amazing all these companies that are represented here. And incremental innovation they bring forth. And the industries they disrupt. And the taxi industry definitely knows what disruption is. Autonomous cars, electrical cars. We're talking diminishing vehicle ownership. And cloud-based technology that's enabled a range of companies. To pop up. These companies are called on-demand transportation companies. Now, taxis have always been on-demand. Even before it was cool to say on-demand. But one thing is sure. They need to step up their game. Because new standards are being set. And ordering a cab through an app just sets different expectations. Than the typical and the conventional way of booking a taxi. That's something I'm going to get back to. But basically, for a lot of taxi companies over the coming years. It's going to be an adapt or die situation. And this is where drivers steps in. And where we... I'm sorry. I'm just going to take a quick... This is where we, more than just helping these taxi companies. Give their customers the best booking experience possible. We step in and we want to make a strategic partnership. To look at the taxi product. And say. How can we make this relevant in the future's urban transportation landscape? I'm going to talk about that throughout this talk. But I think a good way of priming all of that. Is talking a little bit about the difference between Uber and driver. I'm saying Uber because that's the one we all know. But obviously there's a whole range of on-demand transportation services. And the main difference is that we have chosen to work. With the strategic partner that's the taxi central. The three stakeholders at play here. Except for ourselves. Are the passengers that want to get from A to B quickly. Preferably cheaply and safely. Then there's the taxi drivers. Who are often independent business owners. And work on a quite autonomous level. They make a contract with taxi centrals. That promise to send them loads of bookings. And because they take care of that. And a whole lot of other practical matters. They take a percentage of every fare that a driver drives. So all of these parties are our customers. And they have all quite specific needs that we need to address. And that's something that I'll be talking about more today as well. So but let's switch gears. Because I don't want to talk too much about strategy and stuff like that. I want to get into the design of stuff. And the way I want to prime you. Is to have a look at the design. And have a look at what booking a taxi looked like not too long ago. Yeah. That one was empty. Damn it. Okay let's get dry here. And maybe just call a taxi company. Does anyone know a number of a local company? No? A couple of different ones. Maybe I'll just Google it. But I caught that one. So let's give them a call. Ah hello. I would like to order a cab at. One second. Does anyone know the address down here? No? It's called Tollbollen. Tollbollen? Yes. Okay. Thank you. Now I'll spare you more of my acting skills. But it's clear that there's a lot that can be improved on this experience. And there's a lot more. For example if I would be going outside now. After I hung up with the operator. I would be forced to stand outside and try and find this cab. Because otherwise it would probably drive on. Because the cab driver just wants to get people in his cab. And drive them. I never heard from the operator about what this trip was approximately going to cost me. So I'd have to be worried about if the driver was driving me around. And what the fare in the end would be. And if I would be getting ripped off. Now getting into a cab also means that there's a reasonable chance I'm not going to get stellar five star service. As we all know. So these were some of the issues that we were working with. And a couple of these experiences while traveling around. Were sort of the foundations for one of our co-founders to start what was then Click A Taxi. And finally turn into Driver. And I'm going to quickly run you through that history. And then I promise we're going to get into the design of things. So there was Click A Taxi. And the idea behind that was very simple. To make it simple to book a taxi anywhere you are in the world. The first version of that platform was built and trialed in London. And by then it was with a model that was very much like Uber. We started around the same time as these guys. And it was to connect passengers and drivers directly. With us as only intermediary. We were the network orchestrators getting these people together. Now fast forward to today. And our mission is to drive down pollution. And the price for urban transport. And we've chosen to do so with Taxi Central as a strategic partner. We're 25 people. We have an award winning product suite. That we deliver to these taxi companies. And their drivers. And we have offices in Vilnius, Copenhagen and London. And driving people around in those same locations. Now our answer to. That little drama thing that I just put up for all of you. Is right in front of you. It's an app. And you can book a car. We already know where you are. So you don't have to worry about that. If you want to you can add a destination address. So that we can give you an approximate fare for that ride. You always know the status of your taxi driver. So you know when to go out into the rain to get into the cab. And last but not least. You can rate your drive. Which allows us to make sure that there's a certain level of service. That you're being given. There's a lot more features that I'll be getting into. But this is sort of the meta overview. Of how we solved the issues that I was just trying to play out. But the app is just the tip of the iceberg. There's a whole lot of other stuff going on behind to make this all happen. I already discussed this three stakeholders that are at play here. So what's important I think. To mention here is that running a taxi business. Is quite a bit like running a restaurant in a way. There's complicated logistics at play. And the good thing about being an operating manager in a restaurant. Is that all of it happens under one roof. You can make sure that your chefs are in time for prep. You can see the food come out of the kitchen. You can see your waiters interact with your customers. And if anything seems to be going off. Because you can sense if your customers are happy or not. You can walk up to them and start a conversation. Now for a time. Now for a taxi business. This happens at the scale of a city. And it's therefore much harder to get a good sense of. For example the service level that you're delivering to your customers. So these are some of the products we've built. To solve that issue. Quickly from left to right. We have a mobile point of sale system. A mobile point of sale. A driver app. A user app. We have some booking options for corporations and hotels. And we have a whole lot of range of products. Of which I'm showing two here. For the taxi centrals. A call center. And some fleet management modules. Now. What's important for the taxi central. Is just like the taxi driver. To get in a lot of bookings. And to be able to do that. You need people to come back to you. So you need to deliver a stellar experience. And this is why in short. Design is becoming so important these days. Because it's the best way to distinguish yourself from the competition. Besides that. Efficiency is important to these taxi centrals. And we have a whole lot of back office tools. To help these companies become more efficient. We use live traffic data. We use historic data. To know where bookings normally come up. And in that way we help taxi centrals. To disperse their fleet better. And a lot of this is happening. Automatically. To gain efficiency. We've tried to get these taxi companies. Who are usually pretty risk averse. To step into the future with us. And automate their businesses. But automation is very scary. Especially when you have 200 taxis. That are driving. On a Saturday night. And a lot of money goes through the system. So letting all of that. From your operators into the hands of some machines. Is a scary thing. And we built some tools for them. To sort of help. And trust automation. And I'm going to show you two of these tools. The first tool. Is the booking log. On the left an old design. On the right a new design. Luckily the sun's gone down. So I hope you can get a bit of a sense of it. It's a chronological list. Of bookings coming in. In a city. And what we did here with design. Is we added some color coordination. So that it's easier for operators. To read from top to bottom. And understand what kind of booking this is. We've also. And you can't see this on the slide. Added a progress bar. In the middle of every booking. So it's easy when you scan from top to bottom. To see is this booking just ordered. Or has it been paid already. And when we're done with all of that. We realized okay we made this better. But it's still an information overload. Imagine sitting there for seven hours in a row. Staring at a list that is ticking away. It sounds like a tough job. And it is. So we wanted to make that easier. And we designed. What we call the dispatch viewer. Once all of a sudden. All the letters and all the text fade away. And we visualize some of the key data. That you need as an operator. To make sure that your business is running smoothly. Remember. Our machines are making sure the business runs. An operator's task now is actually. More to look at if everything goes well. And if it doesn't. To act upon it. So you can see bookings coming in. You see your cars. And you see if you have a good spread over the city. You get alerts. If there are any anomalies. For example someone booking. And having to wait longer than X amount of minutes. It's an eagle eye perspective. And it helps you keep check of your restaurant. Your taxi business. The other thing that we do for operators. And after that. I'm going to switch. To drivers. And passengers. Is that. As strategic partners. We talk with them. About costs. And. If there's one thing. That new on demand transportation services have done. It's to bring the price down. Of urban transportation. And that's something that taxi. Taxi companies have to consider as well. as well. If you're in an app and you get two options to get home from a Saturday night out with your friends, it's the left vehicle that you pay $120 for or the right vehicle that you pay a $200 for. Let me make it even specific. You get picked up in a Nissan or in a Mercedes. Who would go for the Nissan? Who would go for the Mercedes? So we have a couple of luxury animals in the crowd, and I understand, because that's a nice experience as well. But pressing cost is important, and we do that by talking strategics, but we've also done that by removing a lot of bespoke hardware that taxis need, like taxi meters and a lot of other stuff. And we removed the need for expensive servers and we moved everything up in the cloud. So to run a taxi business with Driver, you need a couple of laptops in a room and you can play operator, and you need an iPhone 4S in your cars to connect to the cloud. And that's it. So this is cost that in the end can save the customer money. So much for the operators. Let's take it a step closer to the experience that you have as a passenger. I've repurposed the touch point map, something I like to use to see where there's opportunities for improvement in our service. I've simplified it, I've left out the operator. There's a lot of sub-steps in between I skipped. But basically, with all of these steps, I wanna illustrate some of the challenges that we've solved with the help of design. So in Quick, Drivers are always searching for fares. Passengers wanna book a ride. They need to find each other. Then a driver needs to navigate the traffic while getting input from, for example, an operator. The passenger needs to enjoy the ride. And finally, payment needs to be handled and everyone's on their way again. What's really important about these services, important for the passengers is timing. As I mentioned before, new standards are being set, and it's just not good enough anymore to wait 20 minutes for a cab. People don't have the time for it, and it's not necessary with today's technology. So in the first step, I would like to focus a little bit on some of the considerations we've had and the challenges we've had with regards to getting a cab to you as quickly as possible, and more importantly, getting you to point B as quickly as possible. Depending on the pickup location, there is a max ETA after which people usually close our app and try something else. The ETA is something that we show you the moment you open the app. You have a little bubble, and it will say, hopefully, if everything goes right and we've done our job well, about four minutes for a cab to come and pick you up. So what we need to look at here is two things. There's the actual waiting time for a customer, and then there's the perceived waiting time. And what we found if we go through these steps here, which is the passenger books a ride, it gets sent up to our dispatching engine, then it gets sent out to the best matched driver, the driver that's closest by, but also the driver that has the best passenger rating. That driver needs to accept the trip, or he can cancel it, and then we'll send it to the second next. So we send the confirmation to the passenger. And within this flow, there is a huge bottleneck, which is basically down here. We need drivers to accept a trip, and that can sometimes take a while. We give a driver 12 seconds to accept a trip or not. If only the second or the third driver accepts a trip, and this happens because drivers are independent business owners, they're not on payroll, and they can't be forced often in most taxi companies to take a trip, you can easily be 30 seconds along the way, and you can maybe shave off milliseconds by how quickly you send it to the driver, but you got a real issue there. So what we wanted to work with instead is perceived waiting time, or basically what we're forced to work with. So the first and foremost things we want to do is make the app look good. A good experience can cover a lot. As Donald Norman said, attractive things just work better. When you wash and wax a car, it drives better, doesn't it? Or at least it feels like it does. And I can give a little shout out to Fleming here. I don't know where you're sitting. He actually laid a lot of the foundation for the beautiful design in the driver app when he worked with Driver. Then, as I mentioned, we immediately make sure that we're very transparent about what is going to happen, and we set the expectations. So if there's four minutes waiting time, we're going to show you four minutes waiting time. Of course, this is an approximation, but we work really hard to make sure that that's something that's accurate. Then what we also do is the moment you place a booking, for first-time users, we actually explain to them that it's okay to put your phone in your pocket now, and you can do something else important. And we'll send you some notifications. The first one is that we found a car for you. The second one is that the car is actually getting close to your current location, so it's time to get out into the rain and make sure that the cab can find you. This reduces perceived waiting time a lot, especially when people learn to trust that system. And then last but not least, when we find that waiting times are unacceptably high, this is something we can get as feedback towards Taxi Central, and within our strategic partnership there, we can help them figure out how to solve problems of coverage. I'll get back to that as well and tell you how we do that from a dispatch point of view. Now, another thing I just wanted to point out with regards to perceived waiting time is where the UX people in the crowd are going to get really excited, and it's these small points of friction that are going to help them improve this experience by a millisecond or make things more clear. And that's basically about these buttons that were also mentioned in the Clue talk, for example. It's what kind of graphics do you use and how easy do you make it, what kind of affordances do you use for people to understand how to get through your app. Because the quicker you can get the job of booking done, the lower the perceived and actual waiting time. So you can compare these two things and I'll tell you which one I like better. But where are you going to press to book a car? There's good things to say for both. The most important thing here is experimentation. A lot of stuff has been said about that already today, so I'm not going to go into that. Then there's the dispatch part of things. Again, actual and perceived waiting time. We could make sure that our engineers made sure that our dispatch is super, super quick. So we're talking a matter of seconds before we found out which driver is actually best matched to go and pick you up. And this is based on some different scorings. We have a behavior score that is, for example, how often does a driver actually accept or reject the trip, and we have a passenger rating score. How well is a driver rated by the passengers? If we have two drivers, both two minutes away from a passenger, and one has a better passenger rating, then he's going to get the job. And this gives natural incentive for drivers to actually make sure that a customer that pays up to 500 kroner for a trip gets the service that he deserves. Besides that, we also have an app for drivers, and what we do here is we help drivers find out where to get their next fare. There's two ways that a driver can get his fare. You can do this by either drive the streets and go hunt for one, and this is what drivers love the most usually. It's something that's sort of a culture thing, and it's something that they're used to from before all this technology made other things possible. And then he can get a dispatch trip. So before a driver gets a dispatch trip, we're trying to sort of nudge him to work with each other instead of against each other, so instead of standing in a line of 20 cabs in front of a hotel to drive to a place where there's a good chance that a booking is going to come in. We have a lot of historic data that helps us find out where normally bookings are coming up at certain points of the day, and we of course also work with live data to help people understand, or to help drivers understand that better not go stand in line here, but drive up here. At that point, we dispatch the trips to the driver, and we ask him to accept the trip. They get 12 seconds, something that visually is also shown by a countdown, and drivers are in a very noisy context as well. It's something that I'll get back to, but basically it was very important for our driver app design to maybe not be as sleek and white and nice as possible as the passenger app, but to instead have big buttons and be able to get clear text that's easy to figure out when you're actually supposed to have both eyes on the road, something I'll get back to. And then last but not least, when a driver has accepted a trip, we make sure that we notify a passenger, and we do the same when your car is arriving. We can do that both through push notifications and SMS, and that depends on what the taxi central decides to do. And again, here, there's a huge strategic benefit of working with taxi centrals, is that they have years of experience of operating at a certain location. There's always certain social, economic, and cultural reasons to do things one way or another way, and this is what taxi centrals help drivers do, if I can turn it that way around. So then there's the next step, and at this point, a car driver is going to be able to get to the destination, and the car has been dispatched, it's on its way to the passenger, and it's time for the taxi to find the passenger and vice versa. Now, location-based technologies have made this pretty easy, but we're running into some issues here throughout a driver's life cycle. And one of those issues was that the moment that a driver accepts a trip, he'll get a location to drive to. He might decide to turn around and go there, and ten seconds later, he gets back to the destination. And then he gets a notification saying that the trip is canceled. You can imagine that as autonomous business owners, they weren't very happy with this, and they made it very clear that they weren't very happy with that, and it was an issue we had to deal with. One of the things that we tried to do was actually to figure out, okay, maybe there's a lot of people that haven't tried this app out before, so let us let them play around a little bit. Let them press the book button. We're going to actually show them that a booking has been placed, but meanwhile, we're holding up their booking in the dispatch engine, and we're not going to send it out for another ten seconds to a cab driver to make sure that if you change your mind, you do it before a cab driver turns around and starts driving towards you. Now, that was one solution, but obviously, it increased the actual waiting time, and it didn't feel quite right. So instead, what we chose to do is to work a bit more with the driver's life cycle, and we're going to work a bit more with psychology, if you will, and when you book a cab these days, from the moment you log into the app, you will see the five closest taxis to your location. So what that does is, together with the ETA we show you, you get a good idea of what the waiting time is, and at that point, you sort of accept, okay, I'm going to have to wait for these four minutes that I booked an app. Then when you book, we're actually going to show you where the driver is, and show him on your way to the customer. And what that does is basically to help you to realize that there's a human on the other end of the line. It's not just a remote control that you can just press and see if a cab shows up. It's someone you're going to disappoint while he's on the way to come and pick you up. And we saw a quite radical decrease in cancellations after that. Then last but not least, we basically also worked with a stick, and we told people, you can't cancel after five minutes. And people got to get out of jail free cards, so if they would cancel after five minutes for the first time, we told them, okay, this is fine, it's your first time you use the app, and next time you do this, we'll have to charge you 50 kroners, or whatever amount, for the inconvenience of the cab driver. Now, I was just told I have five minutes, and I'm going through my presentation a little bit slower than I should, and I'm going to do it at home. So let's see how far we can get. But basically, at this point, a driver has to navigate from A to B, and you have to enjoy your ride. I sort of already touched upon enjoying your ride, and that's through this rating system that allows us to fish out drivers that give bad customer service. If it comes to navigating, we've done a lot of user research together with drivers. We have a driver group that gets sent new updates to the app and drives around with them for a while before we push it out to everyone else. But mostly, I think, when I first came to Driver about half a year ago, I was really shocked looking at this app, because all I was used to was the passenger app, and I thought it was ugly, and I was going to redesign it, and it was terrible. And then, slowly, I started to realize why it looks the way it does, and that's basically to make sure that you keep your eyes on the road. There's this very popular slogan online that you read on a lot of blogs these days, which is the whole, no interface is the best interface. And this is especially true when you're driving a car. So one of the things that we've done here that I want to show before I move on is I really like this thing where, when a driver's driving, he can keep his eyes on the road. He knows that the bottom of the screen makes him end a trip, so he hits it. But in some situations, we actually have to present drivers with two or three buttons. So at that point, just blindly hitting something is going to cause trouble, because you might tell the app to do something that you didn't mean to. So I'm going to play a quick, short video to show what is required to actually end a cash trip now. And it's to hit the button and to hold it for a while. You see the loading bar there? And in that way, we allow people to sort of hit, maybe quickly glance, eyes back to the road, and make sure that they end the trip properly. Right. So now we're at the end. We've got your credit card on file, so you can actually just get out of the cab and move on to do what's important to you. So also this part of the journey, we've made considerably easier. So I'd just like to close with some quick personal reflections. And that's, first of all, it's going to tell you that I'm a product designer at Driver, and I'm still not quite sure if I'm a product designer or a UX designer or what it is. And that might have something to do with the fact that we're a small company, and we need to all stretch ourselves to do stuff. But it also has something to do with that I think it's important to find a balance between, on the one hand, reducing friction, and on the other hand, looking at a complex system as a whole and figuring out how you can create momentum, desire, and stimulating action. As a product designer, you're supposed to find, find places in a customer lifetime where you can improve his life and bring a new feature or product to the market. But doing that actually requires quite a big UX understanding, an understanding of your customer and what he or she goes through. So to close off, I think all of that should be taken a bit more holistically, and we shouldn't take our job titles too serious. And we should realize that design, and now I'm going to come with two clichés. I love clichés. I hope you don't hate them too much. Design is not one big decision. It's the sum of thousands. It's the sum of thousands decisions, thousands small decisions. And those decisions are taken by everyone else than you as well. A thousand decisions means 20 decisions as a designer, 40 from an engineer. And so it's important to break open the conversation as a big team and talk about design. And that's why I think and I hope, that there's more than just designers in this crowd today. And if not, then see your role as a designer also as a role that has responsibility for initiating discussion about design with non -designers and stimulating that conversation. For example, as we do on Slack, where you have a channel where you load up all sorts of different ideas, wireframes and finished products. Yes, I think that's where I'm going to close off. Thank you. Thank you. We only have time for one question, but Nicky will stay around a little bit later, so you can ask him afterwards. But does anyone want to have the precious one question for Nicky? Hi, I'm Julie. I'm from SurPosteria. It's one, but it's actually two. So just out of curiosity, what taxi companies are you working with now? And how has it been working with the actors of this industry? So we're working both in UK, in Lithuania and in Denmark. We're working with, for example, a cab company called Phoenix Taxis. And my experience, I can give you the whole list if you want. After the talk. My experience with working within this industry is that it's just really, really nice to meet people that live a completely different life than you. And I'd never dreamt to work in the taxi industry. But it's a great challenge to actually sort of force yourself to stop thinking about how you think things should work. But to work together with taxi drivers that spend nine hours a day or even more in a car and see what's difficult for them. What's difficult for them and how you can help them improve that. Thank you. Thank you so much. Thank you. Excellent. My notes. Excuse me. My notes are gone. Anyway, now it's time for us all to say thank you to our online viewers. The show ends today at this time. We'll be back tomorrow at 9 a.m. online. Same link. I was wondering if everybody would turn around and wave at the camera and say bye bye. Bye. Tak fai day. Excellent. Tak. Tu bism tak. Bye bye. Take care. Bye bye. Bye bye.
