Travel Entrepreneur and Triposo founder Douwe Osinga - Episode 5

Douwe Osinga is a travel entrepreneur, currently based in Berlin, Germany. Douwe founded 2 travel companies: World66 (sold to Internet Brands) and Triposo. In between his entrepreneurial efforts he worked at Google, where among others he was part of the team that launched Google Wave. In this podcast interview we talk about Douwe’s path to becoming a travel entrepreneur, how the ideas for World66 and Triposo were born, and about some lessons learned from the launch of Google Wave, among others.

In this episode we talk about:

  • How the idea for World66 was born, one of the earliest UGC sites, long before “web2.0” became a thing, and before sites like TripAdvisor and Wikipedia were launched.
  • How Douwe’s personal website helped him get noticed by Google
  • How to run a company consisting of a globally distributed team, and how their meetups help with team bonding and spur new innovations.
  • The nascent trend of “no-UI apps” and how these apps point towards a possible future of mobile
  • The benefits of launching a new feature as a stand-alone app vs integrating in the main app
  • Why there are so many Dutch travel entrepreneurs :)

Links mentioned in this Podcast:

 

 


Transcript

Joost Schreve:

Hi everyone. Welcome to the fifth episode of the kimkim podcast. My name is Joost Schreve, and I'm co-founder and CEO of kimkim. Today I have Douwe Osinga on the show. Douwe is a fellow Dutch travel entrepreneur, who started World66, an online travel company, I think in the late 90's. Then worked at Google for several years, and in 2011 started his second travel start up Triposo. Welcome Douwe, it's great to have you on the show today.

 

Douwe Osinga:

It's great to be on the show.

 

Joost Schreve:

Great, yeah it's always fun to talk to fellow Dutch travel entrepreneurs. There's quite a few of them actually.

 

Douwe Osinga:

Well it's like they say, we have a lot of foreign country.

 

Joost Schreve:

That's right, that's right. It's hard to stay in this tiny little country right.

 

Douwe Osinga:

Yeah, you fall asleep on the train and you're somewhere else.

 

Joost Schreve:

Exactly. So I'd love to maybe hear a little bit about your background. Tell us how you got into travel and becoming a travel entrepreneur.

 

Douwe Osinga:

Yeah sure, you gave the highlights away of course already. I guess travel is something that I grew up with. My parents always liked to travel, and they took us in their car all over Europe. They would drive all the way to Spain, or Turkey, or southern Italy, or whatever. You know, I think me and my brother, we both enjoyed that, and I've always been interested in technology too, and so combining the two seemed like a logical step.

 

Joost Schreve:

Makes sense. And so at what point did you feel like hey this is maybe actually a career as opposed to just something I like?

 

Douwe Osinga:

I had my own start-up before. I did World66, but we sort of got into this consultancy mode, and that's okay, you know it allows you to make some money and explore technology. That company still exists over in Amsterdam. But it's also limited in the sense that, as an entrepreneur you do feel that you would like to have your own product and set your own course. In the late 90's it became clear that the internet was more than a fad, although two years later it seemed like that was all a mistake of course, but yeah, we were just before that whole bubble. When it seemed that this was really something, me and my brother, we started thinking, can we turn this into a travel guide. We hit up on the idea of what you would call today a travel wiki. Except that Wikipedia didn't exist at that time.

 

Joost Schreve:

Right so this was in like in 1999 or something?

 

Douwe Osinga:

Yeah I think we started in 98' but then for real in 99'. So we went together with our, then girlfriends, on a trip through central Asia. We were in Kashgar, which is the outermost bit of China. They had these Turkic people there, and it was just really hard to get information. Because there was a Lonely Planet for that, but it was an old book. It was just obvious that the best way to get to information for travel was to talk to other travelers. That's all right, but with the internet in the back of our head we thought like, well what if we could get everybody to just dump their information in a structured form. That's where World66 sprung from.

 

Joost Schreve:

Right, and this was also before TripAdvisor even was founded right?

 

Douwe Osinga:

Yeah, yeah this was a really long time. Showing my age here.

 

Joost Schreve:

Maybe tell us a little bit about how got started with that? I guess you had a technical background so you coded the thing yourself?

 

Douwe Osinga:

Yes, yes indeed. We started, we built the thing, in those days, on top of Zope. Which is a strange, long abandoned I think, content management system that allows you to write code on top of it. It's fairly suitable for a travel wiki, especially if you want to put some structure in there. Then we tried to build a community around it, and you know we got somewhere. We got some traction. Of course, in the beginning especially, it was hard to make money, but then luckily Google came out with their AdSense platform, which worked really well for us. Then at some point Google also asked me if I would join them. Around that time we got an offer to sell the company, and that's what we ended up doing.

 

Joost Schreve:

Nice. How did you get in touch with Google, and how did they reach out to you?

 

Douwe Osinga:

I had been, there was this thing that Google hacks, that people do. There's a book about it and there were some other things. I experimented with that a little bit too. You had these little tools online that let you do interesting things with Google. Later I heard that somebody at Google said, "well we should either hire this guy or send him a letter that he should stop doing this" I guess they went with the first one.

 

Joost Schreve:

Nice.

 

Douwe Osinga:

I have not updated it for, well, I don't know eight years or so. A lot of these things are based on API's that might or might not work anymore. There are still cool things there, but it might be a little broken.

 

Joost Schreve:

Right, cool. At Google you worked on search, pretty hardcore technology.

 

Douwe Osinga:

Yeah, in line with the whole travel theme of this. I started in Google Zurich and I worked on search, and for the core search group. Then after three and half years I got the opportunity to work in India. I worked there for about a year on mobile, but since this was for India, we launched this product, which was a kind of mixture between Twitter and Google news. It ran all over SMS, that way we could actually reach a lot of users in India.

 

Joost Schreve:

This was before Mobile really took off I guess right?

 

Douwe Osinga:

The iPhone had already come out.

 

Joost Schreve:

Okay, I see.

 

Douwe Osinga:

For most people in India, data rates were very slow in India still, and the great majority of people with mobile devices in India had dumb phones. This was a great way to give them some access to a bunch of information. The product was called SMS channels, and you could create these channels, or subscribe to channels. Of course the cricket channel and the Bollywood channel, they were extremely popular.

 

 

I did that for about a year and then moved on to Australia, Sydney, where I joined the Google Wave team. We also had a lot of fun. That project, for those who remember, ended up being canceled. At that point I thought it might be nice to go back to running my own thing.

 

Joost Schreve:

Great. I mean I definitely remember Google Wave. Very cool technology and actually a lot of the elements that were created there are now kind of present in all our Google products right?

 

Douwe Osinga:

Oh not just Google products I think. I think if you look at products like Slack. I don't know if you've seen the presentation of Apple, when was it, day before yesterday?

 

Joost Schreve:

Okay, yup.

 

Douwe Osinga:

One of the things they showed, is that, if you now enter a URL in an iMessage that points to a YouTube video it will automatically expand into a YouTube widget.

 

Joost Schreve:

Got it, got it.

 

Douwe Osinga:

That was definitely one of the things that we did in Wave, what was it, seven years ago. I think a lot of these things that you see in collaboration products all over the place, you know we had them a long time ago.

 

Joost Schreve:

Right, exactly. When you look back at something like Wave, like that project, I guess it must have been kind of a wild ride emotionally. Like excited about building something great and then maybe a little bit of disappointment with it not getting any traction.

 

Douwe Osinga:

Well

 

Joost Schreve:

Any learning from that?

 

Douwe Osinga:

That's a good question. Yeah exactly. It was really nice to work with a small-ish group of people that were very talented. Somewhat removed from the day to day operations of Google with a lot of freedom, and be able to really explore all kinds of new technologies. I do think there's two things that I learned. One is, if you want to sell a product you have to sell it on one argument. Wave had so many different things and ideas that, if you look at the presentation that we started out with, I think a lot of people came back with, "this is awesome, but what is it". It works much better if you have one thing that you're really good at, and then over time you can add other stuff. Once you, you know, if you look at Facebook, they have all kinds of things, but they started out with one convincing argument.

 

 

The other thing is that, and this is also not rocket science or anything, that you have to find your audience. I think for Google Wave there was an audience of small groups of people that collaborate, probably in different locations, on different subjects. Kind of like how the whole Slack thing started. I think if we had focused on those people, instead of coming out on this wild, big presentation, we could have convinced those people, and worked from that basis. I think that's true for a lot of product development. You have to find your unique feature, and you have to find your core users. Once you identify those two things, and you connect those things to each other, then you have a good basis to work from.

 

Joost Schreve:

Right, right. Exactly just kind of start small and hyper focused, and then grow sort of once you have really nailed that initial product market fit I guess, and then expand from there. As opposed to just try to do it all at once.

 

Douwe Osinga:

Yeah, right, and once you're there you can add features, and you can explore new user groups, and it all flows from there.

 

Joost Schreve:

That's great, I mean it's very recognizable also for me at my entrepreneurial journeys.

 

 

Great. I guess after that you left Google and started Triposo right, or was there something in between still?

 

Douwe Osinga:

No, no that's correct. In a sense that is correct. Again after we sold World66 we had always been, we kept thinking about travel and technology. When the iPhone came out we were like, "huh, this is the way it really should be". A travel guide need to run on a mobile device. Having a travel guide that runs on your laptop is kind of useless, cause you can't really take it with you. We talked a lot about that, and we prototyped a little bit, but you know, as long as I was working for Google that would, of course, not really go anywhere.

 

 

Then when Google Wave was canceled, it seemed like a good moment to really jump into this, because at that point also, aps where really taking off. All the devices had GPS. They were getting fast enough to actually draw maps big enough to have offline capabilities. Which is one of those things we started out with Triposo, in the beginning, it's of course really important for travelers that they can actually take their stuff with them. So it seemed like a good time to start doing that.

 

Joost Schreve:

Right, exactly, it makes a lot of sense. You got together with your brother again?

 

Douwe Osinga:

Yup, and we also convinced one of my fellow Googlers in Sydney to join, and we got a designer that we knew from a while back. We had a smallish team with whom we set out to travel. Interesting angle of course there was, that since I no longer worked for Google I could not stay in Australia. The first half year or so that I worked at Triposo I was technically homeless, and I guess, technically unemployed. That didn't really stop us.

 

 

We started out with this culture where we tried to optimize everything so that you could work from anywhere. Which is really nice, and fitting for a travel company of course.

 

Joost Schreve:

Right, exactly. Yeah, give us some details, like where was the team based? You guys were all over the world, and still are I guess?

 

Douwe Osinga:

Yes, so we're mostly in Europe, although I just spent about half a year in Thailand to run a project there.

 

Joost Schreve:

Nice.

 

Douwe Osinga:

One of the interesting things about that is, of course that, with technology you can really work from any location. Time zones are a problem, but apart from time zones you can really do it. The social aspect still is tricky. Cause if you, over email, and Slack it's easy to misunderstand the other person, even over a video call. What we do, is we bring the whole team together every three months or so, in one location. We brainstorm, we talk about what went really well in the last quarter, what didn't go so well, how can we improve. We present strategies for the next three months, we brainstorm about those, come up with ideas, and then we bring those ideas to the prototyping phase of this, where we actually try to build some of this stuff into actual little products, that we then integrate into the main product or sometimes it remains has a separate product.

 

Joost Schreve:

Right, and so what are some example locations that you guys do those jamborees at?

 

Douwe Osinga:

Yeah so, the first one that we did was really nice. It was in Marrakesh, in Morocco, where we had a house in the old town. It was in January and we had been relying on the weather forecast off our iPhones, but it turns out that those are not really good. The weather forecast for the iPhone had been saying it would be twenty degrees and sunny, but it turned out it was fourteen degrees and sunny. Which is still okay, except for that these old Moroccan houses, they have this inner courtyard as sort of this central area. This is where we would all sit around with our laptops, but it would basically be outside. We had a little heater under the table, and we would just be cold all the time.

 

 

We've, of course, done this for many things. Another thing that I think was really nice, if somewhat unpractical, is we rented a bus, one of those roadie buses. Then powered it up with a bunch of 3G WiFi connections and drove around Spain and Portugal, and were just writing code from the road. Then we would just drive the bus to a beach, hang out on the beach for a bit, and drive it to a town, and then hand out in the town a bit. Meanwhile we would be working.

 

Joost Schreve:

Nice, that's a pretty good way to combine travel an work I guess.

 

Douwe Osinga:

Yeah, it's not the most practical way, but it's a really good way to bond with the team and get everybody to know everybody.

 

Joost Schreve:

Right, and so how many people are on the team now, how many people join those trips?

 

Douwe Osinga:

Not everybody can always do it, but we're thirteen people or so. It's a nice group to do this with. It's a little bit too much for most airbnb's but you can usually find something.

 

Joost Schreve:

Right, right, that's great. You guys do a lot of experiments then I guess when you're on those trips. Like hackathons?

 

Douwe Osinga:

Right.

 

Joost Schreve:

I just read about and tested your latest effort. Is there any notable feature, or anything that came out of it that you'd like to talk about?

 

Douwe Osinga:

Yeah, so like I said, we try to pick a theme for this sort of thing. One thing that had been fascinating some of us, is to think about aps that have no UI. I mean obviously they have a screen, but where the main way to interact with these aps is to just start them. Because, that is in a sense the most simple app that you can imagine. We decided to brainstorm around this and see what you can do. We built a bunch of aps that are around this theme.

 

 

One of the things that is still not quite done, but that we're working one, is a currency converter. As soon as you start it, it tells you the currency of the country where you are, and compares it to the country that your phone is connected to. It knows these things so you don't have to enter anything. It uses the camera to try to find prices, and anytime it spots a price it will just automatically convert it to your home currency. That's a really good example of an app that does not require any iteration other than starting it.

 

Joost Schreve:

That's really cool, that's actually really getting into the vision for mobile: it knows where you are, it knows where you're from, and it has sensors to figure out context.

 

Douwe Osinga:

Exactly, exactly, and another example of that is Walter, which is a travel compass, where we started thinking about what is the… If you would use the travel guide to its bare essentials, and you strip away any way to interact with it, what can you, what is left? What we did is we created this app that shows you on a compass like interface, what's around you. It does this basically only in colored blocks. It will tell you that if you walk into that direction, after twenty minutes you will meet, there will be bars, and in ten minutes there will be shopping, and just behind you in a five minute walk there is sightseeing. It doesn't tell you anything other than that. But if you are in a new city, and you're not really planning to see any specific sights, you just want to walk around, but make sure you don't end up somewhere where there is nothing, then this is a really good way to explore a city.

 

Joost Schreve:

Right, right. I mean I guess it's pretty radical, like ideally it would sort of be married with an approach where you can also get some more info about things. Do you see this type of feature as something that would be a stand alone pap or do you envision integrating it into your main app?

 

Douwe Osinga:

Right, excellent question. The thinking with these aps, before when we did the hackathon we would look at the results of it, and then think about, can we integrate this into the main app? Integrating new features into an ap that already had a lot of features is really hard to do in a sensible way. Building it out as a stand alone app first, and seeing if people like it, if it gets any traction, and then consider if you can bring it back into the main app seems like a much quicker way to get feedback on your idea.

 

 

In this case I could imagine that you can sort of flip your travel guide into a wander mode, which allows you to randomly walk around without really knowing what it is. Then as soon as you are done with that or you really do want to know what reviews are for this restaurant, or whether the church you are standing in front of is worth going into, or the museum worth paying for the entrance fee. Then you can just drop out of the wander mode and you get this detailed mode where you actually know all these detailed things.

 

Joost Schreve:

Right, right.

 

Douwe Osinga:

Another app that we built, if I can still smuggle that quickly in, is the landed app. Which is basically another way of stripping a travel guide to it's bare essentials. One of our guys, he joined me from our Thailand project, then you said, "you know, I don't want to fly back, I'm just going to take the train back." and we said you're crazy. But he did it, and it took him a long time.

 

Joost Schreve:

He took the train from Thailand back to Berlin?

 

Douwe Osinga:

Yes.

 

Joost Schreve:

Okay great. That must have been quite the experience.

 

Douwe Osinga:

It was quite the experience, but one of the experiences that he had, I mean he didn't want to do this. He didn't say look guys I'm not going to work for two months, I'm going to travel. What he said was, I'm going to take the train on the weekends and during the week I'll just work. He would just take the train on the weekend, go somewhere, and then in that city figure out how he could work there. He noticed that he was always doing the same things. He would always try to find a prepaid data card, he would always try to find a good deal on a hotel, figure out how to get from the train station to where he wanted to go, learn a little bit of the language.

 

 

Essentially what he then did was try to find, and this information is already in the Triposo main guide, it's just you have to find it. The landed app has only this information, and it sits in the background, and it waits for you to well, you know, land somewhere. As soon as you do that, it sends you a little notification saying that it has information about where you are, and it tells you this information. It will tell you exactly the things that he would be looking for. Where do I get a prepaid sim card, how do I get from the airport to the town, what is the currency rate, where can I get a good deal on the currency. If we have good hotel deals, last minute hotel deals, it will also pop up those. That is the entire interface of the thing.

 

Joost Schreve:

Nice, and places to get WiFi I guess would be another one that many people are looking for always right?

 

Douwe Osinga:

Yes, yes. That is not in this first version. But yeah, you could totally see ... Another thing that we had been thinking and sussing out in this version are local specific apps for travelers. Because some aps work everywhere, but often times there is a localized app that just works better.

 

Joost Schreve:

Right, exactly.

 

Douwe Osinga:

Cause Uber is nice, and it works a lot of places, but in a lot of places there is a local taxi firm that does slightly better.

 

Joost Schreve:

Right, absolutely. Cool, I'm going to make sure that we link to those aps as well, so people who listen can test out those aps as well.

 

Douwe Osinga:

Excellent.

 

Joost Schreve:

Final question, a bit of a broader question. I'm personally very invested in this, and interested also. So my question is, how far do you think we are in the quest to kind of really make mobile travel guides and mobile travel aps really ive up to their full potential? I've worked for seversal years at TripAdvisor where I launched City Guides, and then worked on the main app. I still have this feeling that there is a lot more potential than what is currently in the market, there has been progress, but it has been slower than I expected, and many people expected. Love your view on that.

 

Douwe Osinga:

Yeah I totally agree, I think, and you touched on this real briefly in the beginning, the whole vision that I think we both share, of having a travel guide that knows who you are, that knows where you are, and has all these sensors and figures based on that what is the next best thing for you to do? That is something that we've made progress towards, but we can do a lot better still. I totally agree with that. I think it's also that people, when they travel, they want to get a unique and authentic experience. Just showing them lists of the best places is just not going to get them there.

 

 

One of the thoughts that is also behind this Walter thing, and I'm not sure that this is the direction, and it's probably not, is to do the reverse. Take away information from people, so that they're taken outside of their comfort zone. Often that is the best way to really experience something new, that goes beyond the ooh, these are the seven top attractions according to Trip Advisor, I have to see all seven of them, and then I've checked them and I can move on to the next town.

 

Joost Schreve:

Yeah exactly. Totally agree, totally agree. Lots of things still to be built, which is great for people like us I guess, so we won't be out of a job anytime soon hopefully.

 

Douwe Osinga:

Exactly.

 

Joost Schreve:

Great, Douwe it was really great to talk to you, and some very interesting ideas here. I'm going to post the links a lot of the projects here in the notes as well. Thanks for joining us today, it was really a pleasure to talk to you.

 

Douwe Osinga:

Yeah, you too. Excellent to be on your show.

 

Joost Schreve:

Great. Thanks Douwe. Goodbye

 

Douwe Osinga:

Goodbye.