move things forward really quickly. If there was a problem you could fix it overnight. You could just be always shipping. At Facebook we would ship every day.
The difference in mobile is that you really have to have these things tied up in a bow when you put them into the App Store. If you don’t you risk two weeks of your users having a buggy experience, which can reduce your App Store rating, which reduces your distribution, and causes press stuff to happen. So it’s a really interesting, major difference from the way you approach Web software. And it slows down the iteration cycles. It causes you to have to take a much more careful approach to not just packaging up and finishing up a release, but in considering the priorities for each release, how much you are going to put into each release.
We found many things that we didn’t expect in the beginning. For the first six months we would do a release every week, just like we would on the Web. And we found actually that users were getting burnt out on receiving a bug fix update every week. So we started packaging up larger releases. Because when we found that when that little update thing comes up on the phone, users get really excited and they hope that something really awesome is in there when they go to look at the Path icon in the update, so we found that putting in one big awesome feature and then a few fixes and things that people have been asking for, turns out to be a really good way to do updates. Or you go full blown and you go for a big dot release like we did with 2.0 and you overhaul the whole thing. And you drop a whole bunch of new stuff on people. But we found that, packaging things up in a much more considered and deliberate way is the way to go.
And I will tell you, as I said, it has been a painful process for us to learn. Our first version, we released it in a very, in the traditional Web way, that if you are not embarrassed by your first product, then you waited too long. So we put it out there. And it wasn’t super great. It was definitely a minimum viable product. We didn’t even have comments. I remember journalists asking me, when we launched—“Really, you’re launching without comments?” So we did what we do on the Web. Of course we are going to add comments. But we found that it was really hard to get those iterations in, on a fast enough cycle that the market understood that new things were happening.
So our first version was, I don’t call it a complete failure, but maybe a 70 percent failure. And so we have kind of learned over time how to package up features and things that are substantial improvements in a way that doesn’t tax the team too much, doesn’t make the consumer wait too long. But I will honestly say, we are still learning. We are doing some big releases here in the next few months. Some of them, I think, we wish we could have gotten them out earlier. Maybe we would have shipped pieces of them earlier if we didn’t have to package them up into once nice thing.
So I would say that on mobile consumers are receiving things much later than they otherwise would have because of the processes that you have to use to develop them. But I think that developers who take the time to put in really high-quality work, that is what you have to balance against. You have to take more time. Consumers are getting things less often. So in between, you have to have a really disciplined focus on really adding a lot of value between each one. Sometimes we do well, sometimes we miss. But we have gotten to a pretty good cadence now.
X: On that subject, since we’re talking about the mechanics of product management in the mobile world, there are related challenges. It’s not just about how often you push a new dot release. It’s also about things like onboarding and retention. So, I wonder what you’ve learned about how to break through all these barriers. One barrier is just making sure someone opens the app after they’ve downloaded it. Then the next barrier is trying to get them to come back every day. Those are all things you can measure, but things it must be really hard to influence.
DM: Yeah, that’s a big question. I’d say the hardest of all of those is the new user experience. We call it the NUX. Immensely hard on mobile. For a very simple reason, because mobile is like the old world of software in that everyone goes into a store and downloads the app. And everyone comes into the app in a totally new state. We can’t pass any data into that. We don’t know who you are. When you first open up the application we have no idea who you are, where you came from, why you are there, no friends, nothing. It’s not just hard, it’s impossible to personalize the experience for you when you first land on the first screen.
And so, designing an experience that’s completely universal for every kind of user, coming through the new user experience, it’s as close to impossible as a problem can get. We’ve gone thorugh hundreds of iterations of what our new user experience looks like. You know, for us, because we’re a social product, the real key thing that you have to achieve is to help the user understand why this new social product exists. For us, it’s setting a lot of context. We tell the user this is a place for family first, close friends second. And here, we’ll help you find those people, and if they’re not here we’ll help you invite those people. Then we teach them about the journaling aspects and the value we can add there. If the user doesn’t get connected to anyone, our chance of losing them is close to 100 percent. Because we can’t figure out who invited them, where they came from, it’s very hard to close those loops through mobile. I anticipate that it will probably get better, but right now we are still in 1999 on mobile.
X: How will it get better?
DM: I don’t know this but what I hope for is, Apple and Google are the two big platform providers in mobile right now, and what I hope is that as their platforms evolve they sort of create ways for developers to personalize the new user experience. And do linking between the experiences and things. The one way that mobile is very different form the Web today is it’s very hard to link data between two apps. It’s a very hard thing. Whereas the Web was built on this fundamental principle of linking between pages and documents and data. And so, it’s still very hard to do that on mobile. There are ways to do it but there are no standards, there’s still a lot of work today. This is a key problem for everyone. It’s a real distribution blocker. There are all of these things that could be done to improve it. I think that they will. It seems like something that would be important. But it makes it very hard.
So that’s one piece. To your second question on retention and this type of thing: I think mobile actually is really quite amazing, because it’s with people more you actually have more time with people than you had [on the Web]. The sheer volume of time that people have their phones with them is
