February 1st, 2013
What makes a user want to download an app in the first place? Ideally, it’s the promise of fulfilling a goal or need for the user. With the hundreds of thousands of options available, and the immediacy of the mobile context, you have a small window of opportunity to engage your user. If users can’t easily use your app, they simply won’t.
Mobile devices are stuffed full of sensors and cameras, yet forms tend to be the prevailing means of input on these devices. Creating an account, uploading a profile photo, and registering, right as you first open an app creates an incredible barrier to use. Luke, in his app Polar, opted to let users experience the app itself. Because users see what it does first, there’s a greater conversion rate.
Luke says that with 80% of mobile app use time being focused on entertainment, the phone has really just become another screen in our lives, like the TV. Games and social networking make up a lot of the time users spend on their phones. Luke suggests that companies consider this aspect of entertainment when developing their own apps.
Jared Spool: Hello, everyone. I want to welcome you yet one more time to another episode of the SpoolCast and today, today, today, we have our returning champion, Luke Wroblewski, to come talk to us. Luke has won every episode of the SpoolCast. The high-score file is his. I don’t know, how many shows have you done, Luke?
Luke Wroblewski: That’s a great question. I’m guessing it’s around 10-ish by now?
Jared: Yeah, it’s got to be. It’s got to be something, yeah.
This time, we have Luke coming back because he’s got news to tell us about his new company. He’s going to be speaking at the upcoming UIE UX Immersion Mobile Conference, which is going to be April 22nd through 24th in Seattle, Washington, and he’s going to be talking on designing mobile input.
Today we’re going to talk about his new venture. This is a new thing. If you haven’t heard about it, it’s really awesome. It’s called Polar. It is this little phone app that is very cool. I hear you guys have hit a major milestone this week with Polar.
Luke: Indeed. The application allows people to collect and share feedback through little polls, hence the name Polar. The major milestone we hit last night, actually, is we crossed one million votes. One million votes is a pretty big number, but the more interesting thing about it to me is that the second half-million of that total came in the past eight days, whereas the first half came in the first 11 days. People seem to be voting an awful lot.
Jared: It’s really cool. I’ve played with the app a lot. It’s one of those things I fire up when I’m looking to kill a little time. People are posting pictures and, “Do you want this? Do you want that?” You just click with your thumb and all this stuff.
You’ve been working on this for quite a while now. How long have you been working on it?
Luke: My co-founder, Jeff, and I got together around the March time frame, and we were sort of batting around, “What do we want to build? How do we want to build it?” I think we actually started making things around April or May. That took us up until November. Six-ish months, something like that? Seven-ish months?
Jared: For years, you’ve been telling other people how to build mobile apps. And this is your second one. You had Bagcheck before, and now you have this one. What are the big things that you had to sit back and say, “Yeah, this is what I tell people all the time. This is what we’re going to do”? What were those big lessons that you were able to apply?
Luke: At a high level, I don’t think I have anything to share unless I’ve tried some things and learned stuff, right? At a very big-picture level, I’m always trying to make things, so I can learn and actually have the lessons firsthand, so that any information I’m sharing with other people actually has some basis in reality.
The interesting thing about that is I think it really comes from me being a designer. The designers I work with, at least most of them, and the way I work is, if a designer’s faced with a problem, what we do is we just get our hands dirty and jump into it. We start to iterate. We start to do little sketches. We start to do little prototypes.
Through that process of building and making and designing, you learn a bunch of things. You uncover connections you didn’t know were there. You uncover problems you didn’t know were there, and you go on tangents that ultimately lead you to more valuable stuff.
I think it’s very, very healthy to always go and do what you’re talking about, because it level sets you, right? It lets you know if you’re actually talking about the right thing and also gives you a whole new set of stuff that you uncover, and ultimately you’re like, “Oh, I learned this. I want to tell somebody about it!” So it’s good.
Jared: Of those things that you’re trying out, was there anything that you were surprised at either how well it worked, like it surpassed even your expectations?
Luke: Yeah. Let’s talk about a good example and a bad example, something that I thought would work and didn’t, and something that I thought would work and is working. Actually, I’ve got a whole slew of these insights, which is kind of cool, so maybe we can talk about a few of them.
On the side of things that are good and are working, for a long time I’ve been advocating this idea of “sign-up forms must die.” The general principle there is, instead of having the first-time experience with a Web application being “sign in, fill out this form, register, create an account,” before you even know what the thing does and how it does it, let somebody in and let them get going, learn about what you’re doing and actually interact with stuff.
That was the approach that we took with the mobile app, Polar. If you download the app, you can open it up and you can just start voting, voting, voting, all you want. There’s no form. There’s nothing out the gates. As you interact with it, we’re actually storing all of your results in a unique token, if you will, or a unique-to-you set of events.
If, at any point in time, you decide, “Hey, I kind of like this. I want to create my own poll, or I want to leave a comment, or I want to share,” something beyond this initial voting process, we’ll actually take all the votes that you have and map them over your account so you don’t lose any of your activity.
What that’s done for us, I think, from a data perspective, is actually pretty compelling. I saw, there was recently a post written by an entrepreneur who was talking about the problems they faced building mobile-first companies. The way he defined mobile-first is “build a native app first.” Which, I don’t want to get into the debate of those definitions, but that’s what he was doing.
He outlined some of the problems, and one of the problems he highlighted was, just because somebody downloads your app doesn’t mean they’re actually going to use it. He cited some numbers from a big app manufacturer that has over 100 million downloads, that their stats showed 50 percent of people that downloaded it didn’t even open the app.
Other data that I’ve seen, when I worked with a really large, very popular mobile app earlier in the year, who I can’t reveal who that is, their sign-up flow in their mobile app had something like 25-percent conversion, which is really low, right? It’s hard enough to get somebody to download your app off an app store when there’s so many hundreds of thousands of apps.
And then the other piece of it is OK, you finally got that download out of them. That was hard enough. People download like one or two apps per months. Actually, it’s two and a half apps per month. You got that download out of them. That was a lot of work. You had to work really hard. Now you lose 50 percent of people from signing up because you have this login screen upfront.
What we did with Polar that’s different is you open it up, you start using it. The numbers that we’re seeing — and this is in our first three weeks or so — is 85 percent of people that open the app actually vote on it. Much, much better than the 50 percent or the 25 percent or those kinds of dropoffs.
That’s an example of me taking one of these ideas that I’ve talked about for a really long time, “kill sign-up forms,” putting it into action and actually having some data behind it and seeing what happens in real-world use and experimenting to see what we can learn and how we can make it better.
Jared: Yeah, I completely get that. I watched my fiancée play with it. She downloaded it. I was like, “You’ve got to see what Luke’s doing.” She was like, “OK, fine. What is this?” “It’s polls.” She’s like, “OK. I don’t think I need to use this.” Then five minutes later, she was like, “Oh yeah, I find myself voting on things.”
And so now she’s using it.
Luke: Yeah. I think you don’t get that value if you don’t let people in right away. Everybody’s naturally a skeptic, probably, right? Like I was saying, there’s just so many things out there fighting for our time. How much effort is somebody going to really put in if you just keep throwing barriers in front of them?
I download this app. “OK. Connect to Facebook. Connect to Twitter. Fill in this form. Upload your picture. Tell us about yourself.” Seven steps later, you still haven’t done the thing that the service or the app allows you to do.
One of the big ideas with mobile, in terms of mobile input, is make it fast and easy. Also, make it fun. The fun factor is actually something that I’ve been talking a lot about recently because a lot of people downplay this. But I have a theory, which now I actually have some data to back this up, which is that I think mobile is another screen for entertainment in our lives. Just like the TV is an entertainment screen, the mobile screen, in many ways, is an entertainment device.
There were some stats that came out three or four days ago from Flurry that showed time spent in the US in mobile apps is encroaching on time spent watching TV, which is just crazy. In addition to that, mobile devices and TVs are actually kind of hand-in-hand. 40 percent of Americans use tablets or smartphones while watching TV at least once a day, which is also huge.
When you look at that, OK, people are using these things a lot. What are they doing on them? Well, 80 percent of mobile app time is spent on games, social networking, and the generalized entertainment category. I think the actual number is 79 percent of mobile app time, therefore, I say, is spent on fun. I think that’s a reality you have to be cognizant of. Mobile is the new entertainment in many people’s lives.
Jared: This fun thing. Is this fun thing something that applies to every application? If I’m working on an app for an insurance company or something like the American Red Cross, or even something like TripIt, which tracks your travel itineraries, is fun really a piece of that, or is there something else to it?
Luke: I guess I love the word “entertainment” more than “fun,” because “fun” implies the wrong thing.
Let’s take some of these examples. I hadn’t thought about this at all in this conversation right here, but you may actually use this as a filter for whether or not you should build a native app for mobile. If you’re an insurance company, and your primary objective is to get information out about your insurance, allow people to sign up, all those sorts of things, do you really need an app? Maybe you need an app to allow people to take a picture when an accident happens and file a claim quickly, but that happens once every 10 years.
Jared: Well, I think, for insurance companies, I could see claim and account management, being able to use my mobile app to find out when my next payment is and what the balance is. Because some folks, when they’re paying their auto insurance, they’re working hard each month to get that payment put together, so they need to know what date it’s going to be pulled out of their bank account or when it’s due. That’s a piece.
Then you’re right, the claim thing. Which, my car is currently in the auto-body shop, so right now I’m neck-deep in auto-claim-related stuff. Knowing what the status of various things are, I could see an app for that. And yeah, I would only use it when I have an accident, but now that I’m using it, it’s a big deal for me.
Luke: Right. I don’t doubt that it’s useful in some of those situations. Like in your case, you got a car in the shot, but how often does that happen? From the company’s perspective, is it worth the effort of building and maintaining these apps. I don’t know the answer to that question.
Jared: Right. Well let’s say the ROI is there. Let’s say that if you can stop someone from calling the 800 number to get, for instance, the information about how many…Let’s take a simple thing that happens in an auto claim. You get a rental car approved. What is that maximum dollars per day of the rental car approval?
I’m standing at the rental car agency and the guy says, “Well, what’s your insurance company going to pay, the maximum number?” I have a choice now. I can either go into the app or I can call the 800 number.
The 800 number has a cost to it. The app doesn’t, or as much. You’re talking about an up-front cost to develop the app, versus an ongoing cost to answer that 800 number.
So the ROI could be there. So let’s say the insurance company says, “Yeah. That would save us a ton of money”. Now, where does this entertainment thing come in?
Luke: In each of these cases I think you have to look at what are those use cases, right? You’ve mentioned a couple of good ones and that might justify the ROI. But if I was that insurance company, I wouldn’t expect people to be using that app every day. I wouldn’t “blow” it all on mobile. That’s the lens that you look at the majority of use at.
The things that are going wild on mobile is things like social networking and games because that’s the things people can pick up at any point in time throughout the day. If your goal is to create a ton of engagement with people through mobile devices and you’re an insurance company, you may be swimming up the wrong creek with those kinds of use cases.
Jared: OK. But here’s another one, and maybe I’m an exception because I’m traveling so much. But I use TripIt multiple times per week. I’m looking up my travel itinerary and different details about it multiple times per week. Where does the entertainment factor come in there?
Luke: Well, I would say that that’s not in the entertainment factor. I would say that piece is in the other 20 percent of mobile time spent. If you look at a chart here, they have in there, 10 percent is utility. I would characterize that as probably utility.
You have productivity. You have news. You have health and fitness. You have lifestyle and you have other.
But when you look at the big portion of the pie, when something is consuming 80 percent of it, it’s something that a lot of companies should at least consider. It may not be that they do anything in that space at all. Clearly there’s traction in mobile as an entertainment device.
It doesn’t in any way negate the fact that utilities are useful on mobile. But it’s less likely there going to become an everyday, multiple times a day kind of engagement which is what everybody’s dream is for a mobile app they create.
Jared: Well, is it? I’m wondering if that’s…
Luke: It sure is out here in Internet land.
Jared: Well, OK. Yeah, but you live in Silicon Valley. Things aren’t real there. I’m wondering if…
Luke: Yeah. If you’re a news site you want people to read the news every day.
Jared: Absolutely, yeah.
Luke: If you’re a fitness training program where people are tracking their calories or activity, you want them to use it every day. If you’re a productivity tool, like a calendar manager or something like that, you want people to use it every day.
I don’t think it’s just one niche of things that people that are in it, not just like the Silicon Valley social app thing, that has that goal of daily engagement. I think there’s lots and lots of domains that are aiming for that.
Jared: Yeah, but I’m thinking of a lot of the big blue chip companies that have things that could warrant having a mobile presence. They have desktop apps.
One of my clients is in the yoga studio business and people could use their mobile apps to make appointments to go see the studio, to find out, if they’re traveling, where the closest studio is. That’s not an everyday thing maybe but it could be a once a week thing. It could be a once every other week thing.
Luke: Yeah. In all those cases you’ve got to justify that cost versus the benefit you gain from it. I totally do not mean to diminish any of those kinds of use cases. I think they all make a ton of sense. I’m simply pointing out that if you look at where the bulk of the pie, the activity in mobile is happening right now.
People are spending a lot of time.
To be fair here, entertainment apps and utilities actually gained some market share at the expense of social networking and games. That might be an interesting trend to look at.
Last year games were about 50 percent of time spent in mobile apps, while social networking was about 30 percent. This year games is down to 43 percent and social networking comes in at 26 percent. What you’re describing actually could be gaining a lot more traction, meaning utilities have started to come up.
That may be a sign of a larger trend too which is mobile’s really starting to mature now. It used to be everybody’s like, “What the heck is this thing? What can we do there?”
Now, just this week, we’ve seen a couple of big companies release some things which to me are really mature mobile products. Flickr put out something new. Google maps for IOS came out.
They’re really well designed in most cases. They’ve really though through mobile use. They’ve taken advantage of touch. They’ve taken advantage of cameras. They’ve taken advantage of the gyroscope. All the stuff that I’ve been talking about for years is starting to come to fruition.
Not to toot my own horn, that’s a result of me or anything like that, but it’s very cool to see it as somebody’s who’s been excited about this space for such a long time.
Jared: I’m wondering if we’re reaching that stage where in order to put something out now you have to have a higher level of sophistication than you used to have to have, both for native and web apps. In order to just play the game, to be part of the game, you have to really have stuff together. Because you will stick out like a sore thumb and look so 2010 if you don’t have something that takes advantage of this latest thinking.
Luke: Or, God forbid, you look so 2008, right?
Jared: Yeah, for sure. Or, God forbid, it’s looks like Whap right with the press one for our main navigation, yeah.
Luke: 2008 were the first round of IOS apps that came out for the iPhone.
It’s interesting that you say like the bar has been raised because in my mind it feels that we’re finally getting started. We’ve had this period where what most people were doing with mobile was port over their desktop experiences or take all their existing web knowledge and fit it in these small screens and weren’t really making mobile things.
Now, when you look at the types of products that are coming out, they are very clearly mobile. The new Google maps for IOS, it takes advantage of the gyroscope, you can move your phone around. I use it as a digital compass to orient you. It gives you 3D space to move through. It’s got street view that uses the gyroscope.
The way you send feedback is you shake the phone. Which maybe that’s silly but they use tone of gestures and they had this great quote when they talked about it. The director or product manager for Google maps said something like “There are no menus. The map is the user interface”.
And for me, a guy who’s been super excited about this stuff for many years, I’ve been like, “Let’s use all these sensors to rethink the interfaces. Let’s use touch. Let’s kind of create different kinds of experiences based on what these devices can do.” Now it feels like that skill set is in place and people have learned how to do that.
The apps that are coming out now are just way, way ahead of what we had even a year or two ago. I think that’s great for everybody, because, like you say, it raises the bar for the industry. But I also feel like now it’s the start of this truly mobile set of services and applications as opposed to things in which they’re trying to figure mobile out or trying to retrofit themselves into a mobile use case when they didn’t have one from the beginning.
Jared: It feels like where we are now is at this really neat turning point in that we’re going to get a chance over the next few years to really invent a new vocabulary of what you can and can’t do in mobile. That’s not even including the hardware we haven’t seen yet. That’s just dealing with the rich capabilities of the devices that we have now.
Luke: Right, exactly. People will raise this stuff, “Well, now you’ve got 50 percent of people in the US have a smartphone and 30 percent globally. Isn’t this mobile thing starting to play out?” Again, I thought, “Well, no, it just feels like we’re getting going.”
A, 30 percent globally, that’s a small number, relative to who you can reach, and then, B, as you were describing, right now, the toolkit’s in place, the ideas are there, and the thing that’s changed is now we have examples. Now we have proofs out there of people doing this stuff and being successful with it. You start to build on top of that. The foundation’s there. Now we can actually start building.
Jared: When you were building Polar, I remember seeing some of the early stuff, and it was dead simple, and you were just getting it out there and having people play with it. You built it very incrementally. You didn’t try and come out with this very rich set of functionality in the first release. You tried one thing and then you tried something else. Was that frustrating, to take all that time, or was it energizing?
Luke: Couple of things in play with that. A, the way we released the app was pretty measured. What we did in the first iteration is we put it out to tens of users. Whenever we put it in TestFlight, it got 40, 50 people in there, learned a bunch of things, made some changes. Then we didn’t really launch it, but we put it out.
It’s got a couple people in there, so we got to the hundreds. So it wasn’t really a launch, but we were testing the waters at a different scale. Instead of tens of people, we got to hundreds of people. Again, we learned some things, made some changes. Then we did our actual launch, got to the thousands level, learned some things, made some very quick and, in some cases, tough changes, and now we’re moving to that next level and doing that same level of iteration.
I really like that process because I think you learn different things each time you make that order-of-magnitude jump. It’s a totally different experience when there’s 10 people, then 100, then there’s 1,000, there’s 10,000, and onward and upward. That process I actually like.
The other piece to that question that you raised, in my mind, is things that look simple on the surface, I think that’s a real big deception, because in order to make something very simple and clear and focused, like there’s so much stuff going on on the back end.
In fact, Anthony Kosner at “Forbes” wrote this great little article, and I applaud him for getting at the stuff that’s going on behind the scenes with our app, because he titled the article, “Polar: A Simple App that Shows…” I think it was like, “how much actually goes into making a simple app.”
Jared: Yeah, yeah. I like that article.
Luke: He outlined all the stuff that we’d been doing, and that actually has a big impact. When you drill into that, you realize, it seems super-simple upfront, but man, it takes a lot to get there, right? I appreciated his perspective on that.
Jared: What surprised you, that didn’t quite work out the way you expected?
Luke: There’s one thing that we did that I think is still a failure and we need to address it. Similar to that Google Maps example I was telling you earlier, where they don’t have any menus, right? They make the map the UI.
We have, in our interface, it’s a photo kind of a voting interface. We didn’t really put any buttons that say “vote,” and we don’t have any calls to actions and “vote on this thing.” Instead, we want people to interact with the content, touch the picture or touch the side that you want to vote for.
When you do it once, then you get it and you get going. But what happens when people share these polls out on the Web, we get feedback that’s like, “Well, I didn’t know that I could vote. There wasn’t anything that told me I could vote.” While we have this dream of “Let’s make the content the UI,” or “Let’s get rid of this chrome. Let’s not have buttons. Let’s not have instructions in our interfaces,” there’s still this real-world reality check, that when somebody sees something in a browser and there isn’t a control there, they get paralyzed.
Many of them get over the hump and they click or tap or whatever, and they figure it out instantly, right? But they still need that first, initial push of either directions or an explicit control to interact with stuff. There’s probably similar things going on in the Google Maps app for iOS, because there’s a lot of things that are hidden.
Ironically, somebody actually made a poll on Polar that was like, “Have you found the 3D mode yet? Do you know how to activate it?” [laughs] It was a poll like, “Have you found the hidden gesture-based feature inside the Google Maps app?” We haven’t hit that point yet where people are just comfortable with, “Oh, everything’s part of the UI,” which I’d love to get to.
Jared: Yeah. When things are hidden like that, you run into the problem that I’ve been calling “socially transmitted functionality,” where the only way you can learn about something is by actually clicking on it, or hearing about it from someone else to know that you’re supposed to click on it. A lot of the “pull to refresh” or “this magic swipe at this particular moment” is how you learn how to do that.
The funny thing is, I’ve been watching people do this, and it’s like, “I can’t figure out a way to delete things from my address book.” “Oh, you just swipe across and you press delete.” It’s like, “How did you figure that out?” “I don’t know, somebody showed me.” It goes on and on.
Luke: We get that stuff all the time. I started developing little hacks around it. I’ve actually written these two things up. I’ll give you two examples. One of the things we have in Polar is pull-to-refresh to get the results list to download. What we did there is we teased a little bit of an image when you actually pulled on the screen, so it looks like something’s happening. It’s actually the bottom of a bear’s butt.
Luke: As you keep pulling and pulling, you start to see that this bear is hanging onto another bear by his shorts, and that bear’s hanging onto the rim of the top of the screen, and as you pull-pull-pull-pull-pull, more of that picture gets revealed to you. It’s actually your socially- transmitted functionality thing, taken to the level of, “OK, we want you to socially transmit it.”
What we’ve seen with that is we actually have people making polls saying, “Have you seen the full bears picture yet? Have you kept pulling to refresh so that you can see the whole thing?” It got people talking about it.
Another example that we do is we have this create a poll where you enter text at the top for what question you’re asking, and the keyboard comes up. For a long time, people were asking us, “How do I make the keyboard go away? How do I make the keyboard go away?” Certainly, you can just swipe down across the screen and the keyboard goes down. But, again, hidden.
What we ended up doing is, when the keyboard comes up — there’s a profile picture up at the top — we replace your profile picture with a “keyboard down” icon. It slides up as the keyboard slides up. You hit it and then the keyboard slides down.
I bring that up as an example of a just-in-time action, right? Shouldn’t be there at all times, but when the keyboard’s up, we’re going to show it to you really quickly and give you a way to get out of there. It’s not that that control is always on the interface. It just appears when it’s relevant, to make these hidden gestures more visible to you directly.
Jared: This is very cool. I can’t wait to see how you are taking the lessons from the stuff that you’ve been talking about so much and put it in and start publishing what you’ve been learning, because I think that it’s great that you’re using this as a laboratory for your own techniques and your own methods and then can turn around and talk about what’s working and what’s not for you, and then people can try it out themselves. You are a good man, Mr. Wroblewski.
Thanks. I’d like to put that on the tombstone if I ever get one. That’ll work out nice.
Jared: Yes, yes.
Thanks for taking the time to talk to us today.
Luke: Thank you, sir.
Jared: Excellent. I want to thank our audience again for spending the time listening to us. Hey, if you get a chance and you’re listening to us on the iTunes, please, by all means, take a moment and fill out the little rating on the iTunes, because that really does actually help us get feedback on what we should do better and can do all sorts of things. We pay attention to that.
You can check out Luke Wroblewski. Of course, he’s written the fabulous book “Mobile First” and “Web Form Design,” but he’s also going to be speaking at the UX Immersion Mobile Conference, which is going to be April 22nd through 24th in Seattle, Washington. He’s going to give a full-day workshop there on designing mobile input, and we’ll talk to you there. It would be awesome to see you there.
Thank you very much for coming and listening to us today, and as always, thank you for encouraging our behavior. We’ll talk to you soon.