Paul Kemp: Welcome to another episode of The App Guy podcast. I'm your host, it's Paul Kemp. I do go around the world - this is a pure global show, and we help everybody in the world who is an aspiring app entrepreneur, app developer, indie developer, small developer, maybe working on side projects... This is the podcast for you, do stay tuned. We try to bring you the best content that we can possibly get from the people that know what they're talking about.
I've got someone who really knows what they're talking about today, because he is the co-founder of Pyze, and we're going to talk about the current state of the App Store, we're going to talk about maybe obstacles that you're facing as a developer - I think that will be relevant - maybe how to create apps, get traction, the role of automation, and a bit more about Pyze.
Let me introduce Prabhjot Singh - he is the co-founder of Pyze and he's here on The App Guy Podcast. Welcome to the show, Prabhjot.
Prabhjot Singh: Thanks, Paul. Glad to be here, and excited for the opportunity to talk to you.
Paul Kemp: Wonderful to have you on. Let's just start with Pyze, an elevator pitch. How would you sum up what you do at Pyze?
Prabhjot Singh: Our mission is pretty simple at Pyze. We solve the biggest problem that any app developer faces - people download a lot of apps, because there's a lot of apps on the App Store, but people use very few applications, so we solved the challenge of retaining users by automating analysis, engagement and personalization for mobile apps, so that anyone can have the same sophisticated tools that big companies like Facebook, LinkedIn or Zynga use to make their app sticky. Whether you're a Fortune 1000 or you're two guys in a garage, we can give those same capabilities to you.
Paul Kemp: It's really interesting you mention that, because Facebook is definitely the dominant player on the App Store, they always are top of the charts. Do you think there's room for indie developers to get traction and have success with their particular apps on the App Store?
Prabhjot Singh: The deck is definitely stacked against the indie developer today. If you go back years ago to the beginning of the App Store - I'm talking 2006, 2007, 2008 - you actually saw a lot of indie apps being successful and topping the charts. Unfortunately, over the last few years, we've seen what I like to call the corporatization of the App Store, where you've got these big players that are completely dominating the space, so it's very hard for the indie developer to break in.
One of the big issues is data. The big companies can build these big data platforms, they're collecting all kinds of information about you and me, and they know exactly what kinds of ads to show us, they know exactly what content to show us, they know how to personalize the experience for me, so that it's an enjoyable experience.
Unfortunately, indie developers just don't have those same capabilities. They're just focused on trying to get a product out the door, probably trying to achieve product market fit, but they don't have to data scientists, they don't have access to these intelligence platforms that are using machine learning, and all these other fancy technologies to really individualize each person's experience. Without that, it becomes really hard to compete, and you compete for screentime, really.
Paul Kemp: Yes, that's a good segue to talk about the current state of the App Store. The Apple App Store, the Android App Store - how do you see it right now in terms of how it's looking?
Prabhjot Singh: It's still a wide-open market in the sense that -- we often like to talk about the app stores as the Wild Wild West. As it sort of matured, there is more predictability in terms of the type of revenue that you can generate, and app stores are still the best way to get distribution for a product easily in the history of software. Because you can immediately get access to billions of devices by uploading your app in just a couple of places. We've never had that capability before as developers, as innovators, as entrepreneurs.
If you think about, how many apps are being downloaded, 2016 was a record high; how much time is being spent in apps? 2016 was a record high. But also, how many apps are being uploaded to the app stores? 2016 was a record high. So not only is the opportunity the hottest that it's ever been, the competition is also the hottest. As with anything, everyone wants a piece and you have a lot of people with great ideas that are executing, but I think the app stores still present a great opportunity to get services out there, a great opportunity to generate revenue.
Paul Kemp: In terms of the obstacles that maybe you see developers facing - certainly small developers, indie developers that tend to listen to The App Guy Podcast - what are those obstacles as you can see it?
Prabhjot Singh: There are two or three things that we hear again and again when we talk to smaller app publishers, teams of maybe 1-5 people, or 10 people even. One is understanding who their users are because you can't really optimize your application or get a product market fit without really knowing who your users are. Unfortunately, the app stores create this firewall between the publisher and the user, where they know everything about the user, but the publisher actually knows very little.
You might use some analytics solution that will tell you where they click into the app, but that's about it. So that's one of the big challenges that all publishers face: who are my users, what is their behavior? Once you know that, the next big challenge is "How do I engage with those users in a meaningful way?" Through push, through in-app messaging, through e-mail, through SMS/MMS...
These are all challenges for small app publishers because the services that enable this generally cost thousands of dollars. I've built apps in the past and I've had apps with millions of users, and I still wouldn't be able to afford paying a few thousand dollars a month to send out push and in-app messaging and those types of capabilities.
This really gave genesis to the idea of Pyze. I had a problem myself, not understanding my users, not being able to communicate with my users, so the idea of Pyze was really to level the playing field and give those capabilities to everyone. Then we've added a bunch more features to enable personalization, so that you can really start to personalize the experience of each user.
Paul Kemp: It's so interesting to hear you speak about more focus on engagement than downloads. We've had over 500 episodes of The App Guy Podcast, and in the early days it was all about downloads. But the way I see it, a download is just one person downloading at one point in time; they may never come back, so I love to hear about engagement.
I would love to know now how can we create apps that do actually get traction? Any tips for us?
Prabhjot Singh: That's a really good question, you hit the nail on the head. Installs is a vanity metric. Unfortunately, a lot of the metrics that we measure in the industry are vanity metrics. Even a metric like daily active users, which was considered the standard in the market - it's even used in earning calls for these public app companies, and so forth - I think is a fairly unimpressive metric.
Let's say there's an application, you download it, I download it; you use it three times a day for 20 minutes each session, and I used it for one second. We're actually both counted as the daily active user number or the monthly active user number.
When you get a download -- and you can buy downloads all day long. If you've got enough of a budget, you can pay Facebook, you can pay Google and you can generate as many downloads as you want.
The real metrics to care about are engagement and retention, and the way you drive retention is by focusing on it at the design time. "How do I want to create cycles of engagement within the application, so that I keep users coming back?" Facebook's great at doing that. My friend Ted just uploaded a photo; well, I want to see what it is. I might get an e-mail, or someone might share his photo and I'll get a notification... It's a great way to get me re-engaged. When I'm there, next thing I know, I've lost an hour looking at Ted's entire wedding album. [laughter]
So how do you create that for your application? That's important to think through in the beginning, but as you get users, there's a lot of cool things that we can do to ensure that users stay sticky.
For instance, we know that a third of the people who download any app will ever only use it once. You've literally got 5-10 seconds of that first app experience to make it count. You should automate an onboarding campaign to give that person a relevant message that tells them something about the application, something that's relevant to them, so that they get hooked and they want to come back.
Similarly, when you start to have a whole slew of users, it's important to start categorizing those users into different groups. You might have loyal users, you might have dormant users; you might have users that are at risk of attrition; ticketing app might have VIP ticket holders or discount ticket buyers... You essentially start grouping these users into meaningful categories and then there's different business objectives for each user. You have to define different user journeys for each of them, so that you can push those users towards that objective.
Someone who is really loyal but has never bought something - I might group them as an under-monetized user. Then you can institute a push in-app MMS campaign to drive them towards purchasing something. Similarly, someone who's a dormant user, I might send him a push notification to remind him why they downloaded the app in the first place.
That's the stuff we do with Pyze - we'll automatically segment the user base and give you the ability to start interacting with each group of users in a meaningful way.
Paul Kemp: I so love what you said there about under-monetized users, and I'm sure that many people listening to this now have many under-monetized users, including me. I launched an app that almost reached a million users, and I think every single one of them was under-monetized.
We're all very busy... You mentioned earlier that there's a lot that the big boys do that we can't do ourselves, because we're smaller teams, but I wondered what we could do in terms of automation. What's the role of automation and how can it help us with getting traction, and all these other things?
Prabhjot Singh: I live by the rule of "Automate Everything", whatever you can, and that's important. Let's take your app, for example - you had an app with a million users. The rest of us, when we have access to basic analytics tools like Google Analytics or Flurry, or any of these types of solutions, we have to do a lot of manual analysis ourselves to create segments of users, and then analyze those segments for things that we care about - engagement, monetization, or whatever.
Now, when you've got a million users, how do you even come up with the right segments and groupings of users to analyze? It's impossible, and that's all you'd be doing, if you in earnest started focusing on that. So it's important to have a solution that can automate segmentation, so you can easily understand what makes your users different. You have some loyal users and you have some dormant users, right? What's the difference between those users? What makes a dormant user dormant, what makes a loyal user loyal? Because I want to go out and get more loyal users, obviously.
In a game you might have a group of people who are spending a lot of money and people who are not spending any money. Kind of like Facebook has these look-alike audiences, we can then start to engage users that fit a particular mold. That's automating analysis. Now you've got a grouping of users that meet different criteria that we understand, and we can see what aspects of the app each type of user is using. Maybe I need to make some changes in my application to move those users along.
Paul Kemp: Actually, it's fascinating you're talking about segmenting each group, because I can imagine that you do get lots of different users who can fit into many different categories, some of which you already mentioned. But are there any other ways to automate, other than maybe putting the users into different categories? Any other suggestions about automation?
Prabhjot Singh: Yes, absolutely. So once you've put users in different categories, you can then start to automate engagement with those users. There are going to be different engagement paths for someone who's a loyal user, or a seasoned ticket holder, or a discount ticket buyer for, let's say a ticketing app. Once you have that understanding of "Okay, these are the different groups of users", you can then start to automate... Almost think of it like a drip campaign for mobile, where you're sending out push messages or in-app messages to drive users to a particular objective, whether it's a purchase for under-monetized users, or getting them re-engaged for users at risk of attrition.
That's also another type of automation that can be done, and it's important to do that, because to do analysis against users, especially if you've got a million, two million users on a monthly basis, it's almost impossible to figure out what push messaging you should send to who, and when you should send it.
Paul Kemp: That seems like a good chance, Prabhjot, to talk about Pyze. You are a growth platform, you help developers grow, get traction, segment... But maybe we could go into more analysis about what it is you do for developers and how people can use your platform.
Prabhjot Singh: Sure. As I mentioned, we essentially automate segmentation - the analysis piece - and then we enable automation of engagement, and then personalization. Those are the three key things that we do.
We spoke about the analysis earlier, in terms of being able to identify what are the trends and patterns at a macro level, across the entire user base - what aspects of the app are being used and not being used. Once you have that understanding and you can start to spot, "Okay, these are the trends" and "Oh, here's an anomaly. What does this anomaly mean?" Once you have that understanding, you can then have the segmentation of users and then you can start to engage with those users in a meaningful way.
We do a lot of the heavy lifting, like if you've got a million users, we know the best time to reach every single user of the application, so that they'll be most receptive. If you and I are using an app, you're using it at two in the morning, I use it at two in the afternoon, and I wanna send a push notification, it will automatically send it at the right time to both of us.
Those types of intelligence capabilities are built in. Then you can set up campaigns to trigger on events; if you've got an eCommerce app and someone's adding an item to a cart and then they abandon that cart, maybe we should send him a coupon for that item, so that we get them re-engaged and convert that user.
You can set up those types of automated actions with Pyze. You can do analysis to say, "Okay, show me all the users that have high engagement in my application, but they haven't spent a dime with me." Now I can take that group of users and I can say, "Okay, let me reach out to those users right now." We enable that analysis in real time, across millions of users if you've got them.
And finally, personalization. You can say, "Okay, I'm going to start personalizing the experience of these different user groups. Maybe my loyal users will see a different interface than my not-so-loyal users", or "My seasoned ticket holders will see a different UI, so they can pick their VIP seats, versus people who buy discount tickets."
You can easily start to morph the application to be very centered around each individual, so that it becomes a personalized experience for everyone.
Paul Kemp: I love it. It's a shame this didn't exist years ago, because it just seems like what we were talking about at the start - the big companies have the big resources, and they have these departments almost that are doing this stuff that you're now enabling the smaller companies, the indie app developers, and everyone really, who can now use your platform to do all this big data analysis and automate it, I'm guessing at a fraction of the budget that it would cost. That's really fascinating.
Prabhjot Singh: That's right. Necessity, as they say, is the mother of invention in many regards. For myself and the co-founders, this was a pain point that we had, and we thought it didn't make sense to solve it for one application because it wasn't cost-effective; we could build a platform. Pyze is free to get started with, and our lowest tier actually is free for up to a million monthly active users. We want to help everyone be successful, and if you're successful, we know we'll be successful as a business.
Paul Kemp: That is fantastic. So how can people go and sign up to Pyze? Where's the best place to go?
Prabhjot Singh: Pyze.com - P-Y-Z-E. You can just click on the Get Pyze button and get going.
Paul Kemp: And for all those English out there, in my native tongue it's P-Y-Z(zed)-E. [laughs]
Prabhjot Singh: That's right.
Paul Kemp: Pyze.com. Prabhjot, it would be wonderful to know how to reach out to you personally as well, because you're so inspirational as a founder, and I'm sure that you've inspired a lot of people. How can people reach out and connect with you?
Prabhjot Singh: Absolutely. You can reach out to me on Twitter @psinghSF, or you can shoot me an e-mail at ps@pyze.com.
Paul Kemp: Wonderful. Thank you so much for coming on The App Guy Podcast and sharing hugely valuable content. We look forward to using the platform and seeing all the success stories from people who have come out and it's changed their business. Thank you very much!
Prabhjot Singh: Absolutely. The pleasure is all mine, and yes, certainly an open invitation to all your listeners - feel free to reach out to me. I often give advice on app concepts. Also, product market fit discussions is something I can weigh in on. Always happy to help in any way I can.
