← Sub Club by RevenueCat
Sub Club by RevenueCat
How Mojo Increased ARPU 60% In Just Five Months – Michal Parizek, Mojo
00:21:06
Loading summary
Transcript39 lines
- [00:01]
A
Welcome to the Subclub Podcast, a show dedicated to the best practices for building and growing app businesses. We sit down with the entrepreneurs, investors and builders behind the most successful apps in the world to learn from their successes and failures. Sub Club is brought to you by RevenueCat. Thousands of the world's best apps trust RevenueCat to power in app purchases, manage customers, and grow revenue across iOS and Android and the web. You can learn more@revenuecat.com let's get into the show. Hello, I'm your host, David Barnard. Today's conversation is shorter than usual and will be featured in revenuecat. State of Subscription Apps Report. Each episode in this series will explore one crucial topic and share actionable insights from top subscription app operators. With me today, Michael Parisek, Senior Growth Product Manager at Mojo. I'm on the podcast. I talk with Michael about the experiments behind MoJo's 60% lift in average revenue per user, why a winning paywall in Japan completely failed in the US and why not relying on day one for most of your revenue is actually a strength. Hey, Michael, thanks so much for joining me on the podcast today.
- [01:22]
B
Hey, thanks for having me.
- [01:24]
A
You've read a blog post on the RevenueCap blog a while back about how Mojo increased average revenue per user 60% in five months. And I've been wanting to have you on since reading that blog post because there are so many little things in there I think people could take away from and not everybody's going to read the blog post. And it's kind of fun to kind of dig deeper into the things behind what ended up in the blog post. So let's start there. What are the things that led to that 60% increase in average revenue per user?
- [01:52]
B
There was actually like a bunch of experiments we did on mainly payroll and pricing, and there were particularly free experiments which kind of stood out and brought that pretty, pretty good lift in arpu. One of them was sort of like a yearly plan as a default. So initially, like on a paywall, we've showed the yearly and monthly plan next to each other on like the very first screen, very first payroll screen. And then we've tried putting the monthly plan under sort of like a view all plans button. So it wasn't really visible on a first glance. And that really helped drive our yearly plan adoption a lot. Like I think like a 15 or 20% percentage point. So that helped a lot, actually increasing the new revenue, obviously.
- [02:41]
A
Anything else that you, that was part of that, that you think really drove that success, or you think you just like, I mean, a lot of people experiment with this. But for some people it does seem to work where it's like the, you know, the yearly plan is listed with a monthly amount and then the monthly plan is listed at a much higher amount. Did you, did y' all test that at all of like having the monthly plan maybe like way higher to where it makes the yearly better or. Yeah. What went in that?
- [03:05]
B
Yeah, yeah, we experiment with that as well. So what we've tried and succeeded was actually sort of like a monthly plan anchoring. So we essentially added like a small line next to our sort of yearly plan which says like that price is equivalent of earning like $10 a month. And that actually worked very well in comparison into that sort of multiplan which was obviously higher. The price was like 25, let's say dollars. So that kind of showed a yearly plan is such a. Such a good deal comparing in the costs. So that actually did work very well and kind of interesting wells that actually work very well, particularly in Latin America in Brazil and Mexico.
- [03:45]
A
That's super rich. Why do you think it performed better in Latin American countries? In the U.S. yeah, because in the
- [03:51]
B
U.S. it worked as well, but it was, the increase was kind of, let's say a little bit less. It feels like 10% lift in your revenue in Latin America was way higher. It was actually about something like definitely double digit was like 30 to 40 percentage. And I think actually it worked better there because that's sort of just my hypothesis. But I think because the sort of like purchasing power of those markets is lower than in the US typically. So people tend to care more about like what's the price there and they tend to basically care more about the costs of the dapps and the subscriptions they have. And if they see that they are just doesn't cost much when they actually look at it from let's say monthly perspective, like what they pay by month, it just kind of persuaded more and triggered more the conversion behavior there. So yeah, we were actually super happy to have there and added their pay and online payroll.
- [04:43]
A
Had you already experimented with lower prices in those countries or was it a similar equivalent price to U.S. prices?
- [04:51]
B
No, we did actually. It was on just sort of another kind of key experiment which drove the 60% increase in ARPU that we did a lot of price testing as well. And some of those actually include also Latin America countries like typically Brazil and Mexico which were kind of the top two GEOs on that region. And so we've actually lowered the price so we didn't have the equivalent of the US prices. That's something what I think usually App Store sort of suggests they basically just do the exchange rate. Right. And then calculate say no, no we didn't. Well we originally had that I think in the past, but then we also tested a bit lower prices and it turned out it was actually better.
- [05:31]
A
What was the key metric you were tracking during this? Like in setting up all these experiments, you're trusting price, you're testing, you know, paywall layout, you're testing placements, you know, do you have the monthly and annual, you're, you're testing all these things. What was the unifying metric you were looking at to determine the success of those experiments?
- [05:52]
B
Yeah, so our sort of like umbrella metric for all those summarization experiments was the average revenue per user in the first seven days. So this kind of ARPU ARPUs 7D 7day. And yeah, we've intentionally used it because we. Well first of all we saw more sort of potential in optimizing the new revenue rather than renewals and just because that we want to shorten at the payback period. We wanted to optimize the new revenue because of also sort of supporting the user acquisition loop to allow higher spend, etc. So we want to drive new revenue. As we saw that we can actually scaled it more and we can actually compound it with the user acquisition and get like more revenue in total. And yeah, so that's why we sort of chose the early average revenue per user and specifically seven day, because at that time we had three day trial. So we wanted to have the window to be long enough to cover, you know, those three, four days. And I think it was just seven day. I think mainly because actually revenuecat showed like one of these arpu, like pre default is actually, that's actually maybe the first one is actually what was available is 7day. So we choose like ARPU 7day, I think because of that. So and it actually. So they work pretty well. Yeah. So basically we optimized for the new revenue and as you probably know and everyone else in the app business, like lots of new revenue is coming from the very early days, like the very first few days. So it was a good metric for just tracking new revenue.
- [07:28]
A
Did you look back on some of those experiments and see the impact on retention? So you know, did you knowingly sacrifice some long term revenue for that quick return on ad spend?
- [07:39]
B
Typically where we look at retention or at least some like a proxy to retention, like we use typically 7 day cancellation rate as a sort of proxy for retention rate or for renewal rate, what it could look like. And we typically look at this sort of proxy in when we did like price testing because I, I've seen data that when you test in different prices, particularly higher prices, usually you see sort of higher cancellation rates and lower renewal rates with high prices. That's kind of quite reasonable. So and I remember a couple of tests where we actually like tested a different price and the price actually mostly like a higher price actually turned out to be the winner on the new revenue. But when we actually modeled having the new price for a year long, sort of calculating a bit long long term revenue and we saw that we actually sacrifice in the long term mainly because the renewal rates just dropped because of the proxy 7 day cancellation rate just was way higher than for the baseline price. So then we decided not to do that and kept the original price and sacrifice a bit of yeah, less new revenue but sort of having more new revenue in the long term.
- [08:50]
A
One of the things I talked about in another one of these state of subscription apps podcasts was how a lot of times experiment results don't stack. So you get a 10% win here and a 20% win there and a 15% win here. And then then you look at it at the end and you actually haven't raised average revenue per user by the sum total of all those experiments. You're kind of like getting 10% here but losing 5% there and getting 20% here but losing 10% there. What do you feel like was the key to actually stacking those successes? Were you looking back as you tested each of those to see if it had a negative impact on some of your previous tests to be able generate such a large increase in average revenue
- [09:36]
B
per user essentially like executed a bunch of pay one price tests and whenever we've seen there is sort of like positive results, whenever we've seen there's a statistical significant winner, usually we've sort of retested also in other markets. So usually let's say we've tested in the US or in Europe and if it works there then we've retested in other key markets to make sure that actually it will work there. And it happened a couple of times that something we've fully rolled out in some of the markets and not in let's say all over the world just because we haven't really seen a positive impact sometimes also on Agaji. So we make sure that particular change is actually bringing additional incremental revenue, you know, in all those key segments according to Geo typically. And if we haven't seen that. Then we didn't roll it out. And then sort of on a high level we've monitored the RP7 day typically actually in the revenue get charged to essentially make sure that for the new cohorts users we actually seen that lift in the arp. Of course it's sometimes sort of hard to isolate it from UA changes from other externalities but we try to our best to see if actually after rolling out we're actually seeing that. And yeah, and yeah, luckily we were. We actually saw that we implemented changes and we've seen that it's actually going up the erpu and we were sort of happy to, happy to see that and kept that process keep going.
- [11:08]
A
Were there any surprising results like something that worked incredibly well in Brazil but was terrible in the US or something amazing in Europe that failed in Latin America?
- [11:18]
B
Maybe one surprising thing was something which we tested particularly in Asia, it was actually in Japan. It was like a long scrolling paywall with a lot of information. Like there was lots of social proof or reviews. There was like a comparison between the free tier and the pro tier. And that design actually worked incredibly well in Japan driving I think 20% lift in new revenue. But actually the same design kind of failed in the US actually in the US more sort of, kind of easy to read or more kind of cleaner design with like a slider and the videos in the background and just very punchy messages. Work much better actually than the sort of very in depth and very descriptive design for Asia.
- [12:03]
A
Yeah, that's fascinating. I did want to talk now about placement. So you know, of course, course as with most subscription apps these days, Mojo does show a paywall on onboarding. But I was surprised to see the stats that the only 50% and I say only 50% because in the Revenuecat data we see more like 80% of payers across, you know, the entire subscription industry happen in that first day. So what else have you done to drive those conversions after the. That initial onboarding?
- [12:35]
B
Yeah, that was actually, that was actually the number one thing which surprised me the most when I actually joined Mojo. It was my first sort of mobile, first kind of business job. I did that actually. There's so much actual revenue coming from the very first day. But then I learned actually is very normal and it's even like lower than we just said. Yes, most apps do like 80% and I think actually that the number like 50% is actually a good signal that if the app actually does that, I think it's a very good signal for the app that it can actually drive a good revenue also from existing users. It's likely have a good retention and essentially a lot of existing users. And it's kind of sounds to me like a really very healthy sort of behavior and signal. There's, I think, a couple of things, I think why MoJo has 50% one is that the free tier is, you know, is actually, I think it's pretty good. Also comparing also some other competitors actually in the free tier, you can actually can do quite a lot. So it's not that restricted like maybe some other apps. There's lots of features which is actually available in the free tier, lots of content. So this is one thing. So like the, let's say genericity of the free tier and just intentional. The other aspect is that we've run the paywall campaigns for existing users, which I think not a lot of, maybe not most of the apps actually do. And I think it's actually one of the kind of underrated things which I think most of the apps should do. So essentially it's running a paywall campaign, so actually triggering a paywall and certain behavior for existing users, either as an app open, like we did in Mojo, or after some key behavior event like when you do something key in your app, like, I don't know, sharing something or whatever the key kind of an engagement event is. So that paywall campaign actually drove, I think about 15% of new revenue from the existing user base, which is. Yeah, it's just pretty a lot. And it's super simple to set up and it wasn't really kind of having any negative reviews on that from the users. So it's just kind of a no brainer to have that.
- [14:42]
A
Yeah, that's surprising. So free users, they open the app and immediately see a paywall.
- [14:47]
B
Yeah, they hit a paywall. Yeah. And actually it worked. Yeah. And quite a lot of users actually drove a lot of revenue.
- [14:55]
A
Did you check Churn and other things like, I mean, I guess that's the thing is like you do always kind of trade off. So maybe there was a little bit of Churn, but then the additional revenue kind of made up for that. But any other things you track, like you said, there were no negative reviews, which that's shocking to me. People seem to go out of their way to like you complain about things like that in the App Store reviews, but support retention, like any downsides to this paywall on app open?
- [15:22]
B
It was on the app open, but it was, I think the frequency is set to one payroll per week. So you don't actually, even if you open the app. Yeah, every day, in a couple of days, a couple of multiple times a day. You just essentially have that paywall only being triggered like once a week. Essentially. The frequency was pretty low. So. Which also, I think played some role why users weren't complaining. But I specifically actually asked a colleague from support and he really mentioned that. That user complaining. I did once, like, a quite thorough analysis of all the reviews and tried to find, like, what the users, you know, complain about. Not really from that particular reason, but I was more interested, like, what the user liked, what I didn't like. And I actually haven't really seen any specific or negative about having a paywall a lot of times displayed, etc. So, yeah, so actually it was a good thing for us.
- [16:10]
A
It's surprising sometimes when people are getting something for free, they tolerate more than we assume. Sometimes, you know, like, I know a lot of games and even some regular apps have full screen interstitial ads. I mean, Duolingo is actually a great example of this. You know, they do it very tastefully, but they have a full screen, unskippable takeover ad after a lesson. So I think people do kind of intuitively get the exchange of value there of like, oh, I'm using this completely for free. And I think you can get away with it more if your freemium tier is very generous, like you were saying. And maybe that's. That's kind of the key. It's like if your freemium tier sucks, one, you're probably not going to get much retention anyway. So not many people are going to see the app open paywalls. And then two, like, they feel like they're, you know, pulling one over on you or, you know, getting something getting, you know, they feel like they're getting a lot of value. And so when they see that, that paywall, it's probably less offensive in those kind of situations.
- [17:09]
B
So.
- [17:09]
A
But fascinating how well that worked.
- [17:12]
B
100%. Yeah. I really encourage everyone to essentially just try to just say, yeah, test it A, B, test it, just measure it. And I think, yeah, most of all, you will be surprised that users will not react the way how you kind of offer fear they react. So it's. I definitely would recommend just trying things out.
- [17:29]
A
The last thing I wanted to touch on was experiment velocity. Y' all did a lot of tests. How do you keep up with that? How do you isolate variables? How do you think about testing velocity?
- [17:43]
B
Oh, yeah, there was something super important. I actually think that it's the kind of number one Growth asset as everyone who actually wants to optimize paywalls and pricing. If you have good research, good prioritization, if the velocity is I think even kind of, it's even maybe more important because there's more shots you take, there's essentially higher chance you win, there's a shorter sort of feedback loop, better learning cycle. So it's, it was a key. And yeah, I did a lot of experiments and I think maybe the number one thing which allowed me to do that was having kind of a third party payable platform allowed me to being autonomous and iterate fast and really kind of shorten the cycle of actually coming up with experiment, developing it, launching it from months or weeks to essentially days. So it was so kind of important that I think without it I would definitely wouldn't be able to get that amount of experiments live in. Yeah. In that short time.
- [18:43]
A
That makes a ton of sense. And you know, Mojo is a pretty big app and you were doing a ton of user acquisitions. So you did have a lot of users coming in to be able to do the tests on. But did you have a particular testing velocity? Like would you be launching it on a weekly cadence or even like three days and then wait for the data to make a final decision? But while you're already kicked off another
- [19:03]
B
test, it was more weekly or bi weekly cadence. But what is also important to say that we mostly run experiments. We typically kind of broke down the audience, the new users to sort of free buckets. According to Geo, it was like a US or a kind of US Australia, Canada kind of English US English bucket. Then there was the Europeans and then there was Latin America. And we typically had like these streams of tests for each this specific kind of GEO bucket. And we've run tests in all of them, three in parallel. And each test kind of lasted usually one or two weeks because in those three geo, those are kind of these three key GEO segments we had and we were able to get the statistical significant results usually in those one or two weeks for each of those buckets. So we were able to conduct experiments there.
- [19:58]
A
I think a great way to summarize this whole episode is you should be testing a lot. If you're not, you're leaving revenue the table. Again, it's just such a clear example that you were able to increase average revenue per user by 60% in five months. That is such a huge win for the company, ability to acquire new users and then you know, over the long haul building that subscriber base over time. So such a fascinating lesson and thanks for sharing your insights today.
- [20:25]
B
Thanks. Thanks again for having me. David. It was really a pleasure to talk with you.
- [20:29]
A
Anything else you wanted to share as we wrapped up?
- [20:31]
B
I encourage folks if you want to learn more about your payable testing and monetization, just follow me on LinkedIn to share some of the advices and experience there. So hopefully you can learn something more.
- [20:42]
A
Awesome. We'll put a link to that in the show notes. Thank you so much Michael for joining me today. It was really fun.
- [20:46]
B
Thanks a lot. David.
- [20:50]
A
Thanks so much for listening. If you have a minute, please leave a review in your favorite podcast player. You can also stop by chat.subclub.com to join our private community.