Hound Reminders lets you send reminders to other people via SMS.
I initially launched Hound as a free app — free to download and free to use. I knew this would be a problem for me because my costs scale with usage. Twilio charges $0.0075 per sms sent, more if international. And even more if there is media involved.
Still, I decided to launch for free. Mainly I just felt too lazy to implement payments. Since this was a side project, spending 50 hours to get to initial release was a big time investment. Spending another 50 figuring out Apple Pay or Google Pay or whatever just felt too daunting. I thought to myself, “let’s see if anyone uses it”.
Initially, nobody did. I had a few friends test it out and there didn’t seem to be any bugs. That was nice. And I got about one new user per week all of whom churned. OK, I thought. No big deal, I’m still using it so I’ll just keep working on things I want for myself.
Around 2 months later I got my first power user. She was a nutritionist sending reminders to her clients to take their weight, make their milkshakes, etc. That was pretty cool! Problem being that it was costing me $15/mo in sms fees. I mean, I’m glad she found it so useful, but dang that’s a lot of text messages. Alright, so it was time to start charging.
I was flustered at how quickly she drained my Twilio account, so I reached for the first solution I could think of. I would just try to pass on the Twilio cost to the user with a small markup on messages sent. With a Twilio cost of $0.0075 I would start by charging $0.01 per message. Then I remembered a classic piece of startup advice: charge more for your product. Alright, let’s do $0.02 then.
I built out a credits system: on signup a user receives 50 credits. Each message sent consumes a credit and you can top up 250 credits for $5. I granted my one power user an initial 50 credits, implemented an “almost out of credits” warning email when they dropped below 10, and waited to see if she would top up. She didn’t. Actually she didn’t even open the app ever again.
I checked my SendGrid account to make sure I was sending out the warning email, and also reached out personally to see what was up. No response.
Alright, so I had to think about why this might be. She took the time to set up 6 recurring hounds in the app but wouldn’t even engage me on pricing. Her recipients had been using my “mark this hound complete” feature, so clearly the app was supplying some value to her client relationship. So why?
My first thought was that $0.02 per message was just too expensive. This nutritionist is not a big business, she is more akin to a consumer. If she is sending enough messages to cost her $30 per month that puts Hound on a cost level with Netflix, which did not sound right to me. It just doesn’t seem worth it, even to me the creator of the app. So I went back and changed the price to $0.01. I left it alone and went back to forgetting about the app.
A few months later I had several power users, all following the same pattern. Every one of them would use up their 50 credits within a month and never top up. Further, not a single one of them re-opened the app after their initial session. They would schedule 3 or 4 hounds, let them run until the credits dried up and then just ghost.
I like to fancy myself a “startup” person, so I know that “retention = growth”. So yeah, having 0% app-opening retention is pretty bad. But rather than get discouraged I decided to ask myself “are my users getting value”? Even though these users are all “churning” I could not shake the feeling that they were getting value out of the app. I took another look at my own usage and realized that I, too, had not opened the app in a while. Was I not getting value out of the product?
I have three persistent recurring hounds. One to remind my wife to perform a daily task that she often forgets. Another weekly reminder to my friends to update our gaming systems before game night. And a third monthly task to remind my wife and I to give medication to our dog. Other than those three, I had a few one-off hounds here and there. But the real value to me was having those three hounds on repeat. I didn’t need to open the app anymore.
Further, I got push notifications every time one of my hounds was delivered and then again if someone marked it as completed. My users were having the same experience. If they were not getting value out of the app then surely they would delete the hounds to stop notifications? And further, if the hounds were bothering recipients wouldn’t they delete them? All of this made me believe my users were getting value, but they followed an unusual pattern.
I decided to build my monetization strategy around this theory: that the users who are willing to pay for my service — the ones who need my service — only need to open the app every few months. They likely have a few things they want to get off their chest and into a reminder system, and then they are done until a few months later when things pile up.
With this in mind, I knew I had to respect two things about the implementation of payments in Hound:
(1) I need to get users to pay me the first time they open the app. If my power users are in the set-and-forget crowd, I need payment before they forget.
(2) Payments need to be automatic. Whether it’s a recharging credits system or a subscription or w/e I cannot require confirmation from the user on every payment. Again this is because they apparently don’t want to re-engage with the app for a long while.
I also decided I wanted a free tier. Most users who schedule an initial hound do so as a test. They schedule one to themselves that’s basically “From Me to Me: testing this out!”. Something like that. I didn’t want to force a payment decision before they could send that kind of test.
I took a look at power users (all 6 of them) and saw that in general they had 3–5 recurring hounds active. I decided I would try charging based on “simultaneously scheduled hounds”. So for free you can have one hound scheduled. Once that hound was sent you could then schedule another one, but you can only have a maximum of one scheduled at any given time. Then for a subscription fee you can unlock 5 simultaneous hounds.
I felt this method solved all of my problems:
(1) Users could test the app for free
(2) Power users who then decided to create multiple hounds had to subscribe
(3) There was a natural limit on how much a free user (like that nutritionist) could cost me
(4) Once subscribed, users did not have to re-engage with the app to keep their hounds active
Implementing the subscription was sort of a nightmare. I’m not a native app developer, so I was using the Expo Managed Workflow. In-app payments forced me to eject into a bare React Native app which was a tad precarious. Then dealing with the App Store payments system, oh dear lord. But anyway it all came together and I released my beautiful well-thought out payments system.
And that is the current state of affairs of pricing for my little consumer app. I’ll update this page or release a separate article once I’ve seen it in the wild.