User Power
Value/Post Ratio
658%
- Jan 12, 2020
- 19
- 125
Hi all, I'm am occassional browser, rare poster here. Earlier this year I posted about a music networking app I was building called Artyst. Well it finally launched just last week, so I thought I'd check in and share what I learned.
Hopefully this will be useful to someone thinking about launching an app. Maybe you can avoid some of the mistakes I made. Happy to go into more detail if anyone's curious!
--------
This post is the story of Artyst, my just-launched music networking app. I started working on this app over a year ago, in June 2019. The app finally launched on July 1st, 2020. I’ve spent a ton and learned a ton over that timeline. But I also made a ton of mistakes.
I’m confident I could get this project launched for a fraction of the cost in a fraction of the time if I was starting again from day 1, knowing what I know now. Here are a few lessons I learned.
Start lean… and stay lean
Like many tech founders, I’ve read The Lean Startup and made an effort to apply lean thinking to my work on Artyst. I spent the first few months sketching UIs, paying for app screen mockups on Fiverr, creating personas, launching a small landing page. I got feedback on these from musicians I know as well as by running Facebook + Instagram ads. This provided sufficient validation that the idea was worth pursuing. I believe this process was worthwhile, though I went a little too far worrying about details and revisions — it shouldn’t have taken more than a few weeks.
Unfortunately, staying lean was much harder. I’d defined a truly minimal version of the app to build. But after a few iterations with the developers, the final specs came out as an extremely bloated version of the MVP. I went from “create and search profiles” to “profiles, chat, news feed, featured section, and more with full iOS + Android support from day one”. A huge undertaking that dragged on several months longer than planned, resulting in several disappointing delays for my users. If I had stuck with the minimal version, I would’ve launched much earlier for less money and ultimately created a better relationship with my users.
Lesson: Start lean. But be careful also to stay lean. Watch out for scope creep. Keep your “minimum viable product” to the true minimum.
Build it yourself
Issues with paying for development
Working with developers has been the biggest cost in launching Artyst — in terms of both time, money, and stress.
I’m a career data scientist. I know how to code. But I’ve never done app development, so I went into this project with the mindset of “leave it to the pros” and decided to pay for a developer.
Initially, I went with a cheap developer from Fiverr for a few grand. He did some decent work, but after several crap deliveries and deadline overruns, I decided to move on. I’d wasted a lot of time, money, and stress.
Next, I decided to go with a “real” app development group. After getting several quotes in the $100k range, I finally settled on a “modestly priced” group for $25k. Yikes. But worth it, I thought, because these guys are gonna hit the deadlines and deliver a super professional app. That’s a ton of money, and I spent a lot of time talking to them before making the decision. How could they not?
… wrong yet again. Development went months beyond the expected launch date. Specs were ignored. I had to fight to get clear answers. A lot more time, money, and stress.
Maybe you are better at vetting developers than me. I don’t doubt you. Just be careful. Don’t assume price = quality. And don’t take anyone’s word for anything. Understand your leverage in each situation, understand the other party’s incentives, and always have a backup plan.
Benefits of building yourself
The alternative? What should I have done instead? Build it myself. If you know how to code, great — do that. But if not, you now have a ton of no-code options available. Sites like Bubble and Adalo that let you build web and mobile apps in a drag-and-drop manner.
I really regret not using tools like these from day one. I had concerns like, but then won’t my app be slow and clunky? Won’t the design be sub-optimal? Maybe a bit. But don’t assume it won’t be just because you pay for a developer. Won’t it limit what my app can do? Maybe… but increasingly, not at all. And even if my desired app function doesn’t exist with no-code, there’s likely a good enough substitute to make it worth using no-code for an MVP — you can always transition to code later after a successful V1.
Put simply, the benefits of building yourself (with or without code) are cost and control.
If you’ve got a stack of cash you’re willing to throw toward a moonshot app development, go for it. But most people don’t. I barely made it work. Coding yourself is typically free until you need hosting — and maybe still free then until you scale up. If you’re going the no-code route, most of these sites offer a free plan or at least a generous trial period. After that, you’ll pay per month, a couple hundred dollars on the higher tiers. Either way, much cheaper than paying for a full development.
And then control comes into play. If you’re relying on developers, you’re not in charge of the timeline. You can set deadlines, sure, but what are your options if they’re not met? And most importantly —the part I regret the most — what do you do after the initial build? If you want a bug fixed, a new feature added, a tiny UI tweak, do you want to have to pay thousands more and wait another month or three on your developer? Or can you tweak the app yourself same-day or, worst-case, over a long weekend?
Lesson: Built it yourself quickly and cheaply.
Get real users, provide real value, get real feedback
A consequence of all this feet-dragging on the Artyst release timeline is that I’ve had thousands of aspiring users waiting on this app for months. That’s a mistake. I should have given them something real to engage with from day one (or, at they very least, month one).
Acquiring users
Next to development costs, user acquisition has been my second largest expense. I’ve spent several thousand dollars running Facebook and Instagram ads to get page follows and email signups on the Artyst website. I can’t say I regret this — in the case of Artyst, social media ads have actually worked really well for gaining exposure. But this is by no means the only way to do it.
Most people won’t have the budget to drop this kind of money on ads, which is understandable. And even for those who do, it doesn’t always work well. This depends on where your target users hang out and how easy they are to nail down using Facebook’s ad audience parameters.
Looking back, I would’ve given more attention to non-paid social media marketing — the kind where you engage with users directly. For example, I would’ve followed Instagram and Twitter pages with high engagement among musicians (eg, XXL, Murda Beatz) and engaged with the commentors in their posts. Second, I should’ve made connections in online forums — there are several really good subreddits for different music niches that could’ve given me really valuable feedback and been a great source of early users. You can likely apply this strategy to your own niche.
Valuable relationships with users
While I built a decent audience for Artyst, my main regret is not delivering value much much sooner. I should’ve given them a small V1 to work with. I could’ve even created a forum for early users to start networking and building a community on until the first version of the app was built. It was a mistake to keep them waiting all that time.
Not only would this have shown more respect by not wasting their time, it would’ve also given me valuable feedback that I could’ve used to iterate much more quickly. If there was a UX flaw in my idea of how Artyst search should work, I could’ve learned that in the first month or two, instead of waiting a year later.
I had real users, but I didn’t give them a real product to use, so I had no feedback to use for directing the app’s roadmap.
Lesson: Deliver a real product ASAP. Get real users to give you feedback and iterate quickly.
Conclusion
In a nutshell: if I was re-starting Artyst today, I would do the following within the next month:
This is a 50,000 foot overview of what I’ve learned launching Artyst over the past year. I hope it gave you some useful thoughts to consider as you dive in with your app.
Hopefully this will be useful to someone thinking about launching an app. Maybe you can avoid some of the mistakes I made. Happy to go into more detail if anyone's curious!
--------
This post is the story of Artyst, my just-launched music networking app. I started working on this app over a year ago, in June 2019. The app finally launched on July 1st, 2020. I’ve spent a ton and learned a ton over that timeline. But I also made a ton of mistakes.
I’m confident I could get this project launched for a fraction of the cost in a fraction of the time if I was starting again from day 1, knowing what I know now. Here are a few lessons I learned.
Start lean… and stay lean
Like many tech founders, I’ve read The Lean Startup and made an effort to apply lean thinking to my work on Artyst. I spent the first few months sketching UIs, paying for app screen mockups on Fiverr, creating personas, launching a small landing page. I got feedback on these from musicians I know as well as by running Facebook + Instagram ads. This provided sufficient validation that the idea was worth pursuing. I believe this process was worthwhile, though I went a little too far worrying about details and revisions — it shouldn’t have taken more than a few weeks.
Unfortunately, staying lean was much harder. I’d defined a truly minimal version of the app to build. But after a few iterations with the developers, the final specs came out as an extremely bloated version of the MVP. I went from “create and search profiles” to “profiles, chat, news feed, featured section, and more with full iOS + Android support from day one”. A huge undertaking that dragged on several months longer than planned, resulting in several disappointing delays for my users. If I had stuck with the minimal version, I would’ve launched much earlier for less money and ultimately created a better relationship with my users.
Lesson: Start lean. But be careful also to stay lean. Watch out for scope creep. Keep your “minimum viable product” to the true minimum.
Build it yourself
Issues with paying for development
Working with developers has been the biggest cost in launching Artyst — in terms of both time, money, and stress.
I’m a career data scientist. I know how to code. But I’ve never done app development, so I went into this project with the mindset of “leave it to the pros” and decided to pay for a developer.
Initially, I went with a cheap developer from Fiverr for a few grand. He did some decent work, but after several crap deliveries and deadline overruns, I decided to move on. I’d wasted a lot of time, money, and stress.
Next, I decided to go with a “real” app development group. After getting several quotes in the $100k range, I finally settled on a “modestly priced” group for $25k. Yikes. But worth it, I thought, because these guys are gonna hit the deadlines and deliver a super professional app. That’s a ton of money, and I spent a lot of time talking to them before making the decision. How could they not?
… wrong yet again. Development went months beyond the expected launch date. Specs were ignored. I had to fight to get clear answers. A lot more time, money, and stress.
Maybe you are better at vetting developers than me. I don’t doubt you. Just be careful. Don’t assume price = quality. And don’t take anyone’s word for anything. Understand your leverage in each situation, understand the other party’s incentives, and always have a backup plan.
Benefits of building yourself
The alternative? What should I have done instead? Build it myself. If you know how to code, great — do that. But if not, you now have a ton of no-code options available. Sites like Bubble and Adalo that let you build web and mobile apps in a drag-and-drop manner.
I really regret not using tools like these from day one. I had concerns like, but then won’t my app be slow and clunky? Won’t the design be sub-optimal? Maybe a bit. But don’t assume it won’t be just because you pay for a developer. Won’t it limit what my app can do? Maybe… but increasingly, not at all. And even if my desired app function doesn’t exist with no-code, there’s likely a good enough substitute to make it worth using no-code for an MVP — you can always transition to code later after a successful V1.
Put simply, the benefits of building yourself (with or without code) are cost and control.
If you’ve got a stack of cash you’re willing to throw toward a moonshot app development, go for it. But most people don’t. I barely made it work. Coding yourself is typically free until you need hosting — and maybe still free then until you scale up. If you’re going the no-code route, most of these sites offer a free plan or at least a generous trial period. After that, you’ll pay per month, a couple hundred dollars on the higher tiers. Either way, much cheaper than paying for a full development.
And then control comes into play. If you’re relying on developers, you’re not in charge of the timeline. You can set deadlines, sure, but what are your options if they’re not met? And most importantly —the part I regret the most — what do you do after the initial build? If you want a bug fixed, a new feature added, a tiny UI tweak, do you want to have to pay thousands more and wait another month or three on your developer? Or can you tweak the app yourself same-day or, worst-case, over a long weekend?
Lesson: Built it yourself quickly and cheaply.
Get real users, provide real value, get real feedback
A consequence of all this feet-dragging on the Artyst release timeline is that I’ve had thousands of aspiring users waiting on this app for months. That’s a mistake. I should have given them something real to engage with from day one (or, at they very least, month one).
Acquiring users
Next to development costs, user acquisition has been my second largest expense. I’ve spent several thousand dollars running Facebook and Instagram ads to get page follows and email signups on the Artyst website. I can’t say I regret this — in the case of Artyst, social media ads have actually worked really well for gaining exposure. But this is by no means the only way to do it.
Most people won’t have the budget to drop this kind of money on ads, which is understandable. And even for those who do, it doesn’t always work well. This depends on where your target users hang out and how easy they are to nail down using Facebook’s ad audience parameters.
Looking back, I would’ve given more attention to non-paid social media marketing — the kind where you engage with users directly. For example, I would’ve followed Instagram and Twitter pages with high engagement among musicians (eg, XXL, Murda Beatz) and engaged with the commentors in their posts. Second, I should’ve made connections in online forums — there are several really good subreddits for different music niches that could’ve given me really valuable feedback and been a great source of early users. You can likely apply this strategy to your own niche.
Valuable relationships with users
While I built a decent audience for Artyst, my main regret is not delivering value much much sooner. I should’ve given them a small V1 to work with. I could’ve even created a forum for early users to start networking and building a community on until the first version of the app was built. It was a mistake to keep them waiting all that time.
Not only would this have shown more respect by not wasting their time, it would’ve also given me valuable feedback that I could’ve used to iterate much more quickly. If there was a UX flaw in my idea of how Artyst search should work, I could’ve learned that in the first month or two, instead of waiting a year later.
I had real users, but I didn’t give them a real product to use, so I had no feedback to use for directing the app’s roadmap.
Lesson: Deliver a real product ASAP. Get real users to give you feedback and iterate quickly.
Conclusion
In a nutshell: if I was re-starting Artyst today, I would do the following within the next month:
- Make a small app on Adalo that supports user profile creation and search (messaging could be handled on linked socials)
- Make a tiny landing page with an email grab
- Engage with the music community on Twitter and Reddit (using my personal accounts) directing them to the landing page
- (and probably still run some IG ads…)
- Launched ASAP and begin iterating with real users
This is a 50,000 foot overview of what I’ve learned launching Artyst over the past year. I hope it gave you some useful thoughts to consider as you dive in with your app.
Dislike ads? Remove them and support the forum:
Subscribe to Fastlane Insiders.
Last edited:
