It has been a long time since I last wrote about my journey of becoming (or, attempting to become) an independent iOS developer. In fact, given that the last post in this series was written in September 2014, it has been exactly two years. Considering many have written to me since that time asking about the next part, and since so much has happend since then, I thought it about time I finally wrote that update :)

Recap

If you haven’t read them before, I suggest you read Part 1 and Part 2 before proceeding with this post. In the mean time, here is a very brief summary:

In the first half of 2014 I came to a crossroads in my career as a developer. With the help of my very patient and understanding wife, we made a decision: borrow money from extended family, purchase outright the IP to the SongSheet app I had developed with a friend, and try to turn it into a living. The transaction concluded on 1 July 2014, and from that point on I was an “indie”.

Given the money that we had, there was a time limit: if things didn’t work out by the end of the year, we would likely have to sell our house, pay back the loans, and then figure out something new to do.

I also had a plan:

  1. Aggressively upgrade the capabilities of the SongSheet app itself.
  2. Implement better online marketing, starting with a better website.
  3. App Store Optimisation (ASO) improvements.
  4. Extend SongSheet’s reach to the iPhone.
  5. Possibly experiment with other revenue models (e.g. free with IAP).

So, how’d it go?

In short, I didn’t make it. But it wasn’t the end of the world, either. In summary:

  • We still have our house :) The loans aren’t fully paid off, but we are making steady progress on repaying them.
  • Revenue grew rapidly, but not rapidly enough (see below). The addition of a second app (Reverse Chord Finder Pro) made a difference, but not enough.
  • By necessity, I did some freelancing work. It was easier to find this work thanks to my demonstrated capability on SongSheet. I managed to secure a number of short to medium term contracts (all part-time) that filled in the gap financially, but it took a lot of time away from developing SongSheet.
  • I grossly under-estimated the complexity of getting SongSheet working acceptably on the iPhone (the subject of another blog post perhaps). I literally only released this last week, some 18 months later than I’d originally planned.
  • Revenue is still growing, and I am oh-so-close to being able to be full-time indie as I originally hoped.

Revenue Growth

When I last wrote, this is what SongSheet’s monthly revenue looked like until the end of August, 2014 (all figures are in AUD, my native currency):

SongSheet monthly revenue July 2013 to August 2014

This is what SongSheet’s revenue looks like up until the end of August, 2016 (to put this latter graph into perspective, I’ve highlighted where the first revenue graph fits):

SongSheet monthly revenue January 2013 to August 2016

That’s quite an improvement, I’m sure you’ll agree. It was a steep rise up until early 2015, after which growth continued at a reduced pace. While impressive, the growth was not nearly fast enough; SongSheet was still pulling in less than $2000 / month by the end of 2014, and this was about 1/4 to a 1/3 of what I actually needed to make ends meet if that was my only source of income.

As another comparison, here is a graph of SongSheet’s monthly average revenue by year since its release:

SongSheet monthly revenue by year to August 2016

2013’s average was $264/month. So far, 2016 to date has averaged $2800/month, and is set to finish higher than that if current trends continue. That is a 10-fold increase! I knew there was potential in the app, and I believe even now that I still have only just begun to unlock it.

But that’s not the whole story. SongSheet was joined by a sister app in January, 2015: Reverse Chord Finder Pro.

I didn’t develop the original version of Reverse Chord Finder and it had been around on the App Store for years already. However, the original developer no longer had the time to give it proper attention (in fact, it had not seen an update in quite some time already), so he shopped around for someone to take it over. As it was a music app, I saw the perfrect opportunity to add to my portfolio an app that would sit quite nicely alongside SongSheet.

We concluded a deal that would allow me to pay a small up-front amount, and an ongoing base amount + percentage of royalties for a period of 2 years. This was structured in such a way that I was pretty certain that given the app’s previous year’s average monthly revenue of approximately $550/month I should be making a profit on the app each and every month.

With a little bit of love, Reverse Chord Finder rebounded from its lows, generating an average of $733/month in 2015 and $980/month in 2016 so far. Granted, those kinds of figures don’t change the world, but it’s still another big step towards my eventual goals. Since taking over ownership, I have pushed updates that got the UI to work on modern iPhones, I rewrote the chord finding engine and added a bunch of localisations (I hope to write a blog post on this last topic at a later time).

My hope with Reverse Chord Finder is that I could also do cross-promotion between it and SongSheet; up until now I haven’t really followed through on that (for various reasons), but between now and the end of the year I’m ramping up in this space and am excited to see what will happen next.

In summary, this is a graph for combined app revenue up until the end of August:

Total app revenue January 2013 to August 2016

And this is how average monthly revenue looks for each year so far:

Total monthly revenue by year to August 2016

Not bad, and not really that far short of my goals. All I need to do now is double the last few months’ revenue. Considering that I’ve already doubled overall revenue 4 times already, one more doubling doesn’t seem like that big a deal at all.

Revenue Models

How to monetise apps on the App Store continues to be a hot topic of conversation. My impression of current discussion suggests a near-universal consensus that paid-up-front is dead; only ad-supported, subscriptions or some other IAP model is viable. In contrast to this obvious trend, I have stubbornly (or lazily?) stuck to the paid-up-front model. (I won’t attempt to justify my decision in this regard in this post; I’ll leave that for perhaps another.)

Given that, I have experimented further with the pricing of both apps.

You might remember from my second post in this series that I conducted the following experiment:

  1. I reduced SongSheet’s price from its usual US$4.99 to US$0.99
  2. At weekly intervals, I raised the price to the next tier.
  3. I terminated the price rises when I felt that I had identified the price tier at which I achieved peak revenue.

While lower prices saw higher downloads, the extra downloads didn’t offset the lower price sufficiently. As a result, I settled on a new standard price of US$6.99. Apart from a few short-term price changes, I further raised the price at the beginning of February 2015 to US$7.99 (without running any formal experiment as I had before). After seeing promising results at that level, I again raised the price to US$9.99 at the end of March 2015. It was at that point that my courage ran out and I was reluctant to push the price any further. However, each price increase saw increased revenue, suggesting I may have terminated my original experiment at much too low a level.

It was also encouraging to see that a few of my larger competitors had also significantly raised their prices. Considering that apps such as SongSheet are tools that musicians use all the time (many of SongSheet’s users use it at gigs almost every single day), I think it’s easier to justify the value of an up-front purchase, compared to many other apps that consumers consider to be “throw-away” purchases (e.g. games).

My last price increase occurred just a few weeks ago, in anticipation of the imminent release of SongSheet’s iPhone version (my decision to release the iPhone version as a separate app to the iPad version could also be the subjecct of another blog post). SongSheet for the iPad (now called SongSheet Pro in order to distinguish it from the iPhone version) is now selling at US$14.99. This is a 50% increase on its previous regular price. And it is doing well. I chose that price so that the new iPhone version could come in nicely at the iPad version’s old price of US$9.99.

The main concern I have with my pricing model is what it might do to my overall revenue growth curve. What I mean is this: a lower price might mean lower revenue right now. However, if it leads to more dedicated users of my app, then the long-term growth curve may be steeper as the word-of-mouth effect has more people to work with. Unfortunately, this isn’t something you can just experiment with easily; it would take years of data (I think) to answer this question one way or another.

I performed a similar pricing experiment with Reverse Chord Finder, which had been priced at US$9.99 prior to me taking it over:

  1. I lowered its price from US$9.99 to US$1.99.
  2. Every 2 weeks I raised the price. The two week period was an attempt to somewhat filter out the noise of week-to-week variations.
  3. I terminated the expriment once I reached the original price again.

What did I learn? That, just as with SongSheet, it made the most money at US$9.99, so that’s where I left it.

If you have a paid up-front app, do not be afraid to experiment with price - you might find that you’ve been under-charging for years and leaving a lot of money on the table. (Conversely, of course, you might have been over charging; but you won’t know if you don’t experiment).

I highly recommend you read Michael Jurewitz’s blog series on app store pricing for a great discussion on how to identify the ideal price point for your apps.

Where I’m at right now

Last week (specifically on September 7) I finally released the iPhone version of SongSheet. With it, I also took advantage of the fact that I now have two paid iPhone applications to create an app bundle containing Reverse Chord Finder and SongSheet. So, this is my app portfolio (app prices correct at time of writing this blog post):

Once things settle down from the release I should have a good idea of what the new level for ongoing revenue will be; initial indications are very positive. I also finished up a contract a month or so back and haven’t secured any other work, so in a sense I am full-time indie right now, and loving it :) The next few weeks will be quite important in deciding my longer term future.

In the mean time, my wife and I are getting ready to fly to the US in order to attend the Release Notes conference (September 27-29) in Indianapolis. If you are going to be there, we would love it if you came up and said hi! We’ll also be in LA for a few days beforehand (we arrive in LA on September 23) and in Austin for a few days afterward. We fly back home to Australia on October 3.