Financial Realities of the App Store
December 10th, 2008
There have been a lot of uninformed people spouting nonsense about Craig Hockenberry’s “Ringtone Apps” blog post, so I thought I’d set the record straight with some real numbers from the App Cubby bank account. There are some amazing success stories on the App Store, but for every successful developer there are quite a few who haven’t come close to recouping their investment of time and/or money. App Cubby is doing quite well, all things considered, but it took a “Staff Pick” from Apple to get us over the hump.
Many of the responses to Craig’s post focus on the need for marketing vs. depending on the App Store for sales. While I agree with the need for marketing, the only methods of marketing I’ve found to be measurably cost effective are working with the press and getting featured by Apple, both of which are essentially free, but incredibly hard to guarantee.
As opposed to what many have assumed, blogging about the App Store is not actually a good marketing strategy. Craig may see a bump in sales (though it doesn’t look like it from the iTunes rankings), but in my experience of having tens of thousands of people flood to the App Cubby blog last week, very few spent time elsewhere on my site, and even fewer actually purchased one of my apps. Craig’s motivation (and mine in writing this response) was to talk honestly about the App Store in a way that spurs thought and moves the iPhone development community in a positive direction.
The closest thing I’ve seen to a "business model" for marketing iPhone apps is to advertise like crazy until you get into the top 50 and once you’re there, the top 50 list will start generating it’s own buzz. Then, just throttle the advertising to keep it in the top 50. But that’s not a business model, that’s like rolling the dice at a casino. I could spend tens of thousands of dollars on marketing and end up losing my shirt. Tap Tap Tap made it work, but I’d argue that they are a very unique case. Where To? was in the App Store on day one and Tap Tap Tap was able to use the first month’s sales to experiment with a $56,000 advertising budget. They’ve used those early lessons and another big wad of cash to help make the launch of Classics an incredible success. But with all that money, experience, and a great app, it still took a price of $0.99 to get Classics back into the top 50 once sales started dropping.
Well then, why shouldn’t every developer just drop their price to $0.99 and make it up on volume? If every developer dropped their price, volume would definitely go up, but it would just be spread across the top 200 rather than the top 50, and it would still take a top 50 app to justify months of development. That scenario doesn’t pose any less risk for developers than the current system and once the price expectation is set at $0.99, niche apps will never have a shot at profitability. You may only think 20 of the 10,000 apps are actually useful and/or fun, but each of the 20 million iPhone/iPod Touch owners is going to have a different list of 20, so the App Store should be able to sustain several thousand great apps, not a couple hundred.
Let me put it another way. Wouldn’t you rather pay $10 each for 20 AMAZING applications than pay $0.99 for 200 gimmicky applications with a few good ones mixed in. If the App Store landscape doesn’t change, developers like Craig Hockenberry aren’t going to take the time and spend the money necessary to create some of those amazing apps. Sure, the big game companies, VC funded developers, and big brands like Target will continue throwing money at the App Store, but some of my favorite apps so far have come from independent developers, and I would love to see what amazing apps are currently stifled by the need to hold down a day job.
So, back to marketing... one real business model for efficiently selling low cost goods on the internet is to have a laser focused marketing strategy developed through trial and error by obsessively tracking results. Well, Apple doesn’t allow developers to track sales from link to purchase. Until there is a way to track and optimize marketing dollars, it’s incredibly hard for me to warrant going in the hole on an advertising spree or dropping my prices to $0.99 hoping that it will pay off once an app hits the top 50.
Instead I’ve chosen a more methodical approach, dipping my toes in various marketing strategies and measuring the results as best I can.
Macworld: I’ll start with the earliest marketing attempt, a banner ad on Macworld.com. I knew that App Cubby’s first app, Trip Cubby, was a very niche app, so I worked with Macworld on targeting my $600 ad buy. I bought 30,000 impressions that would be triggered by a combination of iPhone AND business. Those 30,000 impressions netted about 120 clicks (%0.39 CTR, bad, but not terrible for such a niche product). If EVERY SINGLE person who clicked on the ad actually bought Trip Cubby ($10), our income after Apple’s split would have been $840. Without click tracking there is no way to know exactly how many clicks turned into sales, but I doubt it was anywhere near the number needed to break even, much less turn a profit on this endeavor. To be fair, I don’t think that this is the fault of Macworld’s ad inventory. The numbers make a lot more sense for businesses selling iPhone cases, desktop software, and maybe even broad market iPhone apps.
AdMob: The lack of click tracking was still a challenge, but with $5,000 in free advertising I was able to take a more aggressive approach with my experimentation on AdMob. There was a 3 day window where the initial sales boost was slowing down, and no major press hit the web. During that time I scaled my AdMob budget up and down to see how it affected sales:
The AdMob experiment deserves a bit more explanation. When I was first notified about getting $5,000 in free advertising I was blown away. I spent hours crafting the perfect sales pitch and started buying ads the moment the money hit my AdMob account. I spent over $500 (2224 clicks) the first day and couldn’t wait to check my sales the next morning. Well, sales barely moved, and I had spent 3 times more on advertising than I saw in gross sales. So, since the ads were paid by the click and not the impression, I decided that it would be my goal to actually deter people from clicking on my ad. I created ads like: “Trip Cubby • Business trip log $10” and “Business mileage log. Buy now! $10” Surely people wouldn’t click on my ad unless they knew what there were getting into and would be more likely to actually purchase Trip Cubby. Still, no noticeable bump in sales.
I decided to stop advertising Trip Cubby and save the rest of my free advertising for Gas Cubby, which is a bit less of a niche product. After spending the last $3,000 buying ads for Gas Cubby, I still haven’t seen a pattern that would confirm AdMob to be a cost effective marketing solution for my apps. This is by no means a conclusive test of the overall effectiveness of AdMob, but the numbers are pretty telling. The most compelling thing about AdMob is that you can bid as low as $0.10 per click to send iPhone owners directly to the point of sale. It’s possible that AdMob will be more effective if Apple allows app trials. The 42,000 people who clicked on my ads could have download the app with no risk, and purchased it later if they enjoyed using it.
AdWords: I’m admittedly not an AdWords ninja, so I took a very conservative approach, bidding low on highly targeted ads. The ads have throttled up and down as cheap ad space comes and goes, but have added up to almost $600 in 4 months. Without a way to track which clicks turn into sales, it seems foolish to throw money into a black hole bound by competitive placement and fraught with click fraud. Not being able to make AdMob work at $0.10/click makes me even more pessimistic about AdWords. Big companies have driven click rates for iPhone related words through the roof. Even my conservative approach ended up averaging $0.37/click!
Freemium: Many developer’s have chosen to offer an “Lite” version of their app. There are two main freemium strategies: 1. Put ads in the Lite version to make some money, then try and convert users to the premium version 2. Limit the features of the Lite version and use it as a free trial for the premium version. As opposed to what most people think, this strategy doesn’t seem to be paying off for most developers. Robert Marini, a developer for Pinch Media, posted a tweet yesterday saying: “There are apps that benefit from free ad-supported versions, but not nearly as many as people think. Most just shoot themselves...” He ran out of room to type on twitter, but I have a feeling he may have said heart or crotch rather than foot. Since Robert works at an iPhone analytics and advertising company, I’m going to take his word on this one. I’ve been tempted to release Lite versions of my products, but am holding out hope that Apple will allow free trials at some point in the near future.
Press: Having great apps, decent communication skills, and a bit of luck has gotten App Cubby more free press than most iPhone developers. We’ve been featured on: daringfireball.net, macworld.com, zdnet.com, tuaw.com, appleinsider.com, theappleblog.com, gizmodo.com, appcraver.com, edmunds.com, appvee.com, macnn.com, and many more blogs and app review sites. All these articles have helped sales and will continue to send customers our way over time, but the Gizmodo post hit at just the right time to be able to see it’s direct affect on sales:
Spikes are nice, and quite welcomed, but I’m trying to build a business, not a roller coaster! With enough good press spread across a few weeks, an app can build a base, and gain some traction, but a single article is often not enough to propel an app into stardom. As with much of my efforts at marketing, things would probably be a bit different if I weren’t selling niche apps. A game that gets a link from Gizmodo probably does much better than Gas Cubby did.
Help from Apple: Apple does a very good job of insulating developers from the people who actually run the App Store, so I have no idea who makes the final decisions (Steve himself?) and why some apps get featured and others don’t, but I’m forever grateful that Gas Cubby was somehow selected as a “Staff Pick.” The graph doesn’t need much explanation:
Where do we go from here? Since Apple pays monthly (and December sales won’t be paid until late January) I have a pretty close estimation of App Cubby’s year end numbers. Here is a basic breakdown of where App Cubby will be at the first of the year after all the bills are paid:
Total Income: $65,000
- $24,000 - startup funds borrowed from family members
- $32,000 - income from App Store sales
- $5,000 - free advertising from AdMob
- $4,000 - 1997 Honda Accord - my wife and I sold one of our cars to keep the business going
Total Expenses: $65,000
- $29,000 - Programmers: I’m not going to be specific, but I definitely paid less than $150/hr.
- $15,000 - Personal Salary: 80 hours a week since March 6th puts me at around $5/hr.
- $7,000 - Marketing: AdMob, Macworld, AdWords, etc.
- $5,000 - Legal, administrative, equipment, webhosting, etc.
- $4,000 - Artists: Described in an earlier blog post
- $3,200 - Charity: Sent to China for earthquake relief and humanitarian work
- $1800 - Investor Repayment: $360/mo. since August 1st. as per the promissory note I signed with my family
We’re in a great place right now, having already made enough money in the first couple weeks of December to cover my salary and administrative expenses for February (the December check will cover January). Sales have been trending downward and could drop like a rock at any moment if a solid competitor moves into my niche, but whether sales go up or down, there is still a bit of a chicken and egg problem; with the December check already spent, I won’t see my next influx of cash until late January. So, dear readers, should I follow the path of most US businesses and borrow against future earnings to do more marketing? Should I slow down development and do a marketing blitz the moment my January check rolls in?
I’m no marketing guru, but I’ve done as much marketing as I could afford (startup funds ran out in July), and have learned some interesting lessons along the way. If I can find a cost effective way to market my apps, I would be willing to borrow money for marketing, but so far the strategy that has made the most fiscal sense is to spend money on development and hope for good placement with Apple and the press. With our next app scheduled for launch before Christmas, January could be the month App Cubby really breaks lose, but in February the dilemmas will remain; where’s the sweet spot between marketing and development? Is there a truly cost effective way to market $5 iPhone apps?
And this whole post leads to questions I hope are being discussed within Apple. Would free trials help raise the quality and price of iPhone apps? Would providing click tracking to developers spur cost effective marketing? Is the top 100 list in its current iteration good for the platform? How can the App Store be improved to help users with search and browsing?
- • I Was Wrong
- • Give Freemium a Chance
- • Beta Testing
- • Appville Interview
- • Build Your Own App Store
- • Taking Care of Developers
- • The Sparrow Opportunity
- • Goodbye Tweet Speaker, Goodbye Twitter
- • A Letter to Kirby Ferguson
- • The Sparrow Problem
- • Do Your Own Damn Work
- • We Were Wrong
- • Going Pro
- • Stretching Reality
- • Apps, Apps, Every Where
- • Apple Deprecated Flurry’s Treasure Trove of Data
- • OS X @2X
- • Please Steal My Ideas
- • Free and Low-Cost Apps