There are two ways to learn stuff: realizing success to learn what works, and realizing failure to learn what doesn't work. This is about the latter. Specifically, it's about what types of marketing activities are ineffective for promoting iPhone apps on the App Store. Unfortunately (for me), I realized a 100% success rate in learning what doesn't work. I'm posting this because I hope it will be useful to other small indie developers trying to market their wares on the App Store.

When I released the original version of That Word in 2009, I did essentially no marketing¹—and the sales figures reflected that. At best, I achieved a couple sales a day, and it wasn't uncommon to have a dry spell of three or four consecutive days here and there. I didn't create That Word out of commercial interest (I was on “sabbatical”, after all)—so I wasn't too disappointed about the sales.

For That Word 2, however, I decided I would put more effort into marketing to see if I could actually make any real money from the App Store. I started reading several marketing blogs and studied the really successful indie iPhone developers² to try to emulate their secret sauce.

Here's a summary of what I did for marketing:

• I did a press release to announce That Word 2, using the services of PRMac, costing US$19.95.
  
• I ran ads on Facebook targeting people who stated an interest in word games, were iPhone or iPod touch owners, and lived in an English-speaking country. I ran the ad only for three days, which yielded 406,734 impressions and 108 clicks (a click-through rate of 0.027%). The total cost was US$39.92. Facebook offers advertisers either CPC (pay per click) or CPM (pay per thousand impressions) pricing models. I chose the CPC option, with a bid of $0.45, lower than Facebook's recommended bidding price³. My average CPC ended up being $0.37, and the CPM was $0.10.
  
• I ran a one-week ad campaign on the Fusion ad network. I chose Fusion Ads for a several reasons. First, they offer a one-week campaign for US$400, which was an amount I was willing to risk, and less than what other ad networks charge⁴. Second, I believed that the target audience served by the Fusion ad network would be technically savvy and have a higher-than-average adoption rate of iPhones and iPod touches. Third, I saw some of the “really successful” indie developers advertise on the Fusion network, so if it worked for them, I thought it might do likewise for me.
  

  The ad received 112,867 impressions and 614 clicks (a click-through rate of 0.5%); this translates to CPC of $0.65 and a CPM of $3.54.

• I submitted my app for review to a few sites (the major ones), but got no reviews⁵.
  

The net effect of all these marketing activities, at a cost of US$459.87, was a doubling of sales for two weeks, and then a return to normal. “Doubling” is usually pretty good, but when you have only a trickle of sales, doubling the sales is still just a trickle. So, overall, it was a bust.

Lesson learned: there's no ROI on web-based ads for marketing iPhone apps. Your marketing dollars and effort are better focused elsewhere. At a CPC of between $0.37 and $0.65, and Apple taking $0.40 on each sale, your ads need an impossible 30% conversion rate on a $1.99 app just to break even.

I'm convinced you need to get some kind of mainstream press⁶ exposure to have reasonable success. The economics of selling something for $1.99 precludes using conventional CPC or CPM-based advertising at the current market rates; this only makes sense when your product costs much more, or you can spend enough to reach a critical-mass of users and then subsequently benefit from large-scale word-of-mouth exposure. As an indie developer, I don't have the means for a big ad campaign to reach critical-mass in this fashion.

I need to figure out how to get exposure from the mainstream press.