3% ROI for selling my iPhone app using MochiAds | Hacker News Reader

3% ROI for selling my iPhone app using MochiAds

andrewgarrison.com

31 pointsavgarrison17 comments

No one has been able to advertise paid applications on the iPhone platform and get a positive return from the advertising itself. The name of the game is to generate enough downloads to get onto a 'top paid app' leaderboard in your category, which will then expose you to a bunch of buyers who wouldn't have seen your app otherwise. In other words, you're indirectly paying for placement.

All cost-per-click networks will have abysmal conversion rates for paid apps. Cost-per-install networks shift this risk from you to the network, since you only pay per install. However, paid apps advertising on straight cost-per-install networks delivers minimal volume, which makes them not worth doing - while the risk to you is gone, the conversion rate still sucks. Some cost-per-install networks (Tapzilla, Apperang) require you to pay them more than the cost of the app, so they can bribe users to install the app. This works, but it automatically results in a negative ROI on the advertising itself - since they take a cut, Apple takes a cut, and the user gets reimbursed through PayPal. If your spend isn't enough to get you onto a leaderboard and get that sweet organic traffic, you wasted your money. Right now these user-bribing networks are still small, so this is a real risk. Buy from everyone simultaneously.

Happily, there's a lot more options for free applications. If you've figured out how to make sufficient money from your users through virtual goods sales (like one YC company, Addmired, although there are many others), and therefore can afford to advertise, there's many cost-per-install options you can use to effectively buy placement - Flurry (which is mine), TapJoy, MDotM, Free App A Day, many others. Again, the prudent-but-flush developer buys from a lot of sources simultaneously to ensure their application ends up on the front page of iTunes.

The economics of CPC advertising are punishing at low customer LTV, unless you are getting unbelievably cheap clicks. I wouldn't be able to make the math work well for BCC at $.10 a click, and I have four years of conversion optimization and $30 price points to fall back on.

Your quality score must be 10/10 on all your keywords.

Does that necessarily follow? Are you assuming that he does or can get 10c clicks?

I've spent about $8,000 in 2010 six to eight cents at a time. You don't literally need 10QS to do so, but a generally high QS is one factor. Others include Conversion Optimizer, several years of history, rigorous pruning of nonperforming keywords, being in a niche which has lowish competition for the keywords I care about, being able to outspend other competitors through having a higher price point, etc.

Incidentally, I am not very good at AdWords.

I've spent about $6,000,000 on adwords in 2010 six to eight cents at a time, and you're better than you think at it.

Andrew, you completely miscalculated your ROI:

(gain - cost) / cost = ROI

($4 - $100) / $100 = -.96

Your ROI was therefore -96% and not 3%.

Sorry to hear about your terrible returns, but thanks for sharing nonetheless.

Could part of the issue be the ad? I don't mean to offend, but it didn't seem like it would inspire many clicks. You might want to give a better indication of what the game is about, and give a really strong CTA (call to action). Something like "CLICK HERE" usually does the trick.

I think there's some real merit to this comment. In reading the original post, I didn't realize the image embedded in it was the ad itself and not just a screen from the game. You might consider consulting someone experienced in creating ads next time around, if only to factor that out as a consideration.

Getting clicks is not his issue if they are happy to drain a hundred bucks in a day. The issue is that he needs to convert about 20 - 30% of the clicks to purchasers to make the math work at his price point. That is extraordinarily high.

Ah -- you're right. I now see that the app sells for 99 cents.

To slightly rephrase: even without counting Apple's and LinkShare's cut, the developer would still need a 10% purchase rate just to break even (10 cents * 10 people = 1 purchase) (which is just as ridiculously high as a 20-30% conversion rate as far as I know).

You probably paid for a lot of clicks from people who were curious but don't have devices, and of course you might lose people on your page in the app store.

I would suggest a better creative and more importantly, since you're PPC put the app's price on it so casual browsers and people who don't buy apps won't click thru - "ON SALE!! $normalprice" in one of those red starry-circle things might help a lot.

You should check out Tapzilla - http://tapzilla.com (YC S2010) - you only pay for actual app installs rather than paying $100 for 994 clicks, which leads to 4 installs (I had no idea install rates per click were as low as 0.40%!). They just launched on Techcrunch last week (http://techcrunch.com/2010/10/13/tapzilla-offers-daily-deals...).

Edit: I can't claim to be an expert on mobile advertising, but it doesn't seem like traditional CPC/CPMs really work in this space. I'm on a mobile device, which means lots of accidental taps, and if I randomly click on an ad that catches my eye, chances are I'm not going to want to install the app from the app store right then and there (although I guess this also means your 4 installs could be a low stat since users might return to install the app later without going through the Linkshare affiliate link).

Andrew, thanks for sharing your numbers man. Raw numbers are one of the most valuable things for learning.

3% ROI isn't great from a cash perspective, but there's other benefits you're getting - those 100,000 impressions are potentially worth something for people recognizing your product later, then there's getting more reviews, potential word of mouth, etc. Also, someone that checked it out first from the ad might buy later, especially if they get repeated exposure.

The danger with such a low return is that if you're not watching carefully, you could run in the red pretty quickly and lose money. But as long as you're above break even and paying attention, your real ROI is probably a bit higher than that 3% considering the other positive effects.

Edit: Wait a minute, you got 4 sales... do you mean you spent $100 and got $3 in return? I was thinking you got $103 in return... if you're in the hole $97, that's utterly abysmal and discard the rest of my comment. No positive secondary effects gonna make up for that.

I'm really embarrassed that I used the wrong number for the ROI. I fixed the article, but unfortunately I can't change the title of this link on Hacker-News. Hopefully I don't lead people astray making them think that MochiAds is a viable option for advertising their iPhone apps!

Has anyone had any success advertising their iPhone apps? I'm disappointed MochiAds didn't yield a higher ROI. It seems like their ads could be more effective, considering they are blasted at the user, in the center of the content they are looking at.

I had depressing ROI with advertising my iPhone app on AdMob. I can't remember the exact figure, but depressing is the word I can think of to describe it.

Reply on news.ycombinator.com