[espadagroup]

Pay for exposure already exists within the app store. Large companies pay around $10-$20K a day depending on the category on crap ad networks (won't name them) that incentivize the users to download the companies app in order to "get to the next level" or something like that. The downloads are incredibly cheap and you get thousands and thousands of them. The app store ranking is almost entirely based on download velocity, so your ranking goes up. Once you get in the top 12ish the organic downloads start pouring in just from being near the top. The value per user of those organic users is so high that it more than off sets the cost for the crap/incentivized downloads. Almost everyone who can do this, does this.

[physcab]

Just to nitpick a little. It's not that those organic users are so valuable that everyone who can promote does just to get them. It's that organic users + incentivized users lowers your eCPI and can make your app (usually a game) profitable. But you can totally be profitable with just incentivized users as well...you just need to make sure your retention and revenue per user are higher than your costs in acquiring them :)

[derefr]

> The app store ranking is almost entirely based on download velocity, so your ranking goes up.

I wonder how much the App Store's rankings would change if Apple only counted a download "impression" once the app had had, say, 10 minutes of (non-contiguous) runtime on your device? These perversely-incentivized game users would be filtered out of the metrics, but all the drive-by "download, look around, oh this app isn't what I wanted, delete" users would as well.

[unthunk]

This wouldn't work well for all apps. For example, I went on a trip to the Rockies in November and wanted to check the elevation at the lodge I was staying at. I downloaded and ran elevation app to find out. Took only moments to find out and even with fiddling around the the app while hiking I probably spent no more than 5 minutes on it but it was completely useful for the purpose I had downloaded and used it for. Should limited use/one-time use apps not be counted simply because they weren't used for an arbitrary amount of time?

[kevingibbon]

Lets say you need 20k DLs to get top 5 in a notable category. Lets say you get another 20k of real DLs from the app store ranking. Incentivised DLs are $0.35 each, but they are Dls and not real users. So $7k for 20k DLs.

You usually want a LTV of 3:1. So your LTV would need to be $2.10 per user. These users are way worse than acquiring them from other channels because they only briefly see your logo and a brief description. Let's say they are only 25% as good. So now you need a regular LTV of $8.40. Very few apps have a LTV of over a dollar.

[gyardley]

The "let's say they are only 25% as good" is a bad assumption. The real downloads you get from increased app store ranking are as good as any others you'll get.

The ratio - 20K real downloads from 20K incentivized, or 1:1 - is also a bad assumption. Here the actual ratio varies wildly, depending on how much the incentivized installs actually influence your app ranking. It's a feast or famine situation - spend too little, and your ratio will be closer to 1:10, since your ranking won't have moved enough to influence organic traffic. Spend sufficiently, and your ratio will be closer to 10:1, if not higher.

The LTV of 3:1 is also a bad assumption - I've worked with firms who would spend at breakeven all day long, often because they wanted to grow their user base into an acquisition or use it for future in-house cross-promotion. Many others are happy to make (say) $1.50 on $1.00 spend.

My experience agrees with the parent comment - anyone that has the budget to buy incentivized downloads does buy incentivized downloads, except for a small proportion of developers who strongly disagree with the practice. It simply works too well.

[kevingibbon]

Obviously on a case by case basis. I've seen app downloads that come from rankings to be lower quality than those that come through other channels.

If you are trying to build a real business a 3:1 LTV is very reasonable.

[ableal]

> You usually want a LTV of 3:1. So your LTV would need to be [...]

Lifetime value of a customer ? ( http://en.wikipedia.org/wiki/Lifetime_value )

[kevingibbon]

yep.

[espadagroup]

Plenty of apps do; retail and real estate, just to name a few.

[spontaneus]

Facebook just recently started doing this as well. You can now pay to get your app promoted in people's news feed.

[kevingibbon]

Completely different to AppGratis. Facebook allows you to target (based on gender, location, likes, workplace) and advertise your app. Users have no incentive to DL your app and actually have a need for your product.

I've had a lot of success on FB mobile install ads.

[hboon]

Mind if I drop you an email regarding FB mobile install ads? Have been experimenting with it, but not much success. Want to pick up a few tips.

[potatolicious]

This isn't quite the same thing. Facebook is letting people advertise apps on news feeds, they're not forcing downloads via a game mechanism.

What GP is talking about is the much sketchier model of blocking (or greatly slowing down) access to an app unless the user goes and downloads an affiliated app. This is artificial download inflation by bending the user over a barrel, not advertising.

[physcab]

I don't think Apple will approve an app that blocks a user or slows them down unless they install a 3rd party app. What usually happens is an advert pops up during normal app usage, the user has to exit or let the ad go away, then they can continue using. It's a normal advertisement. Its not unlike when the NYTimes does a full site takeover for some Buick commercial and you have to minimize the ad before you can continue reading. But going out and buying a Buick certainly doesn't decrease your access time to NYTimes articles.

[nkorth]

I've seen many games that let you "earn" ingame currency by downloading apps or completing surveys. A few of these games _indirectly_ require you to download the incentivized apps by providing very few coins in the actual game, or making the ads the only way to get coins (other than with real money). Apparently that's sneaky enough for Apple.

[zevyoura]

How does this work? My understanding is that apps cannot see what other apps are installed on iOS unless the device has been jailbroken.

[fnayr]

For iOS apps there's something called "custom url schemes," which is basically a string that every app is self-assigned (if at all) that can be used to open up the app from within another app (or check the app's existence).

[Termana]

I'm not sure if this is how the specific ones being talked about might be working, but a simplified way of doing it would be that when APP A is installed it connects to a server and tells it that it has been installed with UNIQUE DEVICE ID, that way APP B can connect to the server and see if APP A has been installed on UNIQUE DEVICE ID. You can imagine ways to do the same with viewing ads etc.

[pdpi]

Not exactly slowing you down if you don't get a 3rd party app but e.g. Draw Something will give you extra coins to watch ads for other apps.

[espadagroup]

Yeah, though that's still controlled by Facebook, since you're actually paying them rather than some third party.

[tayome]

Facebook's app value proposition is not to be an app store - it's a social networtk

[jjellyy]

If this is happening those ad networks should be banned. Incentivizing a user to download an app is against app stores TOS.