Android Income Report | Hacker News Reader

Android Income Report

droid-blog.net

32 pointsandroidoka35 comments

I'm a successful app developer and I can't help but feel that this kind of approach is not the right way to go about.

App-stores are "hit based", at that level of revenue you should be thinking, "how do I make this go up an order of magnitude", not "how do I optimize my ads to squeeze a little more revenue out". Just taking a look at some of the reviews your apps have... you should be improving quality before optimizing revenue.

One of the reviews:

"Update: Facebook Ad is covering up my ship! I want old version back"

Yea, that certainly makes me question his focus :S

Hi soup10,

I totally agree. My intentions were more research driven than by greediness. Please see my comment at the top for more details.

Best regards Johannes

One pet peeve I've got with Android and Google Play is that my country is still not supported in Google Checkout, so I can't register for a merchant account.

This means that I cannot sell apps, so the 3 options I have would be to:

1) go with an ad-supported model

2) enable in-app payments with a third-party provider, such as PayPal, but this is tricky because it goes against the guidelines of Google Play

3) publish the app on my own website, with payments through PayPal, being the only sane solution for me in case I don't want an ad-supported model

I'm extremely disappointed that Google, after 4 years of Android being on the market, is still not allowing my country. And I wouldn't have a problem selling iOS apps in iTunes. Not only that, but there have been numerous reports of extremely poor customer service for Google Checkout, so even if you're blessed to live in an approved country, you can still get fucked.

And then people wonder why the ad-supported model is preferred for Android.

You should really give SlideME a try, some have been pleasantly surprised by it. Maybe even Amazon's marketplace thing.

Which country?

Are you sure you cannot sell stuff even without merchant support? Here in my country (Brazil) the merchant account is also not supported, but they give me the option to use an AdSense account instead to receive money.

It's far from ideal, and makes me have to jump through additional hoops just to sell apps. But it's allowed.

I'm extremely disappointed that you'd choose to try an obnoxious, intrusive SDK like SellARing. Not only are you exposing your users to it while you app is installed, but they leave their SDK installed on the user's device after they've deleted your application and left it a one-star review for doing something so disrespectful. I imagine there is also a large number of users that are just not savvy enough to uninstall this SDK after-the-fact, so you've effectively hosed them.

I appreciate the need for app developers to make money off their work, but this is not the way to do it.

EDIT: I misread their product description - this SDK doesn't survive an uninstall of the offending application. I was assuming they'd use a permission like "install applications" to drop another application. At least on this count, you don't have to worry.

> they leave their SDK installed on the user's device after they've deleted your application

How does this work? I come from the iOS universe and statements like this scare me out of ever installing anything from the Play Store. Is installing an app not clearly reversible? I can't find anything useful for "uninstall sellaring" or "remove sellaring" either.

Is there a golden rule to keeping Android 4 clean?

> they leave their SDK installed on the user's device after they've deleted your application

I have worked on an Android app extensively. I am pretty sure that's not possible.

You're right - I misread the product description (updated parent post).

I was assuming they'd be doing something tricky like asking for INSTALL_PACKAGES (http://developer.android.com/reference/android/Manifest.perm...) and dropping something on the device. Thankfully they don't do that.

As someone who has hacked on Android before (ie: http://unrevoked.com), I wouldn't be surprised if there were ways to silently install packages, or even just tricking the user into allowing an install of a further package.

Quick googling leads to this: http://jon.oberheide.org/blog/2010/06/28/a-peek-inside-the-g...

Android apps can't have that permission, only the Market gets that permission.

No offense, but you're spreading some pretty false FUD. Unrevoked uses multiple exploits and requires a lot of user interaction and a computer to initiate the process.

Further that last link basically describes how Play remote installs applications and is all under the assumption that someone has somehow MITM the SSL connection, something I'm presuming is not at all easily done. (Further it's going on two years old, I wouldn't be shocked if Google is now signing their INSTALL_ASSET messages, Play has changed a LOT in the last two years).

Apps are sandboxed overall similarly in iOS, WinRT, Android, WP7. Uninstalling them, uninstalls them. It even removes all data attached assuming the dev doesn't manually put data on the SD card instead of using the API to store data on the SD card.

True. INSTALL_PACKAGES is a permission with protectionLevel=signatureOrSystem. All other applications can do is take the user to an install-Activity hosted by the OS. For my intention for installing SellAring, please see my answer above.

FYI, it is possible for unprivileged apps to invoke the INSTALL_ASSET functionality themselves. One such example described here:

http://blog.duosecurity.com/2011/05/when-angry-birds-attack-...

Another variation of that attack is still unpatched, allowing any app to invoke INSTALL_ASSET. Certainly that's not intended functionality and is a bug that will be (eventually) patched, but I wouldn't classify it as FUD.

That is indeed a little scary. Thanks!

One thing I'm always curious about when I see revenue from mobile applications is: does advertising on a mobile device really work? I'm hesitant to sell anything unless I'd be confident selling it with my name attached to the label and I'm not sure mobile ads are something I'd consider valuable.

The two primary issues are that 1) mobile devices slow down / severely discourage any online form entry and 2) the age old question of "how many clicks are misclicks".

1) is the killer for me. It's rare for me to actively click on an ad and even if I'm drawn to an ad whether I'm on laptop/desktop or a mobile device makes quite a large difference on the ultimate decision.

I guess you could argue that it's up to the advertiser to ensure their ad money is well spent but unless you're just spreading awareness (or possibly other applications?) I'd personally not feel happy selling mobile ads.

Thanks for sharing your numbers though! This sort of openness is something I quite enjoy seeing. Best of luck hitting your $2k goal next month and I hope the $4k goal is reached soon after :)

You are right to wonder. Mobile ads largely do not work. I tired admob as a publisher and a as an advertiser, results are poor. As a publisher I saw ad revenue on the order of $5 for 1000 active monthly users. As an advertiser I sunk in a few 100 bucks and saw I think one purchase of my apps that I advertise. Mobile ads are destroying value. I have once met a person at a conference who said he was making money in mobile ads. Turns out he knew peopple at abig companies willing to spend money on brand advertising. I still run the admob ads In my app, and I see that most ads are brand ads, and ads from companies willing to buy their way into minds e.g. Google or Microsoft.

It leads one to wonder if the mobile ad payment argument is simply a problem with Google. That is, Google's strength is in advertising, and it works for Google due to their size and reach, but that capability may not expand to their app developers. Certainly, some are making the argument that there's more money to be made by pushing your app through a traditional retailer than a traditional advertiser. (Erm, that last word may be a poor choice there... I just can't think of a better one.) e.g.: http://www.engadget.com/2012/03/31/flurrys-analytics-apple-a...

A while ago, the madvertise SDK had a bug which resulted in ads not being scaled. This resulted in very small ads that were clicked by accident almost all the time. After a while, this bug was fixed. When looking at the conversion rate for my house ads, it was totally obvious that the non-accidental clicks resulted in a respectable conversion while the accidental clicks had almost no positive effect. I think the google AdSense for mobile do a good jobs here because the user has to click them twice before anything happens.

It kind of blows my mind that any developer would willingly integrate something like Sellaring. It sounds like the kind of ad network no sane person would ever want anything to do with, and if an app I used integrated it, I would uninstall that app on the spot. I'm not shocked it had such a negative impact; I wonder if he spent any time considering how his users would feel about it?

Almost all ads to some degree are something that "no sane person would ever want". The difference is that Sellaring makes ads that are not easily ignored. In contrast, the on-screen ads on Android's AngryBirds are easily ignored even if animated and although I don't run AdBlock I rarely consciously see ads on webpages. The human brain is an amazing filter.

I'd say the advantage to Sellaring is that 1) the person actively listens to the message (it's more difficult to filter out audio ads as they're listening for the call to start) and 2) it's using up "dead time" if the ring was going to go for longer than 10 seconds anyway.

I can see both points of view but if you're not paying for the app... It also provides a strong incentive to upsell to the paid version of the application that loses the advertising as well.

> It also provides a strong incentive to upsell to the paid version of the application that loses the advertising as well.

It provides a much stronger incentive to immediately remove the app and leave a 1-star review.

Exactly this. If a form of advertising would actively annoy me if it was on my phone, I do not put it in my apps. Most users understand the need for ads in applications. They're willing to have a 50px banner because they're pretty easily ignored. Most users are not tolerant of apps reaching outside their scope and pushing ads, like this Sellaring and whatever the other one was that put ads in the notifications (AirPush maybe?).

To inspect the addons used by apps there is a brilliant app called "Addons Detector" which you can find out about the analytics, advertising, developer tools, social gaming, push notifications etc libraries used by the apps installed on your Android.

I was really interested when I first ran it! I have no connection to app, just a happy user.

For the millionth time:

   If you are not paying for it, you're not the customer;
   you're the product being sold. [1]

This is the inevitable end game of developing for a platform where the majority of users refuse to pay for applications. Also, note from the dude's blog that he offers an ad free version; for the price of $4.16 people refuse to be ad free. The install numbers are instructive: 740k vs 1046. I'll let you guess which is which =P

[1] http://www.metafilter.com/95152/Userdriven-discontent#325604...

Hey, sure, I made considerations on how my users feel before integrating this kind of ads. I'm speaking about different monetization strategies on conferences and while the bad impression of notification push ads was proven by many apps, I wanted to check whether the reactions to SellAring were as bad (or, in other words, whether the notification bar is considered as personal as the empty auditive space before a call being accepted). They were and that's what I'm able to tell the other developers now based on a foundation of knowledge.

Bad? _Bad_?

The whole concept of SellAring is completely absurd. Why would anybody tolerate a random app putting ads in their phone calls, when the app isn't even running? Whose insane idea was this?

I cannot imagine ever using an app you've written, knowing that you thought this was a good idea. What other absurd bullshit might you put on my phone without my knowledge?

Hi gwillen, well, it seems like there are people tolerating it, otherwise SellAring wouldn't exist. Same for push notification networks like LeadBolt and Airpush. I totally understand that you're mad though, a cellphone is a very private device. Do you personally feel that the notification bar should be considered equally personal as the space before someone picks up?

Best regards Johannes

Bit confused here. You had 1.3m Madvertise requests and 100k Madvertise impressions. Where did the 1.2m requests go (I'm assuming 600k went to AdMob. How about the other 600k)?

Probably just unfilled. Before filling my remaining requests with House Ads, it was not unusual to only have a 50-75% fill rate on Admob. Maybe Madvertise is different (I've never used them) but it could be that no advert is served.

madvertise only fills in some european countries. In case madvertise doesn't fill, I use AdMob. For madvertise I use a 30 seconds refresh rate, for AdMob 60 seconds. This explains the differences.

Best regards Johannes

3D invaders is really fun. You should be selling an ad-free version for 99 cents. (Maybe you are already? I didn't see anything indicating that, however.)

I'm glad you like it. There's no ad free version available at the moment though.

These methods will not work for long, and it's not a basis for a business. The adverts are extremely intrusive and irritating for users.

Thank all you guys for the great discussion!

Reply on news.ycombinator.com