pay for ad-mob advertising

So I’ve wondered if paying for banner ads with ad-mob would get you a decent amount of installs. I just released a new app so decided to try it to get a jump start on “seed” user base or maybe even get into the top new apps category. I’ve heard of people spending $50 or so and getting a few installs, but what if you put a decent amount of cash into it? So with my new app I made an admob campaign and spend $360 dollars (not a lot, but more than $50).

Here is my campaign info:

  • $360 ad balance
  • targeted all android devices, android 2.1 up, North America and Western Europe, Carrier traffic, no demographic targeting.
  • two ads, a text ad with app icon for tile image, and normal banner ad with image.
  • advertising a free app.

Here is the results:

4,136,739 impressions
36,020 clicks
350 app installs.

An abysmal failure I would say. 350 installs is nothing for $360 dollars. How could I get only 350 installs with 36k clicks on my ad. What this means about ad-mob either:
a) massive click fraud
b) massive accidental clicks
c) my app or app description so bad no one would install it.

I tend to think its a or b not that my app is that bad. One major point to consider here though, I think, is the price I paid per click .01 maybe if I had payed more I would have gotten clicks that lead to installs? I would like to try the experiment again with a default bid around 0.30 and see if I got any installs. Still though at .30 per click I would only generate 999 clicks for $300 so most users I would get would be 900 more likely to be around 100 even with good click through to install ratio. I wonder if there is a point like around 5 - 10 cents that you could get a lot of clicks that led to installs?

Anybody else has experience with paying for advertising?

Hi XdebugX,
have you tried tapfortap?
You get 1000 impressions for free and if you integrate their SDK into your app every impression you show on your app earn a point to display your ad into another app.

At least you can try because you don’t need to pay anything.


Yes, I’ve signed up with tapfortap but haven’t integrated them into any of my apps yet. I also signed up for applifier and have integrated them into one of my games. Still early for applifier but not noticing any increase yet though.

I’m thinking though if the click through to install rate for free cross promotion is anything like ad-mob it won’t help much. It took over 4 million impressions to get 300 installs with ad-mob.

Maybe it was that the new app I advertised is kind of a niche market (it’s a stop alarm app for people who ride trains or buses). But then again, why would they click an ad for “Stop Alarm” if they didn’t want to install it, its a free app. My ads were not misleading. I think theres just too much ad fraud or accidental clicks.

Hi XdebugX,
it looks more fraud than accidental clicks. Anyway your app looks really niche market app and probably your need more control where your Ad is displayed. :frowning:


It would help if you could show us what your creatives looked like and what app they advertised.

That being said, I know that the pay-per-install crew is paying $1.40 per install for some apps. So your cost of $1/install is not bad!

Your big problem is not being confident you could monetize the installs at > $1 return on each install :slight_smile: If you could, then for every $1 you spend, you get > $1 return… it is like printing money … Angry Birds style! :slight_smile:

Even if you can’t monetize each install at >$1, if you have a very large budget you can still make money from paying $1 per install. That is if you can pay for enough installs to get you into the top of the new section. Then you also get alot of free installs that wouldn’t have happened if you didn’t pay for installs. But the problem is indie developers don’t have enough of a budget to force an app to the top.

I’m convinced that on a small scale it is hard to make banner advertising for apps pay off.

Here is the banner and text ad images:

And tile

With the text “Stop Alarm, Never miss your stop!”

Here is the market page for the app, which was where it took them when they clicked on the ad. Was targeted for android devices, so opened in the market app. Stop Alarm

I could see paying $1 per install if it was a paid app. But it’s free, why would anybody click the ad then not install it? And with a free app it’s almost impossible to make much money with 350 users. Making less than 0.40 a day with ads, will take me 2 years to make my money back at this rate lol.

But I do agree with that, if it gets your app into a top new free list than it would be worth it. Which is one of my goals with the campaign. Which has worked sort of, in the play market, it is in position 38 of top new free for transportation category. This is because transportation is such a weak category, theres not many good apps in it.

I recently tried a bunch of impressions with house ads on Robo Miner, just to push my other games a little bit up. Not so much impressions, it was just 700 imps/day with a ctr of 4% so ~30 clicks per day.

Guess what -> Downloads does not really change at all. Minesweeper: Unlimited! still has only around 5-15 installs a day.

Because the CTR is much higher than on regular ads there is no fraud and no accidental clicks involved, but as you can see, there is still not the expected result.

That means that the problem is not the ad but something AFTER the ad which does not convert the clicks properly

Maybe I need to overwork my marketing graphics alltogether :slight_smile:

What I want to say is: Its not always fraud or accidentals … admob would have to shut their doors if they have that much fraud clicks nooone would use them to advertise their products.

Good test to verify that is to use one of the pay per install offers. With rather small investment you can run a campaign for a while and see how the clicks/installs ratio looks there.
I used to try AppBrain with my game and the ratio of people coming to the app page to downloads was around 3/1 so pretty decent. I even managed to improve how the app page converted using this ratio as a feedback.

One thing to note is that in such cases people clicking through are actually looking for software to install vs people who are just curious enough to click the banner without a real need for a new game/app.

Your CTR seems ok (0.87% for a banner advert!). The real question is … why are people not installing ?

Some more data:

In Oct 2011, I tried paid installs on to jump-start WordHero.

  • 1761 clicks
  • 1300 actually loaded the Google app store (some people backed out before even seeing the click result)
  • 403 installs

So 22% of clicks resulted in an actual install (highly targeted advertising, with graphics matching what they would see on the Google Play store). I think this is reasonable.

You can work out cost-per-install as CPC/0.22…

I would suggest that what you do is spend in $50 increments and do some A/B testing.

Like so:

  • Make 2 adverts

  • Run each advert with minimum advertising ($50?) to get some data

  • one with higher CTR is better advert

  • Repeat with a 3rd advert

  • Now use your best advert and

  • Change the landing page & text (keep a record!)

  • Change only one thing at a time: eg: text or header graphics each day (at the same time!)

  • See how this changes your install numbers

Basically, you are creating a controlled feedback loop to improve your Play presence and increase the number of installs. The obviously will improve your organic install rate too!

I had about the same success running an admob campaign. One thing that I did notice was that i’d set my budget for the minimum, just $10 and that money would be eaten up in under a minute. Just 60 seconds and there were 1000 clicks. How many installs did I get from that? Less than 5. Seemed rather suspicious to me. Then a friend of mine told me to remove Russia and China from the list of countries, said he heard that’s where most of the click fraud was taking place. I have a spelling game so I removed every country except North America, UK and Australia. Then it took a full 24 hours to run through 1000 clicks and i got 80 installs.

I had targeted North America, Australia, New Zealand and Western Europe, thinking I would cut down on click fraud. There must be a few countries in Western Europe with a lot of click fraud though. If I try the experiment again I will only target N. America, Australia and the UK.

It can still be hit and miss. I personally dont think it was worth it. The number of downloads you need to hit to make up the money you spent is pretty high. I’d wager you probably need 1000 downloads to make 50 bucks
Someone mentioned tapfortap earlier. I use them, I think it’s been worth it so far. I was losing users until I installed it and now I gain everyday. I don’t gain a lot but definitely better than before. They have a 1 to 1 earn/spend ratio and they give a lot of free credits. I got an email from them today saying how they’re going to get something like 5000 free credits every week. In the last 2 months since I started using it I’ve gotten about 50,000 free credits from them. And every time I get a bunch of credits I see a small spike in downloads. It’s a great ad to put on your menu screens, the places u don’t want to put an admob ad cause the user’s never there for more than a few seconds.

Anyway, that’s my 2 cents.
It’s also a great way to find out how your performing in real time. Cause the credits earned that is displayed on their sites is live, you don’t have to wait the random amount of time that most ad providers have between updates.

If 80 people installed your app through an advertisement and they LIKE your app … they will show it to their 100 friends (if they got friends) and that friends also have friends (maybe).

At the end it is all up to your app :wink: You just need that 1 user who has 1000 friends :slight_smile:

If I had gotten 80 installs per $10 that would have been 2880 installs, I could live with that. But 350 installs is awful. I wonder if ad-mob refunds advertisers when they filter click fraud. Sounds pretty unfair to me if they don’t.

How can you proof that there was clickfraud? You simply can’t. Go ahead and blame people how DO clickfraud or placing their banners to force false-clicks to make money :wink:

Go and play Fruit Ninja - the free version - you will notice, that after a game the ad comes up with a small close button underneath … tons of false-clicks there… because you are tapping wild on the screen to get the fruits and if you fail the ad comes up immediatly while you are still tapping wild around - false click :wink:

you see :wink:

Beside that, the conversation rate is low - that’s what it is, everywhere with nearly no exception.

They dont refund anyone.
But hey right, what youre hoping for with an ad campaign isnt to get a million installs. Because with a few thousand clicks that just wont happen. What u want is to get the ball rolling, have word of mouth take it further. I wish there was a better way to get visibility. But without a “whats new” or “what’s being played right now” list. Our options are limited.

Can someone please explain to me who benefits from click-fraud OTHER than Admob ?

Person paying for adverts - nope
Person displaying adverts - against Admob TOS

Even so, do they fake a device connection, pull the advert and then ‘click’ it ? Thoughts/comments ?

They do detection (I’ve heard), but I not unless it’s clearly rampant. Go ahead and launch your game and click on a couple ads. You’re cpm for that day will raise, and you’ll be owed a few extra cents. I suspect that they only target the people that do this hundreds of times.

admob has no benefit of clickfraud, because admob aims for a long term income and that depends on a good reputation and (obviously) advertisers.

Everyone knows, that it is not so hard to get kicked by admob, they wouldn’t do that if they aim for clickfraud…