What’s the most important thing when it comes to buying app installs?

Cost-Per-Install (CPI), right?

You’ve probably read lots about how to lower your Facebook ad CPI. Or how someone managed to get super cheap app installs by targeting Asian or European countries.

It seems like an endless quest to achieve the lowest cost per install possible. But is a lower CPI always better?

In this post, I’m going to use a case study to show you why you need to focus on more than just cost per install for paid user acquisition. Keep reading to find out how to get the best value for money from your paid user acquisition budget.


Why a Low Cost Per Install Isn’t Only Metric That Matters

Using cost per install as your only metric for measuring the success of an app install campaign doesn’t give you the whole picture.

By only looking at CPI you are missing one important thing. That is the value of the users you are buying!

You see, cost & value are two very different things. Cost is simply how much you paid for something. Value is how much your purchase is actually worth.

Is buying low-cost app installs really a good thing if the value of those installs is basically zero?

The constant obsession with getting the cheapest installs possible with no regard to their value reminds me of those spam messages I get on Twitter every day.

You know the ones. They usually look like this: “Buy 10000 followers for $20” or “Get 20000 real Twitter followers for just $29”. I completely ignore them, and I bet you do too. That’s because we both know that the users will never read your Tweets, let alone interact with them. That makes them absolutely worthless.

Most of the time when it comes to mobile app installs, value is the amount of revenue the new users generate. But this isn’t the only way that a user can generate value.

You might want to buy installs purely to collect usage data, check for bugs, build a critical mass of users etc. That means each install you buy is generating value in a form other than money. What I’m saying here is that there is a time & place for getting cheap installs regardless of the monetary value of those users. You just need to understand when that is.

In the case study I’m going to show you, we will assume that the goal of the ad campaign was to generate revenue. That means that we are measuring the value of the installs purchased based on the revenue they generated.


Measuring The Value of Paid App Installs

Measuring the return on investment for your mobile app install campaigns is relatively simple as long as you are prepared.

What you need is the ability to track mobile app ad attribution. This is the process of attributing users to particular advertising campaigns. This allows you to then measure the performance of each campaign.

Analytics tools like Flurry & Google Analytics are great, but they generally lack ad attribution features. Flurry has some ability to do this using custom URL’s, however it is limited.

If you are running Facebook App Install ad campaigns you can’t use Flurry or Google Analytics for ad attribution. You need to use another tool. The reason for this is because it’s not possible to enter a custom URL for Facebook App Install Ad campaigns. Facebook automatically get the iTunes URL for your app.

Three of the most popular mobile ad attribution tools out there that are also Facebook marketing partners are Localytics, Appsflyer & Adjust.

I won’t go into detail about each of these solutions as they all will do the job. Each has a different pricing structure & feature set, so I suggest selecting the one that suits your individual needs best.

Once you have an ad attribution solution in place you can separate users into segments based on the ad campaign that they came in through. You can then measure the life time value (LTV) of each segment.


The Case Study – How Much Value Did Cheap App Installs Deliver?

Let’s dig into the case study & see how much value the low cost installs delivered in revenue.

Because it’s not possible to accurately track which campaign delivered advertising revenue, I am only including in-app purchase revenue.

Just like in my previous post about how high value Facebook App Installs generated 4x more revenue than organic installs, I ran the advertising campaign for 1 week & measured the 2 week life time value (LTV) of users acquired through that campaign.

cost per install facebook ads


Here are the numbers for the campaign, including cost per install:


    • In this case I purchased 376 app installs at a total cost of $109.
    • That gave me an average cost per install of about $0.29.
    • Installs came from Mexico, Malaysia, Chile, Italy, Philippines & Romania.


And here are the revenue numbers over a 2 week period for those users:


    • Total in-app purchase sales: 3
    • Total IAP sales revenue: $25.49

itunes connect cost per install roi


Ok, so we can see already that the results weren’t great. Let’s look at what these numbers mean in a bit more detail:


    • The average revenue per user was just $0.068. While I paid $0.29 on average for those users. That leaves a $0.22 loss for every install I purchased.
    • Approximately 0.8% of users actually became buyers. Or another way of looking at this figure is on average each user made 0.008 in-app purchases.
    • Average spend per paying user was quite high at about $8.50, however I question whether having only 3 purchases is enough to provide a relevant estimate here.



I hope these results help to shed some light on just how important it is to understand the value of the users you are acquiring & not just how much you pay for them.

Without an understanding of value, you might think that buying installs for $0.29 each is great. But clearly that isn’t the case here. I’m not in a hurry to go and buy more of these installs with the hope of achieving a positive return on my advertising investment.

It’s time app developers shifted focus to increasing the value of installs, while minimising cost per install.

What’s your primary focus when it comes to buying installs for your apps? Are you looking at the value of the users you are buying, and if so, what are you seeing? Go ahead and share in the comments!


  • GetSocial

    Great article! The average CPI is outrageous, isn’t it?! This is one of the biggest problems we’re solving right now too via smarter invite tools to encourage more “word of mouth”.