measuring lifetime value of users in different segments

We already covered the how, who and when of calculating LTV (lifetime value). This post is about the Why. It should have probably been the first post as Simon Sinek famous saying “You should start with why”. Anyway, better late than never so here we are.

Companies calculate their user lifetime value as soon as they start ramping up their marketing efforts and especially if they need to raise money in order to do so.

Fundraising and LTV

One big reason to know your LTV is so that you can present to investors and appear on top of your game. Investors normally want you to have a grip of your KPIs since they want to know you are going to be smart about how you will spend their marketing dollars. The basic formula of user acquisition is that you want your CLV (Customer Lifetime Value) to be bigger than your CAC – Customer Acquisition Cost. This is not enough on it’s own but definitily the place to start – the most basic requirement. If you can show that you can consistently have CAC < CLV, investors would want to invest in your company.

Marketing – LTV by segment

The need to do grow via paid marketing is another big reason to start figuring out your LTV. This time you would want to figure out the LTV of specific segments. The basic formula of having the lifetime value exceed the customer acquisition costs still applies here but you should think about it per segment. In other words, you want to be able to calculate the value of each user so that you can find the segments with the high CLV which makes it easier to make the user acquisition formula (CLV > CAC) work. Alternatively you can try to figure out what channels are less expensive and figure out the user lifetime value of these channels and see if the formula works there.

Measureing lifetime value in order to improve it

The last reason to figure out your LTV is that you can improve it. The basic idea behind any analytics and measurement practice is that you can’t improve what you don’t measure. As you consider changes to your app or service you want to be able to A/B test or at least do cohort analysis on users that reached before and after a big change. The parameter you should consider is the user lifetime value – focus your optimization efforts on this parameter and try different things while keeping the changes that increase your LTV. This should help you reach a point where the user acquisition formula works and you can start growing fast.


Feel free to share:
App Monetization, Video

Rewarded video ads means that the user gets in-game currency but the publisher only gets paid on install
Video advertising is hot in the mobile app ecosystem today. Everyone is talking about it and many app developers are doing it. Even more so in mobile games. The rewarded video format especially is becoming popular by the hour and even Facebook and Google are moving into the space according to latest rumors.

What is rewarded video advertising

Video ads can be shown to users in different places inside the app. If the app has video content, the video ads can appear as a pre-roll. In other situations, the ads can appear in a natural breakpoint. A very popular option these days is to allow users to opt-in to the ad by offering something in return – a reward. This practice of giving a something in return for watching an ad is sometimes referred to as value exchange. The user gets some benefit in the app that is normally a premium feature in exchange for the value created for the advertiser by the user watching the ad. In games, the premium feature is mostly in-game currency.

Developers give the reward when video ad is shown

From the developer standpoint, they are giving the reward to the user instead of selling it for money. For example, a game might credit the user 5 precious gems in return for watching a video ad. The ad-network SDK is usually providing a call back function that lets the developer know that the video view is completed and the developer writes a code to give the reward.

Developers don’t get paid when a reward is given

Here is the loophole – the developer might expect to get paid in return for the precious gems he is giving away. However, that’s not happening. Most of the video ad-networks are not paying for video views at all. They are paying a share of the revenue they are making which means the developer is getting paid mostly for installs that the video might be driving. If the video is interesting, if the user likes the app, downloads, installs and opens it. Only then the developers get paid. Not for his users watching the videos and not for giving away his premium features for free.

Transparency and clarity are a big deal

Some app developers might be ok with this loophole since some of the views generate clicks and some of the clicks generate installs and eventually the developers do earn money. However, the fact that many developers are not aware of this is alarming and shows that ad-networks are not being transparent enough about their practices.

Impact on optimization and paid marketing

The most important impact on the game developers is that they lose sight of what revenue is generated by what users. This is critical for in-game optimization, for mediation decisions and for marketing ROI analysis.

If you’re monetizing with rewarded videos it’s important to know the Ad revenue per user. Check out SOOMLA Traceback – Ad LTV as a Service.

Learn More

Feel free to share:
Fun Stuff, Game Design

guess the game quiz with multiple question marks appearing on colorful background
Fun quiz for Friday. Match the avatar to it’s original mobile app and guess the game they belong to.

Lets see how many of these you get right. We dare you to post your result to Facebook and challenge your friends by tagging them.



Feel free to share:
App Monetization, Resource

mobile ad networks logos including: Flurry, Fyber, Vungle, Mopub, AdColony, Revmob, Chartboost, Inmobi, Tapjoy, Millenial, Supersonic, UnityAds, Inneractive, Startapp and Applovin

A few months ago we did a post called top 20 mobile ad networks that became very popular. This post describes each network in words which is very useful. However, we recently realized that app publishers need a better tool for comparing ad networks to each other.

We collected the main features of 18 mobile ad networks into the google doc linked here. There are 5 Spreadsheets: Payout/Revenue, Mediation Support, Reporting API details, ad formats and General information.

Below are the details of each section.

Mobile Ad Networks Payout/Revenue

This section contains details about what you will be getting paid for: the business model of each ad network, the payment terms, payout methods and thresholds. We recommend also reading – 5 secrets of Mobile Ad Networks to gain further insight about this section.

  • Payment Model – this field indicates the commercial commitment made by the ad-networks – what are they actually paying for. Rev-share means that the mobile ad-networks pay a percentage of what they are making. It also means that you are only getting paid if they made any money regardless of the amount of ads you showed your users. Some ad networks indicate the share of the revenue they are paying. For example – if the table indicates 70% it means that the ad-network will be keeping $0.30 of every dollar paid by the advertisers after their deductions (read about deductions below)
  • Campaign Types – this indicates how the mobile ad networks get paid. Since the payment model is mostly rev-share based you need to understand how your ad-provider makes money to understand what you are getting paid for.
    • CPI – the advertiser is paying only after the user downloaded, installed and opened his app
    • CPC – the advertiser is paying when the user clicks
    • CPM – the advertiser is paying when the user sees the ad
    • CPCV – the advertiser is paying when the user finished watching the video ad
  • Minimum Payout – this field indicates the threshold you need to pass in order to get paid. Unpaid balances roll over to the next month.
  • Payment Term – mobile ad-networks don’t pay you immediately but rather wait a certain period of up to 2 months to pay. NET 30 for example means that revenue that was made during March is paid on May 1st – 30 days after the end of the month.
  • Revenue Deductions – when an advertising company promises to pay a percentage of the revenue they mostly refer to “net revenue”. This means that after calculating how much revenue the ad-network made from the impressions related to your site they start deducting certain costs up to 40% of the revenue. “Sole Discretion” means that the ad network may apply a fixed deduction of x%. Let’s assume it’s 20% as an example this means that if your rev-share is 70% and the ad-network made $1, they may first deduct $0.2 to cover their costs and then the $0.8 is shared so you actually get $0.56 (70% of $0.8) for every dollar made. “Consistent” means that the ad-network is providing the exact amount they are deducting. “Limited” means that the provider is only deducting in certain categories that are provided in the notes below the table
  • Links – for each ad network you will find a link to the publisher terms as well as to the payment FAQ page

Mediation Support

This section provides a table of which mobile ad networks supports which ad mediation platform. a Yes in the table indicates that the ad-network in the left column of the selected row is supported by the mediation platform on the top row of the selected column.

Reporting API

This spreadsheet provides the details about the information available for automated reporting. This is important for companies who try to get further insight into their revenue using their internal BI tools. Here are more details about some of the fields:

  • Timezone – this indicates the timezone of the report and is especially important when trying to reconcile discrepancies
  • Data Update Lag – most mobile ad networks don’t make the data available in real time. This fields indicates how long the data is delayed and what’s the update interval
  • Time Granularity – indicates whether the API allows to get the data fields per hour or just per day
  • Other Dimensions – country, app, zone, user, traffic source, etc – a “yes” in this field means that the data can be queried for this dimension. For example, if the provider has a “no” in the “user” column it means that you can’t get the ad revenue or number of clicks per user.
  • Data Fields – these are the fields you can get back in the response. For example a “no” in the “installs” column means that the ad-network will not provide the number of installs

Ad Formats

This section indicates which ad formats are offered by each mobile ad network. For example a “yes” in “video” column means that the network offers video ads.

General Information

In this section you can find general information about the companies: How long they have been in business, where is the headquarters located, what platforms are supported and also links to their website and other useful pages.

Contribution Guidelines

We are interested in contributions for the spreadsheet. If you want to help us evolve this shared resource and keep it updated you will receive multiple karma points and likely to live a long life. Here is how you do it:

  • Clone the spreadsheet to your own google drive
  • Make the edits you want to make – add relevant references in the comments
  • Share the cloned spreadsheet with – make sure you give edit permissions as it allows me to see revision history
  • I’ll notify you about the progress in email back to the email associated with your google drive. If you want to be notified to another email please provide it in the note when you make the sharing request.


If your company uses multiple ad-networks it’s important to know the ad revenue per user. Check out SOOMLA Traceback – Ad LTV as a Service.

Learn More


Feel free to share:
App Monetization, Video


There is nothing more boring in mobile apps than the Remove Ads button. At least that was the case until recently. The least expected place for innovation just re-invented itself. I have seen the option to watch a single video ad to remove ads in one app and then in another. When I saw it the 3rd time I realized there is a trend here and I’m sure others will quickly adopt this new standard.

Watching 1 Video Ad Equals an Ad Free Session

You can see an examples of what it looks like in the games Blocky Football and Tiltagon:

unnamed (2)










The prompt gives the user choice between paying $1.99 to remove ads permanently or watching a video to remove them temporarily.

Similar approach has been implemented by Spotify that offers a 30 minute ad free session in return for watching 1 video ad.












It’s all about user choice

These ads are likely to become more and more popular since they give users choices. Some users don’t mind the ad interruptions while others might prefer to focus their ad experience in a short time and be done with it. It also signifies a more open and transparent approach to ads. Publishers are telling their users “look we need to get paid to continue making games, it’s your choice how to help us”.


If you are interested in accurately measuring the impact of innovative placemtns it’s important to know the Advertising revenue per user. Check out SOOMLA Traceback – Ad LTV as a Service.

Learn More


Feel free to share:
Fun Stuff

Soombot is our mascot. He’s a robot with programming super powers that can transform into many different objects, shapes and people. We keep action figures of all his reincarnations up in the Soombot gallery. Lets see if you can guess the origins of each Soombot.

Feel free to share:
Industry Forecasts, Research

Zynga_s_Financial_ReportsOne of the things I learned a while ago but was recently reminded of by one of our advisors is that you can learn a lot from the financial reports of public companies. Lectures in industry events, blog posts and news articles are great sources of data but you never know the exact context and how accurate the data is. Financial reports are much more reliable.

Zynga’s ad revenue is growing

The motivation to check Zynga’s ad revenues came because in the past 3 months I’m suspecting that there is a growing trend in mobile games to rely more heavily on advertising revenues. Most of the data points I had, however, were based on unreliable sources so I wanted to check with a source I can trust as well – financial reports.

So here are the results. Zynga ad revenue grew from $74M in 2011 to $173M in 2015. 2.3x in 4 years while their in-game revenues went down 40%.

Screen Shot 2016-04-20 at 5.51.47 PM

Moreover, the ratio of the ad revenue from the overall revenue is also growing dramatically from 7% to 23% in just 4 years. In their last reports Zynga released information that their Q4 and Q1 ad revenues accounted for 30% and 26% of their total revenues. Their Q1/16 ad revenue grew 41% compared to the Q1 revenues in 2015.

Screen Shot 2016-04-20 at 5.54.29 PM

What’s driving the trend towards advertising based monetization

There are many drivers for this trend. Here are a few that I’ve spotted:

  • Brand budgets are starting to go mobile.
  • Video ads bring high eCPMs from brand campaigns as well as app trailers.
  • App publishers are learning to integrate more ads in non-intrusive ways and are feeling more comfortable with increased ad exposure

The first driver is the most important one that will lead future growth in the next years. There are forecasts by eMarketer showing that mobile advertising will grow 3x in the next 4 years.


If your company also makes significant ad revenues it’s important to know the Advertising revenue per user. Check out SOOMLA Traceback – Ad LTV as a Service.

Learn More

Feel free to share:
Industry Forecasts, Marketing

what's the average spend per app on cost per install campaignsAbout 2 weeks ago a friend asked me how much do app companies spend on getting app installs – what’s the average cost per install campaign size. He was pretty happy with the answer I gave him and the research I referred him to so I thought I would share the answer.

Concentration in cost per install campaignscost per install campaigns account for growing part of the total mobile ad spend and forecasted to reach $6.8B in 2019

First, you have to realize that the average spend is really irrelevant here due to strong concentration in the market. According to eMarketer there were $4.6 billions spent on App Install campaigns in the US during 2015. You can assume that the numbers are 2x if you look at the global market so $9.2B. The reality is that the top companies are doing a big chunk of that. You can look at the public companies to get a sense of that:

  • Zynga spent $169M on Marketing in 2015
  • King spent $344M in 2015
  • Glu’s marketing spend was $45M in 2014

The rumor is that Machine Zone and Supercell are spending close to $500M each on cost per install campaigns. The real answer is that the distribution is estimated as:

  • Top 3 companies: $1.2B ($400M each)
  • #4 – #10: $1.2B ($150M each)
  • #11-#50: $4B ($100M each)
  • #51-#100: $2B ($50M each)
  • #101-#200: $0.75B ($15M each)

The average spend per app is around $4,600

If you still want to figure out the average spend you would need to divide the total campaign spend by the number of apps. Surely enough Apple and Google can provide this data. The latest numbers that were released were 1.5M apps for Apple and 1.4M on Google Play – Obviously there is a lot of overlap so we can assume 2M apps in total and an average spend of $4,600 per app.




Feel free to share:
App Monetization, Resource


This is the 3rd part in a series of 6 posts. Here is a summary post explaining all the different calculators.

The methods explained in previous posts assume you are starting with a fresh new app. The life-time modeling method used in posts #2 and #5 for example assumes you only have data on a limited number of days. The method described here is for estimating marketing LTV. Typically when you start marketing your app you already have 6 months of data in one segment (typically on organic traffic) and you are trying to estimate the CLV in a new segment.

Calculator Details


  • Training Data from existing segment:
    • ARPU for existing segment users – day-1 to day-7
    • LTV for existing segment users – 180 days
  • New Segment data:
    • ARPU for new segment users – day-1 to day-7


  • Customer Lifetime Value for the new segment (Marketing LTV) presented as a number and in a Gauge


This model assumes that the ratio between the day-7 revenue of segment-A and the day-7 revenue in segment-B will reflect the ratio in the LTV. For example if you have a new traffic source that has $0.5 ARPU in the first 7 days and you are normally seeing $1 in the first 7 days then the LTV from the new source will be half your normal LTV. This is very intuitive and is actually supported by much more advanced models. The calculation therefore has 2 steps:

  • Figure out the ratio between the day-7 revenue numbers
  • Apply the same ratio to the LTV


  • Simple
  • One of the most accurate models around


  • Requires 180 days of data from existing segments

More methods to calculate marketing ltv

Here is a similar method that can be applied even if you have partial data from the existing segment

If you have time and a spreadsheet you can also fully model out the revenue function


Calculating LTV for ad-supported apps can be specifically difficult due to reporting limitations by the ad-networks. Check out SOOMLA Traceback – Ad LTV as a Service.

Learn More

Feel free to share:


We have a difficult announcement to make. Beginning today we’re winding down the GROW service.

Over the past 2 years we had a huge vision to create a community of all the game developers in the world and allow them to share analytics and insights. We envisioned a shared database that will allow the community members access to the same data tools as the “big guys”.

The GROW network has been growing quickly and we achieved every growth milestone we have set for ourselves. However, we came to realize that the product just doesn’t solve a big-enough problem for its customers. As a company that always strives to generate value for its customers, we’ve decided to shut down GROW to make room for new products that solve bigger problems.

GROW will be shut down on May 31st 2016.

This includes Whales Report, God Mode Analytics, Basic Analytics, The Sync Service, Highway and the Receipt Validation Service (Fraud Protection). [CORRECTION: Fraud Protection / Receipt Validation will keep running until July 31st 2016 – at which point it will be included as part of the new product]

As we are doing the transition it is very important for us to make sure we are giving you enough notice to find other solutions and to make sure the negative impact is minimal.

You can find answers to most of your questions below. If you have additional questions, feel free to email us to


Q: What about SOOMLA Store and Profile?
A: The open source projects will still be available on Github for the community to continue to use and maintain them. In addition, the paid packages in the Unity Asset Store will stay there for the benefit of the community. There are a few power users in the community that agreed to stay as moderators and take over the support for the projects.

Q: What do I need to do to minimize impact on my game users?
A: We recommend removing the Highway SDK from your game before May 31st. SOOMLA will not be responsible for any production issues in games that have the Highway SDK after May 31st.

Q: I use GROW Ultimate. how can I switch to SOOMLA Store?
A: If you’re using one of the GROW bundles you will notice in your game project that they are made from separate packages. Removing the Highway SDK will allow SOOMLA Store and Profile to work with no issues. 

Q: What about the Receipt Verification / Fraud Protection Service?
A: The receipt verification service will stop functioning as a stand alone service on May 31st. [CORRECTION: Fraud Protection / Receipt Validation will keep running until July 31st 2016 – at which point it will be included as part of the new product]

We are planning to include it as part of the new product during June. If you are using the existing service you will need to open an account and integrate the new service when it’s available. Fees for the new product will apply.

Q: Is SOOMLA Shutting Down?
A: Not at all. We have a strong team and investors who believe in the new vision of the company. We are still planning to change the world but in a different way this time.

Q: What is the new product about?
A: The new product is going to be called SOOMLA Traceback. We can’t tell you what it is but we will announce it very soon. If your game monetizes mainly with ads stay tuned for more details.

Feel free to share:

Join 14511 other smart people who get email updates for free!

We don't spam!

Unsubscribe any time



SOOMLA - An In-app Purchase Store and Virtual Goods Economy Solution for Mobile Game Developers of Free to Play Games