Growth is a loop not a funnel. Follow these four steps to scale your game and take it to the top
Traditionally, user acquisition teams and ad monetization teams ran in separate circles and often even worked apart. The former was responsible for bringing in users from various channels, while the latter was responsible for making sure those users purchased a sword or watched an ad. For the most part, communication ended there.
But now that ad monetization is becoming an increasingly important revenue stream, gaming companies are beginning to unify their monetization and user acquisition teams, even assigning one point of contact to oversee both sides of the business - ultimately they are both about driving game growth. Doing so, of course, makes it much easier to make one side of the business work for the other.
In working closely with both sides of the coin, ironSource understands that growth is not a funnel that starts with one and ends with the other, but a loop powered by user acquisition and monetization together. That’s why we’ve unified our own monetization and user acquisition teams, and how we came up with the idea of the Growth Loop.
Think of it this way. To optimize their buying strategy, a user acquisition team wants to know which users are generating high amounts of revenue and why - since it influences which supply sources they’ll go after and how they’ll tweak their creatives. So they ask the monetization team. Meanwhile, the monetization team knows that to maximize revenue they need to acquire high quality users who will generate even more money. So they pass back information to the UA team. And so the cycle goes.
Don’t just use this eBook for best practices, also try to understand how your counterparts approach growth, and conversely get them to understand your business needs, so as a company you can take your game to the top of the charts. Your goal should be to find common motivations so you can optimize the growth process. And the first step in building that relationship is getting to know the Growth Loop. Let’s begin!
Step 1
The Growth Loop starts in an unexpected place. Usually, developers kick off growth by perfecting monetization - confident that once ARPU is high, growth will follow. However, that’s not always the case. As we’ll learn, just because games monetize well, doesn’t mean it will definitely garner the scale to make that monetization meaningful. That’s why before anything, you need to assess the marketability of your game, and understand if players actually want to download it. Small UA tests are a great way to measure this.
Thus, the true goal of Stage 1 of the Growth Loop is to determine the marketability of your game and test user acquisition. There are many methods and techniques for doing so, which we’ll touch on in the following sections. Read on for user acquisition best practices.
Acquiring users organically: A bit about app store optimization
It’s notoriously difficult to get organically discovered in the app stores. There are millions of apps competing for share of voice on the top lists. Before investing in UA, how do you get users to organically download your app?
You turn to ASO, or app store optimization - a marketing process dedicated to boosting app visibility in the stores, which focuses on keyword optimization, conversion optimization, app store page management, backlinking, localization, and review and rating management.
A well-thought out ASO strategy can help land your app a place on the top (relevant) charts, as well as making sure they appear in relevant search queries - without spending any budget.
We’ll never really know how the App Store and Play Store algorithms rank apps. But the black box is not too much of a roadblock, as Apple and Google both provide pretty in-depth guidelines for constructing an app store page - including the do's and don’ts of icon design, screenshots and videos, and descriptions.
Even if you have big budgets, and aren’t too interested in ASO, it’s important to make sure your app store page is optimized for conversion. Why? Users who land on your store page from an app-install ad will want to see that what you have to say matches the ad that just convinced them to click through. Going through the ASO checklist is a good way to make sure your store page is as high-converting as possible.
There’s a lot that goes into forming a paid user acquisition strategy. You need to set your KPIs, decide which ad units to utilize, which supply sources to run on, which pricing models to use, and then which network partner to choose. Where do you start?
Focus on creatives to crack Cost-effective UA at scale
The first step to a great UA campaign is nailing the creatives, since it’s a key factor in being able to scale cost-effectively. Let us explain.
In a network waterfall, campaigns are ranked by eCPM - meaning how much money that campaign is generating for publishers. The campaigns with the highest eCPMs will be the ones to top the waterfall and ultimately gain a large share of voice (and ad impressions) on the network.
How do you make sure your eCPM is as high as possible? You attempt to influence two factors: CPI and IPM. CPI, or your bid, is notoriously difficult to change, as it largely depends on cash flow.
Luckily, IPM is somewhat easier. IPM, which is the installs generated per one thousand impressions, measures the performance of the creative. Basically, the better the creative converts, the higher the IPM. And if your IPM is very high, you can afford to lower your CPI and ultimately scale cost-effectively.
That said, let’s start by going through each ad unit and creative and seeing what they each bring to the table. Ad units in this case are the containers, which hold the creatives or formats. In other words, creatives can be placed into any ad unit.
Creatives
Video ads - Generally, video ads are the simplest creatives to make, and therefore are the most common. There are a few ways developers can use video ads: One strategy is to record 15 or 30 seconds of gameplay (the most exciting 30 seconds, of course), as a way to show users what the game is all about. Another strategy is to depict an actual person playing the game, filming them over the shoulder, as they’re completing a meaningful challenge.
Playable ads - Playable ads offer users a snippet of interactive gameplay. Think of them as “micro-games.” They’re perhaps the most user-friendly and enjoyable ad experience to date. It’s one of the reasons why playable ads perform so well.
In addition to generating high IPMs, they can also drive high-retention users who stick around, since playables do a great job of showing users exactly what they can expect after downloading the game. After all, users get to literally “try out” a level of the game and see if they enjoy it or not. By increasing the potential quality and relevancy of the users, playable ads can reduce app uninstall rates and improve retention rates down the line.
This process can end up saving advertisers large sums of money, since it weeds out users who wouldn’t enjoy the app, minimizing potentially wasted spend on acquiring irrelevant users. As such, even if the CPI for playable ads is higher than a video ad, the ROI can be greater.
Interactive end cards - Interactive end cards add an additional layer of engagement at the end of a regular 15-30 second video ad, which gives users a feel for the app’s functionalities.
They’re usually slapped onto the end of video ads, especially interstitials, to add a little spice and interactivity to a somewhat common ad unit. Think of it like a mini playable ad.
Interactive video ads - Interactive video ads let users choose their own in-ad adventure. The ad plays a few seconds of video, stops to ask the user what they’d do next, plays another video based on that choice, and so on.
It’s almost a hybrid between video ads and playable ads, which is why it reaps the benefits of both. Specifically, they’re highly enjoyable, tell a good story, and show the user exactly what the app will be like - the three ingredients of an impactful ad.
These ads tend to work best for slots games and simulation games, since they require a simpler user interface than other creatives.
Ad formats
Rewarded ads - Rewarded ads offer users a clear value exchange - users engage with an ad and get in-app rewards (like in-app currency, premium content, etc) in return. Generally, advertisers prefer running their campaigns with rewarded ads as it boosts both visibility and performance.
Because users must click to opt-in to watch the video ad or play with the playable, they’re a guaranteed view for advertisers. And in an industry in which viewability remains a growing problem, it’s useful to know that your ad is only shown to users who actively choose to see it.
Users have a high motivation to engage with a rewarded ad and are less likely to find it intrusive. As a result, they’ll remain positive during the duration of the advertisement and respond better to messaging - leading to increased IPM.
Offerwalls - Also rewarded, offerwalls act like a mini-store inside an app, listing multiple “offers” that users can complete in exchange for receiving an in-app reward. For example, users can receive extra in-app coins if they install an app and finish the tutorial, get to level 5 in a game, start a free trial, complete a first purchase, etc.
Advertisers buy offerwall inventory on a CPE basis. CPE, or cost-per-engagement, is a pricing model in which advertisers choose a post-install event as a goal, and only pay for the users who reach and engage in that specific app event. It’s best practice to set the event relatively deep in the app. The deeper the event, the higher-quality the users you pay for will be, although of course it may limit scale. In other words, you won’t be paying for users who just install your app and leave - only users who complete a certain event in your game, that you set ahead of time. Setting the right event by considering LTV allows you to ensure that you acquire users who will generate enough revenue down the line to be ROI positive for the price you acquired them for.
Interstitials - Interstitial ads are full-screen ad units which either display a video, static image, or playable, offering the user an option to exit or skip after several seconds. They’re a great ad unit to quickly inform users about your app while you have their undivided attention, since there’s nowhere else to go for at least a little while.
Additionally, some apps such as hyper-casual games focus the majority of their ad monetization strategy on interstitial advertising. You wouldn’t want to miss out on major supply sources by ignoring interstitial ads.
Banners - Banner ads are rectangular ad units typically placed at the bottom or the top of an app’s layout, remaining in place for the duration of the session.
They get a bad rep - since IPMs tend to be low, fat-finger syndrome is usually the culprit for high CTRs and low CVRs, and of course, banner blindness. But for all the disadvantages, there are some benefits. First off, they’re great for increasing the ‘brand awareness’ of your app. In other words, the more users see them, the more they’ll start recognizing your app - which may prompt them to install from a more engaging ad unit later on. Not to mention, they’re relatively inexpensive in comparison to other formats.
Once you’ve got a few creatives in place, it’s time to check the market. Do users even want to install your app? It’s important to get this answer before you spend budget on a lengthy user acquisition campaign. While organic users will help you understand if there is a market for your app, they won’t necessarily give you the whole picture on the full market potential for your game.
Too often, developers base their go/no-go evaluation of their game on monetization performance, assuming that if ARPU is good, then user acquisition campaigns will be good, too. But in fact, this assessment needs to happen before spending resources on cracking monetization. Why? The simple answer is, the best ARPU in the world won’t make a difference if no one wants to play your game. The more complex answer involves going back to eCPM, which we discussed earlier.
In some cases, a low ARPU (and therefore in most cases a low CPI) game may actually generate a higher eCPM than a high ARPU game, simply because the low ARPU game has a much higher IPM (i.e. can generate more installs per thousand impressions served). Most developers would nix a low ARPU game without factoring in the market response, where scale could make up for lower overall ARPU per user.
A game with low ARPU which generates a higher IPM is one we know the market prefers - ARPU aside, users are installing at a much higher rate. Because the IPM is so high, the eCPM jumps too, securing the game a high position in the network waterfall, and garnering more impressions and an even higher volume of installs - facilitating a virtuous cycle of growth. In the end, the low ARPU game generates more revenue overall because of scale, proving to be the better game business.
By not assessing marketability at the first stage, developers run the risk of killing a potentially high-performing game too early, or alternatively, heavily investing in trying to promote a game the market isn’t interested in.
If you know your game is marketable and have a handful of varying creative concepts to run with, you’re ready for the creative soft launch.
What is it? The creative soft launch is a process to determine which creative delivers the highest IPM or highest quality users (depending on the KPIs), which will then be used during your global launch. Think of it as a practice run for your creatives.
Essentially, you’ll test various creatives in a few markets similar to your target market (if your official launch will be in the US, you would typically soft launch in Canada and the Netherlands, where users behave like users in the US), and continuously eliminate low-performing creatives from the campaign. You can then further optimize the creatives that do well, squeezing even better performance out of each one. By the end of the creative soft launch process, you’ll be left with one or two stellar creatives - which you’ll deploy when launching your game worldwide.
That brings us to the second stage of the Growth Loop, monetizing the users you’re beginning to acquire.
Step 2
Of course, designing the monetization strategy should happen naturally during the development stages, as IAPs and ads revolve around the in- game economy. But now that users are trickling in, either organically or via the paid campaigns you set up, you’ll have a better idea of how to segment, price your IAPs, determine reward amounts, and more. Your goal in this stage is to maximize ARPU, or average revenue per user, while balancing the user experience.
Tips for in-app purchases
Traditionally, freemium mobile games have relied on in-app purchases (IAPs) as their main revenue stream - and although it’s becoming more popular to mix IAPs with other revenue drivers like in-app ads, IAPs continue to be a huge form of income for many game developers. Read on for some best practices for implementing IAPs.
Content
Think about the types of IAPs your users may want to purchase. Go beyond the basic IAPs, which often don’t offer huge value, such as skins or an option to disable ads, and offer items that will actually enhance gameplay. These are items which can be integrated into your core game loop, and encourage dedicated users to go deeper and deeper into your game.
That said, be careful of offering too many paid upgrades. It may be okay for hardcore and midcore gamers, who understand the value, but casual gamers tend to take more convincing and education. You can remedy this by offering rewards just because. For example, after progressing to the next level, reward your users with a few extra coins - a bundle which can also be purchased. This way, you’re educating casual spenders what an IAP is worth (in this case, a level progression).
Pricing
Next comes determining the prices for your IAPs. About 2% to 5% of users make IAPs, so it’s best to offer a range of prices to catch as many potential buyers as possible. The wider the price distribution, the more likely you’ll attract each type of buyer - which is usually broken down into whales (a small minority which contribute to the majority of your IAP revenue), dolphins (mid spenders), and minnows (light spenders, but the majority).
You may be surprised, but some users may spend as much as $99 on a single IAP. For the rest, offer IAPs starting at $1 and go up from there.
Live-ops
First, let’s define it. What is live-ops? Basically, it’s any change added to your game that doesn’t require releasing a new version to the app stores. It can take many forms such as: promotions, events, challenges, and random rewards. It’s best to keep live-ops in mind during development so that you don’t need to completely overhaul entire systems to support it.
Be sure to create a live-ops calendar each month so you can stay ahead of promotions. Include KPIs, too. Is your goal to increase retention or revenue? If it’s revenue, for example, you can run a limited-time IAP sale for an upcoming holiday - especially if data shows certain IAPs aren’t performing well and you want to give them extra attention.
If it’s retention you’re after, you can run a special challenge for your players, asking them to complete a task and then rewarding the winners with a rare item. Still, this challenge could be monetized, too. How? Make the challenge only accessible to users who are properly equipped, which will nudge them to make the corresponding IAP.
Segmentation
Each user behaves and spends differently. Your monetization strategy should reflect that. Use segmentation to divide your user base into various personas and tailor monetization to each. This way, you’ll be able to create content according to what each segment cares about most, and personalize experiences to encourage each segment to purchase. The goal here is to convert non-paying users or increase revenue from existing payers.
You can segment by purchase frequency, in-game progress, payment tiers, etc. For example, offer players who spend an average of $10 two options: a bonus IAP that’s worth $5 and another that’s worth $15. Usually, they’ll go for the $15 option since it has a higher value proposition.
As you likely know, IAPs are just half (or more often, 5%) of the story. In some cases, ad monetization is driving even more revenue than IAPs - especially for hyper-casual games which tend to rely solely on ad monetization. Read on for best practices. Ad units or placements can be divided into two main categories: system and user-initiated ads. The best developers will craft a strategy that contains the right mix of ad placements in order to drive maximum revenue.
User-initiated ads
User-initiated ads are ads that users opt-into, which provide some sort of value exchange - i.e. users choose to engage with an ad and receive in-game content as a reward. Think of it as an engagement tool designed into a feature within the game.
There are two types: rewarded video ads and offerwall ads.
Rewarded videos, which are video ads users watch in exchange for receiving a reward, are integrated right into the core game loop. For example, they can be served after a level fail to offer users a second chance if they engage, they can be served daily on game launch to encourage users to play every day, served as a boost when players are low on juice or energy, as a hint when they’re stuck, etc.
Offerwalls, which function as mini in-game stores that list out tasks users can complete in exchange for a reward, are generally placed on the main screen or the in-app store where you display your IAPs. They’re not in the game loop because these tasks generally take several minutes to complete, and would be too disruptive to prompt in the middle of gameplay.
Being opt-in, user-initiated ads are credited with increasing revenue without negatively impacting user experience. In fact, if implemented correctly, rewarded video and offerwall can actually boost retention and session length, as users come to rely on the rewards they receive to progress further in the game.
The only thing to look out for is making sure rewarded video ads don’t cannibalize your IAPs. To avoid this, make sure you implement rewarded video as a part of the core loop features such as revive, boost, double winnings, hints, lives, etc.
Additionally, it’s good practice to make the rewards exponential, so that if a user receives 2 coins for watching a video on level 10, they receive 200 coins on level 50.
System-initiated ads
In contrast to user-initiated ads, system-initiated ads are not opt-in, and are served whenever you, the developer, decides. These include interstitial ads and banners.
Interstitials are full-screen ads, either displaying a static image or a video, or even a playable ad, which users can x-out of after several seconds. They’re usually placed at natural stopping points in the game, like in between game levels, after a certain number of actions, or before the “next” button in a series of screens.
Banners are rectangular ad units that sit on the top or bottom of the screen throughout the user session. The ads themselves can refresh within the rectangular container.
In general, system-initiated ads are great for increasing ARPDAU and monetizing almost every kind of user - or at least the users who don’t choose to make IAPs or engage with user-initiated ads. In that way, they offer a steady incremental revenue stream.
Over time, however, too many system-initiated impressions can negatively impact retention - which is why we recommend serving a balance of both user-initiated and system-initiated ads. For IAP heavy games, the uplift in revenue from system-initiated ads doesn’t usually justify the decrease in retention or session length. The best way around that is segmentation, which we’ll get to next.
Segmentation
Just like you segment your IAPs based on unique user personas, you should also segment your ad implementation by user groups.
Break up your segments by country and by revenue, distinguishing between users who engage with many ads, users who engage regularly with ads, and users who don’t at all. Then, serve each segment with varying ad units, reward amounts, and capping limits. For example, try monetizing users who are unlikely to ever opt-into rewarded videos or offerwalls with system-initiated ads. On the flip side, make sure to cap system-initiated ads to your ad whales.
Another strategy is to base user segments on app usage, and leverage user-initiated ads as a promotional tool to hook infrequent players. Give them large rewards to tease them into coming back.
Additionally, you can segment based on in-game progress. For example, serve a rewarded video offering giving users just the right amount of coins they need to progress at the exact moment they need it.
Step 3
Now that you have a basic monetization strategy in place, it’s time to analyze and optimize its tipping points. Where can you improve? How can you further maximize ARPU? Let’s start by going over some key metrics.
Key monetization metrics
The first step in optimizing your monetization strategy is analyzing key business metrics. By staying on top of the following metrics, you’ll not only get a better idea of overall business health, but also your revenue potential and areas for improvement.
Lifetime value
LTV, or lifetime value, is a metric which estimates the revenue generated by a single user throughout their entire lifetime within your game. It also tells developers at which point in their lifetime users become profitable.
There are many ways to calculate LTV, but typically developers calculate it in relation to monetization (based off metrics like ARPDAU or ARPU) and retention. For example, they build a curve to understand and what the LTV is at Day 7 and use it to extrapolate what the LTV will be at Day 90.
Effective cost per mille
eCPM, or the effective cost per mille, is a standard metric for evaluating and optimizing ad monetization. It measures the ad revenue generated by 1,000 ad impressions, and is calculated by dividing total ad revenue by total impressions, and then multiplying by 1,000.
It’s a helpful metric for monetization managers to compare ad monetization performance across multiple variables, such as ad network, region, operating system, location, etc. For example, you see that rewarded video generated 400 impressions and $5.00 is revenue, while interstitials generated 700 impressions and $3.00 - it’s hard to compare performance based on these numbers alone.
By calculating eCPM, we find that rewarded video is generating more money, driving $12.50 for every one thousand impressions, compared to the $4.29 interstitials are driving.
Engagement rate
ER, or engagement rate, is the percentage of users who actively engage with ads. It’s calculated by dividing engaged users (i.e. users engaging with ads) by total active users.
It’s unique to ironSource, but nevertheless important, as it tells monetization managers how much of their ad monetization potential they’re achieving. In other words, if only 20% of active users are engaging with ads, developers can conclude they haven’t hit their ad monetization ceiling yet - there’s still room to grow.
There are several reasons why ER may be low. Perhaps the rewarded video traffic driver isn’t visible. Or the reward isn’t valuable enough for users to care to engage with the ads. Of course, engagement rate varies across different genres, but by using benchmarks developers can ensure they’re maximizing ad monetization potential.
Usage rate
UR, or usage rate, is the number of ad impressions per engaged user in a given session.
Like ER, UR is unique to ironSource, but is critical for determining ad monetization potential. If engaged users see 1 ad per session, but benchmarks for your category indicate UR should be closer to 3, you understand that either you’re capping too early or the rewards you’re offering aren’t optimal.
For example, offering 10 extra lives for watching a video is great for ER as it’s extremely valuable, but not great for UR - since users don’t need more than 10 lives per session.
Granular ad revenue measurement
In the past, monetization metrics like LTV, ARPDAU, and ARPU were being calculated solely off in-app purchase revenue, since developers had more visibility into how much in-app purchase revenue a single user generated for them, than they did ad revenue. This was especially problematic for hyper-casual games which relied nearly 100% on ad monetization.
Today, a handful of mediation companies, including ironSource, have made user-level ad revenue data available for developers - giving developers the full picture on their monetization performance and allowing them to optimize accordingly.
As we covered previously, user-initiated ads aren’t just money makers - they can also increase key KPIs like retention and engagement. They just need to be implemented properly and optimized accordingly.
Optimizing towards retention
See the graph above, which depicts retention increasing along with each user-initiated impression served. To perform this test, we set different capping parameters, beginning with no capping for rewarded video and then capping at the first impression, and so on. We found that once we hit 7 impressions per user per day, retention continued to increase, but the difference in retention between each impression was smaller and smaller.
Optimizing towards retention
See the graph above, which depicts retention increasing along with each user-initiated impression served. To perform this test, we set different capping parameters, beginning with no capping for rewarded video and then capping at the first impression, and so on. We found that once we hit 7 impressions per user per day, retention continued to increase, but the difference in retention between each impression was smaller and smaller.
The reason user-initiated ads increase retention is because the premium content non-paying users receive in exchange for engaging with ads allows them to progress further in the game, similarly to their paying counterparts. The key is finding the rewards and values that facilitate game progression, and therefore player retention. It takes some A/B testing to find that sweet spot, which you can perform with LevelPlay A/B testing tool. More on that later.
The compelling benefit of rewarded videos, specifically, is that they can be integrated as part of the core game loop and function as an engagement and retention mechanic. For example, you can reward users the first time they open your game each day, and offer them a rewarded video to double the reward. This not only gives users a reason to play every day, but also engage with rewarded videos every day, too.
Remember to always consider engagement per user age. If you see that users stop watching rewarded videos after Day 7, then further optimization is needed. Maybe, in this case, the reward wasn’t adjusted to the level of the game or maybe the reward isn’t visible enough in deeper levels. Either way, following this metric is an easy way to see where in the lifetime optimization is needed.
Optimizing towards revenue
Theoretically, every ad unit increases ARPDAU, though a bit less with each impression served. That’s because as the session gets deeper, IPM decreases as there are fewer conversions. eCPM as a result also decreases, which ultimately means less revenue for the developer. Still, the revenue contributes to incremental increases.
The interesting part, however, is how user-initiated ads increase ARPU. In the graph above, we see how with each user-initiated impression, ARPU improves significantly, which isn’t necessarily the case with system-initiated ads as we’ll see next.
As we covered earlier, system-initiated ads are great for monetizing a broad range of users. But because of how they impact retention, they need a bit more attention than user-initiated ads during the optimization phase.
Optimizing towards retention
Unlike user-initiated ads, system-initiated ads need to be more carefully balanced against the user experience, since users don’t have control over when the ad is served. Consequently, in the graph, we see that system-initiated ads can eventually damage retention with each impression served.
Fortunately, there are ways to mitigate the negative impact on retention. First is playing with capping and pacing settings to determine the optimal number of impressions per session and the intervals of time between them. Second is finding the ideal number of seconds before allowing users to x-out of the interstitial ad.
If you’re generating revenue at a relatively quick rate, then it may not be bad for business if retention is low. In these cases, it’s less problematic to ramp up system initiated ads. Last is playing with the format inside the interstitial. For example, we know that playable ads and video ads tend to be more enjoyable that static interstitials. See how each format impacts retention for your game and build your ideal mix from there.
Optimizing towards revenue
Like with user-initiated ads, system-initiated ads also increase ARPDAU with each impression, though at a slightly lower rate. But notice how different the ARPU graph looks.
That’s because ARPU is a function of retention and revenue - with revenue increasing and retention decreasing, the ARPU graph peaks at the sweet spot.
There’s no secret equation to finding this sweet spot. The only way is to A/B test. Try, for example, capping interstitials at 3 impressions for one group and 5 impressions for another, and then compare the LTV. How many impressions do you need to serve to maximize LTV?
If you take away one thing from this eBook, it should be this: ignore your gut. Instead, base your monetization strategy on hard numbers and data. How? A/B testing. There are many complicated ways to perform A/B tests in apps. LevelPlay offers an A/B testing solution right inside the mediation platform, making it especially easy to set up tests, analyze the results, and then update the game accordingly.
There are dozens of tests you can perform, some of which we discussed briefly already, including ad units, reward amounts, placements, and pacing and capping. But that’s just the beginning. You should also test your waterfall. Determine whether single instances or multiple instances delivers higher eCPMs, or manual position vs. auto-optimization, or flat eCPM deals vs. network optimized pricing.
Step 4
Just as you’re analyzing and improving monetization, the same needs to be done for user acquisition. Which supply sources deliver the highest quality users? How can you lower your bid but still maximize your share of voice in the ad network? Your goal in this stage is to maximize profit through scale. Before diving in, let’s go over some key metrics.
Key user acquisition metrics
The key to optimizing your user acquisition strategy is analyzing the right business metrics. We can divide these metrics into two categories: performance metrics, which tell us how well your ad campaigns are performing, and quality metrics, which give us insight into the kinds of users you’re acquiring.
Performance metrics
Click-through rate
CTR is the percentage of users who tap on your ad’s call-to-action button. It’s calculated by dividing the number of clicks by the number of ad impressions.
Conversion rate
CVR is the percentage of users who clicked on your ad’s call-to-action and also installed your app. It’s calculated by dividing the number of installs by the number of clicks.
Installs per mille
IPM is the amount of installs a campaign receives per one thousand ad impressions. It can be calculated a few ways - either by multiplying CVR by CTR by 1,000, or more simply dividing installs by impressions and then multiplying that by 1,000.
eCPM
eCPM is the effective cost per one thousand ad impressions, or the ad revenue generated by a campaign. It’s typically a metric used by monetization managers, but equally important for user acquisition, as networks use eCPM to rank campaigns in the waterfall. The higher a campaign’s eCPM, the higher the network will place it in its waterfall, meaning the more impressions it garners. In this way, eCPM represents the campaign’s buying power.
eCPM can be estimated by multiplying IPM by the CPI bid.
In-ad data
There are many kinds of in-ad data, which is essentially the data collected inside an interactive ad covering the multiple touchpoints from impression to click/end. This includes win/lose CTR, exit rate, dropouts, time to engagement, and in-ad completion rate - all of which allow for much deeper creative optimization and analysis than CTR and CVR, for example.
Return on ad spend
ROAS informs user acquisition managers how much profit they made off their ad campaigns. It’s calculated by dividing ARPU by CPI. Today, most user acquisition campaigns are geared towards optimizing for ROAS.
Post-install event per mille
PIEPM is a metric that takes both performance and quality metrics into consideration. Calculating the PIEPM is the only way to measure performance in CPE, or cost per engagement campaigns, in which advertisers pay for each post-install event their ads drive.
Quality metrics
Retention rate - RR tells you how many users return to your game on any given day day, usually measured at day 1, 3, 7, 30, 60, and 90. It’s calculated by dividing the number of retained users on day X by the total installs between day 0 through X.
Average revenue per user - ARPU is the average revenue your acquired users are bringing in. It’s calculated by dividing the sum of IAP revenue and ARM (ad revenue measurement) until day X by the installs.
Remember we discussed in section one of this eBook that a good eCPM means your campaign will be placed high in the network waterfall, allowing you to capture the majority of a network’s impressions?
Since eCPM is affected by IPM and CPI, these two metrics become especially important in user acquisition. IPM is easier to influence than CPI, which largely depends on cash flow and LTV.
Here are a few strategies for maximizing IPM.
Creative optimization
This is obvious: the better your ads, the more people will want to install your game. And the better your campaigns perform, the higher it will be placed in the waterfall and the more impressions you’ll get. But of course, “better” is subjective. That’s why it’s essential to utilize in-ad data when you can. In-ad data of course is only applicable to interactive creatives. Is time-to-engagement low? Maybe the tutorial is complicated. Are users completing the playable but not installing? Maybe you’ve set the difficulty level too hard, and your players don’t like losing. There are dozens if not hundreds of possibilities.
The good thing about creatives is that they are quick and easy to improve, unlike your game, which would take months if not years to perfect.
If you’re not sure where to start, focus on optimizing these three in-ad experiences first:
The right number of levels
Typically, ads are just 30 seconds long, and sometimes even shorter. To make the most of this short time, you need to first determine how much time it takes a player to win, and second how much time it takes for them to decide if they want to install. Too short, and users may not feel compelled to convert yet, but too long, and users may drop off before they reach the CTA to install.
Try A/B testing 3 different versions - a playable ad with 3 levels, 2 levels, and 1. Which iteration balances drop-off to maximize CVR?
The right difficulty level
Making the playable too difficult turns people off. But making it too easy lowers investment and engagement. If there’s no challenge, there’s no pleasure in completing it. Our in-ad data at ironSource shows that difficulty level matters hugely, and users convert more often at a happy medium.
If you suspect your playable may be too hard or too easy, create a few versions with different difficulty levels, and see which one delivers the highest IPM.
The right outcome: Win or lose
It shouldn’t be surprising that people today expect instant gratification, and as an extension of that, expect to win easily. This is why it’s important your creatives have a high win rate. Help users win by including frequent hints every two seconds, for example pulsating arrows that make it clear what’s needed of the user, and glowing clues.
Attribution
How can advertisers be sure which video, playable, or interstitial finally convinced a user to install their app? You can never be 100% certain, but you can absolutely take an educated guess. That’s why there are so many forms of attribution: view-through attribution, click-through attribution, multi-click attribution, etc.
Since we’re talking about ways to maximize IPM, we’ll focus on view-through attribution (VTA), which attributes installs to views rather than ad clicks and tends to deliver 20% more installs than click-through attribution. The general rule is to set the VTA window to 1 day, as opposed to the 7 days that’s typical for click-through attribution. In this scenario, if a user watched your video, and a few hours later went to the app store to install your app, that video would get credit for the install - even though the user never technically clicked on the call to action within the video. It’s important to remember, however, that because views are lower on the waterfall than clicks, if there is a matching click for the campaign, the click will take precedence over the view.
VTA increases IPM for two reasons. First, attributing based on views means there’s no chance for click fraud, which is extremely prevalent today. Second, it catches installs that would otherwise be marked ‘organic’. It’s true that your ‘organic’ installs might drop, but that drop is cosmetic, because these users were never organic in the first place - they engaged with a video within the attribution window, after all.
Suppression lists
It should be obvious to say that there’s no point wasting budget targeting users who already have your app. You’ll never get an install from these users, which also means you’ll drag down your IPM which will hurt the rest of your campaign.
By providing ad networks with a suppression list, which is essentially a list of current app users, you can make sure you’re only advertising to potential new users. This results in more relevant and higher volumes of impressions, and therefore ultimately installs.
There are three kinds of suppression lists: static, dynamic, and custom.
Static suppression list
This is a list of all device IDs that have previously interacted with your game. You can share this list via .csv, .txt, or API.
Dynamic suppression list
This is a dynamic list of device IDs that automatically updates as devices install your game. It’s only enabled through your tracking solution.
Custom suppression list
This lets you target a specific group of users, in a retargeting campaign, for example.
Anti-fraud measures
In attribution manipulation, fraudulent channels attempt to steal credit for organic installs or installs from legitimate sources. As you can imagine, this practice viciously attacks a campaign’s IPM.
There are many ways to prevent attribution manipulation, but the most effective strategy is turning off non-incremental channels, or channels that don’t add to (or even lower) your total installs. The trouble is identifying which channels are non-incremental.
Be cautious of media channels that offer you free user acquisition tests, huge discounts and rebates, CPI to CPA rev share, and have “issues” with integration. In other words, if the channel seems too good to be true, it probably is - so do your due diligence beforehand.
In addition, be sure to monitor total installs each time a new channel goes live. And if you suspect you have non-incremental channels already live, test pausing them and see what happens. Do installs decrease?
As we mentioned earlier, IPM isn’t the only metric to optimize campaigns for. CPIs are the other half of the eCPM equation. It’s true that CPIs are more difficult to lower than IPM, but it becomes slightly easier with automation.
There are several dynamic bid optimizer tools on the market, including ironSource Ads' and Unity Ads' ROAS optimizers, which utilize in-app purchase and ad revenue measurement data to automatically optimize campaigns according to ROAS goals.
The optimizer makes sure it’s always bidding on the right supply sources, auto-adjusting bids every day, ultimately saving user acquisition teams hours of manual work. Essentially, it helps you achieve similar if not lower CPIs with significantly better ROAS results.
Naturally, after step four, the loop circles back to step one, and the cycle continues. By the end of each cycle, you’ll have even more insight into how your users monetize and the supply channels they come from - giving you a better understanding of which aspects in your game need to be improved.
In addition to following the loop, make sure you’re always assessing game metrics against industry benchmarks. Find out where you’re behind, and concentrate on getting better. Is ARPDAU low? Is ER?
The more focused you are on the Growth Loop and where you’re behind today, the smarter you can be tomorrow.
Want more content like this?
Head to Unity Resources - the place for people who love making, growing, and of course, playing mobile games. Get actionable insights on all things game design, development, and business.
Fill out this form to access in-depth expert knowledge and solutions.
