With a focus on genres like puzzles, coloring books, and board games, ZiMAD has established itself as a powerhouse in the development and distribution of captivating gaming experiences across various platforms.
This interview with Kirill Zhukovsky, Chief Product Officer of ZiMAD, provides valuable insights into its unique approach to game development, marketing strategies, and the pivotal role of marketing analytics tools in driving success in the highly competitive mobile gaming landscape. Join us as we explore how ZiMAD navigates challenges, leverages innovative technologies, and stays ahead of the curve in an ever-evolving industry.
Can you tell us what ZiMAD does?
ZiMAD specializes in such genres as puzzles, coloring books, and board games. We successfully develop applications not only for the App Store and Google Play—we also have enough positive cases of development for alternative platforms, such as Amazon, for instance. Despite our vast experience in classic genres, we strive to master new ones. We also have a team that finds suitable games for purchase or publishing in order to extend our portfolio.
Can you elaborate on what makes ZiMAD different from its competitors?
We can boast hundreds of millions of installations of our products and an impressive DAU both in the US and worldwide. While this is the result of years of work, our experience gives us an edge thanks to the deep understanding of our target audience. We have tested thousands of creatives—and we examine what works and what doesn’t every time. Our knowledge is based on hundreds of product A/B tests, dozens of playtests with live audiences, and constant communication with the community in social networks, which, by the way, is quite vast as well.
We have organized quite a lot of marketing and content collaborations with the world’s biggest brands, and we continue to develop this line of work, which is another component of our success. The preparation of such activities usually includes from 2 to 6 months for coordinating the marketing assets and all the legal nuances. But once everything works out, the rapid growth of KPIs is accompanied by a sense of pride for the work done and a feeling of having been in touch with something great.
How do marketing analytics tools contribute to understanding mobile game advertising trends, particularly regarding user acquisition strategies?
To avoid any confusion, I will roughly divide analytics into internal and external. Data on internal analytics comes directly through SDKs integrated in our own products and helps to optimize campaigns on the go, identify successful creatives, and forecast LTV and ROI. From now on, I will mainly talk about external analytics, i.e. systems that allow you to track and study the activities of competitors and influential advertisers in various genres.
It must be said that we have repeatedly found ourselves in a situation where we came up with a successful UA strategy based on a funnel with good conversion rates, but thanks to advertising analytics services, all key competitors were able to copy that strategy literally in a month, so the advantage was lost. So when you are the leader, such services play against you. But you have to constantly fight to be ranked number one as the market is very competitive. And in an environment where everyone is watching each other, we have to use the entire arsenal of tools for external market analytics.
User engagement starts with choosing a strategy. Insights into users’ motivations, needs, and interests help to shape it. Such information is provided by services like Google Analytics, SensorTower Audience Insights, Quantic Foundry, or Solsten.
Reports on top advertisers give an idea of the current state of the market: who, where and in what volumes is buying displays. They also show how fierce the competition for users is and make predictions about whether prices for installs will rise or fall. Performance Index Reports from Appsflyer, Singular, Kochava, and other services are useful for understanding the global situation. They also highlight the Share of Voice of individual networks in the media mixture of advertisers of particular genres, niches, or peer sets. Monitoring these changes is critical because this is where growth points can be found.
External analytics systems help to understand why a campaign’s performance has changed—whether it’s due to creative burnout or, for example, aggressive traffic acquisition by competitors. Based on the information obtained, budgets can be adjusted, and approaches in creatives and messages can be changed.
At the same time, the information about the performance of other companies’ creatives does not tell us much. Numerous impressions suggest that an advertiser (publisher of a game or an app) could afford to buy ads in the given networks and locations for a certain period of time. But this information is rapidly becoming outdated, or it even already is. It’s quite possible that the creative that is now being displayed as the leader in impressions is actually now yielding poor conversion rates, but remains on because the advertiser has been unable to find a replacement for it.
Our experience in communicating with major publishers shows that seemingly successful projects with trendy mechanics, which have been buying 250,000+ installations a month for a long time, sometimes turn out to be unprofitable at the end of the year. We’ve witnessed several examples of such campaigns. In one of them, the publisher recorded a loss of more than $500,000.
Therefore, the boost of installs on a specific funnel visible in the analytics system reveals the essence of the experiment, but does not give a hint about its success. Repeating the scheme without understanding all the details is associated with risks. Long-term trends of 6 or 12 months are more reliable, but in a dynamically changing competitive environment, we cannot wait a year just watching someone else’s campaign.
Understanding someone else’s UA strategy requires studying all the available information, and not just the number and duration of creatives shown. We collect and cross-check data from at least two independent sources. And although a huge amount of information is collected, it can easily become confusing, so any analytical conclusion is nothing more than a hypothesis that needs to be tested.
Can you explain how analytics tools facilitate the analysis of metrics like impressions, CPI, IPM, Install Rate (IR), CTR, and CPM and their impact on user acquisition campaigns?
In the best systems, you can now view the retention and engagement metrics of a third-party app. However, when we double-check them, we often see a large margin of error. So one should not fully rely on this information. As for CPI, IPM, IR, CTR of specific applications, they are not disclosed in public sources. This is the biggest challenge, and, at the same time, a benefit for the industry. Otherwise, marketing turns into a poker game when all the cards of the opponents are revealed. We have seen attempts to add sections with such analytics in advertising networks, but there was either too abstract information, or the solutions were almost immediately covered up because of the risks of trade secret disclosure. Generalized information is more useful for R&D teams, while UA specialists benefit from detailed metrics.
When comparing the campaign with the benchmarks of a peer product, only accurate and reliable information allows finding bottlenecks and work on them. This is why the skill and experience of a UA specialist is so valued: with incomplete information on competitors, they can quickly decide which creatives and strategies work on certain grids and which do not.
Could you elaborate on how marketing analytics tools help in identifying user quality as a top priority for mobile game advertisers, especially in terms of metrics like retention, LTV, and ROAS?
Marketing analytics tools help to determine marketability and niche capacity, but for specific products, not all metrics can be viewed. And those that are available contain quite a significant margin of error.
ROAS and LTV for third-party projects are unlikely to be seen in any system. But some advertising networks regularly publish generalized CPI data broken down by genre and location. If you make a correction for the subgenre, correlate this information with Retention and number of sessions from the external analytics system, and know the approximate ECPM, you can build an economic forecast. The calculation will be rather approximate, but it is still better than nothing.
The integration of AI tools into mobile game advertising has become prominent. How do marketing analytics tools aid in leveraging AI for optimizing ad creatives and campaign performance?
By processing large amounts of user data, AI and ML systems increase the efficiency of traffic procurement. Google was the first one and transferred UA processes to ML algorithms. They were followed by Meta (Advantage + complete campaign solutions). And now, almost any network offers algorithms for automatic optimization for a given goal (be it ROAS, In-app Event, or Retention Rate).
UA specialists along with analysts also rely on AI/ML to build predictions of cohort behavior.
AI trained on information from external analytics services can also help in the creative process. There are advertising video generators that can generate hundreds of creatives in a couple of clicks and take into account some of the information accumulated on converting and non-converting approaches. So far, the logic for determining converting approaches is too primitive to seriously rely on such solutions. Nevertheless, the progress is obvious.
Ideally, AI should cover all aspects of creative creation within a single solution:
1. The analytics system examines all data and selects 10 creative references that have the highest probability of success for a given geo, network, and campaign type.
2. Based on the references, a text AI-module generates the required settings and a scenario. We have already tried such a scenario generator in practice.
3. The scenario together with the uploaded assets of the game itself goes to the image or video generator.
4. Then AI-sound is created, which is also a reality now. Algorithms for synthesizing music and voice-overs in any language produce quite a high-quality result.
5. In the final stage, the video or picture is converted to all necessary resolutions.
Naturally, it should be possible for the UA manager to make a wide range of edits and improvements at each stage. It sounds complicated, but a few years ago it seemed impossible to generate photorealistic images on demand using AI, and certainly not video.
How can marketing analytics tools assist in measuring the effectiveness of offerwall and UGC in video ad formats and tracking their impact on user acquisition and engagement?
Analyzing aggregated data from networks using various services that specialize in offer wall traffic and identifying long-term trends helps to gather some information to improve our own campaigns.
As for analyzing the success of UGC videos in social networks, it is not easy to make decisions based on the data on impressions, because, in addition to the unique charisma of the people in the frame or behind the scenes, social factors also play a role. We had a video that gained tens of thousands of comments and a lot of likes over 5 years, which provided an excellent conversion rate. But if any competitor tries to repeat it, they are unlikely to succeed in the short term, because they will not have the right level of social activity. It’s not as simple as it seems.
The concept of creating “fake games” for user acquisition presents a unique challenge. How can marketing analytics tools help identify and address deceptive advertising practices while maintaining transparency and user trust?
Personally, it’s not so much the ads for misleading mechanics that annoy me, but, for example, outrageous ads. Let’s say, I’m playing a relaxing game before bedtime and suddenly see an interstitial with a screaming head coming out of the toilet. Unfortunately, networks practically do not give the owner of the game tools to automatically block this type of videos; it can only be done manually, which is very inefficient. Another unpleasant factor for us is related to forced transitions to the store without user consent after the display of the ad. Large networks do this among others, creating an uncontrolled flow of users from our games to other games. Force transitions not only annoy everyone, but also distort the data in marketing analytics tools.
And as for “fake games”, the industry is rebuilding, and such mechanics are now integrated into products as mini-games and temporary events, which is perfectly legal, both legally and ethically. In the case of harsh misleading creatives, there is always a wave of negative feedback from outraged users. Even if advertising networks show weakness in such matters, stores can easily resolve the issue on their side by detecting patterns inherent in misleading attitudes with AI systems and severely penalizing violations.
How do marketing analytics tools assist in benchmarking and comparing performance metrics across different ad creatives, platforms, and user segments to refine advertising strategies?
Decisions on building an advertising strategy fall on the UA manager and CMO, but analytics services certainly help to react to trends. Internal analytics allows us to forecast LTV. But depending on the type of monetization and payback period, it can be a very difficult task in cases when we need to make forecasts for one or two years ahead. Unfortunately, mistakes are inevitable simply because there are many external factors, such as advertising demand and social and global trends. During Covid, many advertisers left the market while people were sitting at home and playing games more due to restrictions—many people’s forecast LTV did not match the actual LTV in both positive and negative ways. In case of a negative development of events, there should be a certain margin of safety in the budget.
External analytics systems are also important in this matter: if we see that one of the competitors has increased activity on a certain network and location, we definitely consider this information and make prompt decisions.
What role do marketing analytics tools play in enabling real-time monitoring and optimization of ad campaigns to ensure maximum effectiveness and return on investment (ROI)?
Recently, MMPs (mobile measurement partners) have done a lot in the direction of real-time analytics of ad revenue generated by new cohorts of players. This allows UAMs to respond to changes faster. Such services work closely with clients and are willing to customize functionality to meet their needs. Nevertheless, to obtain the most complete information, we actively use our own BI system, through which we monitor ROI, ROAS, and other important parameters. Looker Studio, Tableau, Power BI, and other solutions can serve as a foundation for such a system.
How do marketing analytics tools adapt to the changing landscape of user data privacy and audience targeting without relying on third-party cookies or identifiers?
One of the areas of development is so-called Interest-Based Targeting based on analysis, segmentation and, where applicable, extrapolation of first-party data obtained in the course of user interaction with the site. It is also noticeable how services are improving LTV prediction mechanisms based on early signals on engagement, retention, and revenue. There is a lot of money at stake, so more and more effective tools are emerging to run campaigns under severe restrictions on user data.
Can you provide examples of specific marketing analytics tools or platforms that are particularly effective in tracking trends across creatives and optimizing user acquisition campaigns in the mobile game industry?
It’s better to start with tools for analyzing creatives and trends provided directly by ad networks: Google’s Ads Transparency Center, Meta’s Ads Library, and TikTok Creative Center. TikTok provides the most advanced service with its flexible search system, real metrics of creatives’ performance, and AI tools in the backend that decompose each creative into elements, allowing you to prepare new scripts and generate new ideas in seconds based on effective concepts.
As for other specialized services, after SensorTower unexpectedly acquired Data.ai, the list of recommendations became a bit shorter. We can also recommend Appmagic and MobileAction. We are in contact with the teams of the top services, and I can say that all of them are very actively developing. There are also a number of ASO analytics systems, in particular Apptweak and Appfollow. For A/B tests of landing pages, tools like Splitmetrics, where decisions are built on a data-driven approach, will be useful. Services like Replai.io promise to increase the efficiency of working with creatives, although right now we use in-house solutions in this matter. And you can’t go far without a modern attribution system, which I mentioned earlier.
Kirill Zhukovsky
Chief Product Officer, ZiMAD
