Principles and approaches to monetization in Belka Games. Advertising, trends and practical cases

Evgeny Gilmanov, Head of monetization at Belka Games

This is the third part of a series of articles about monetization. In the first, we talked about the foundation of monetization and intersections with other areas in the company, in the second — about setting up monetization within the product, finding growth points and generating ideas. In this one, we will go through advertising monetization, draw a portrait of a candidate for the monetization team, briefly describe market trends and give a couple of cases from practice.

Advertisement. Where to start, how to improve

I will immediately give you a link to the description of advertising formats — it will help if you have not encountered the introduction of advertising and do not know the basic terms.

There is an opinion that advertising harms IAP monetization, since it negatively affects retention and cannibalizes the desire to pay. Often this opinion is formed by our experience and observations of the hyper-casual games market, where advertising is imposed almost from the first level. But there are examples in my experience when a well-implemented combination of advertising and IAP monetization made it possible to achieve high payback rates in a short time.

When the question of integrating advertising monetization arose in Belka Games, many players in the casual mobile games market used advertising as a way to monetize their audience.

Our products were initially built according to the balance for IAP monetization and showed fairly good stable metrics, so the first task was to get a profit and not break anything. Hence the main principle:

Important: When designing an advertising monetization system, you must thoroughly understand how it works and keep all documentation up to date. In order not to break anything, you need to understand what can be broken.

Let’s talk about the format of Rewarded advertising. In order for such advertising to work, you need to follow several configuration rules:

1. Moderately useful bonuses for the user. Too little value will not attract attention, too much will break the balance.
2. Timeliness of the offer. If the player has accumulated enough currency, the offer to receive is unlikely to receive the proper conversion.
3. UI. The offer should be visible in the game interface and understandable to the user.
4. Limits and cooldowns of impressions. You have to control how many times a player sees an ad offer, how many times he can get a bonus for viewing ads, what cooldowns there will be between offers and successful ad views.

Based on these rules, we made the first prototypes of the system and conducted dozens of tests to get the best metrics. Of course, we have not received the optimal settings yet, but we regularly improve the details.

What to do with newcomers? When to turn on a rewarding video, when to turn on a full-screen advertisement (Interstitials), how not to break the conversion on the first purchase?

We asked ourselves these and a dozen other questions during the initial integration of advertising. We did a lot of research, talked to industry acquaintances, and began cautiously trying out various advertising formats. This is important at least in order not to reinvent the wheels or step on the rake. There are no universal recipes found — so try and explore the audience.

For us, the main metric here was the conversion to the first purchase and the distribution of players by levels and days of life before the first conversion.

A separate topic is the limits on impressions at specific points. The economy is like this: in most cases, the most expensive display will always be the first. If you infinitely increase the number of impressions per day per user, then sooner or later the price of the impression will tend to zero. That is, your interest is also constantly decreasing, but it is not close to zero, but goes into negative territory, because the player gets some kind of gaming advantage for watching the video, and you get almost nothing.

It should be understood that not all players who have watched ads at a particular point at least once a day will reach the limit limit. And if you decide to increase the viewing limits at some point, it’s worth looking at the volume and composition of the audience you work with and evaluating who you can profit from. I’ll give you a case from our practice later.

Another pattern: the more advertisers you have in the auction for a user, the slower the eCPM will decrease. But the more advertisers there are, the greater the load on the stability of the game. Keep a balance between user experience and money.

Who to take to the monetization team

A candidate, as for any other position, can be obtained in two ways: to raise or to search the market. You will have to search for a long time and it is not a fact that there is a necessary set of skills for your particular business.

What is important (without priorities):

• understanding the market;
• understanding the economics of your product from and to;
• technical background;
• proactive attitude and ability to sell your ideas;
• creative thinking;
• good knowledge or direct experience of working with a product, marketing, analytics, game design.

The last point answers the question of how and from whom to raise the person responsible for monetization. We take a conditional marketer or analyst, focus on developing the necessary skills and performing specifically targeted tasks, and get the necessary skill set for the company.

The question may arise: why should a person go in this direction at all, if analysts, marketers and product managers often close this position in companies? It’s simple — a trick. The task of the marketer is to attract target users; the analyst is engaged in mathematics, reports, research, forecasting; product — designs features, looking for ideas for product development. If you involve them in monetization partially, then the return will be partial — which means you will not know how much you have realized the monetization potential of the product.

Where is monetization going in 2023

Trends in the current major markets for many reasons are now more negative. Many companies from the top gross USA show a fairly strong decline in metrics, both qualitative and quantitative, for a number of reasons — and you know about many of them for sure:

• it is becoming more and more expensive to attract users, and even more so paying ones;
• it has become more difficult to enter even the top 200 of the US gross, and it will be even more difficult. There should be a strong product, a well-balanced monetization system, and a lot of support from traffic attraction;
• platforms are working more and more closely with the privacy policy, and this needs to be closely monitored;
• whales focus on fewer games;
• more and more games from the top gross are switching to the hybrid monetization format. Here it is important to find the perfect balance in the advertising settings for each segment by demography and other important behavioral characteristics of users.

On the other hand:

• the quality of games is growing;
• More and more major PC/console game publishers are becoming interested in mobile games;
• game developers are increasingly conducting large-scale collaborations with brands and stars. And players are increasingly interested in the PR component of the product brand and the company that produces it;
• strong mobile devices in the markets of developing countries are becoming more accessible, as well as high-quality low-cost communications. The trend of increasing the solvency of local players is gaining momentum, and we must prepare in advance to work with this audience.

Cases from practice — instead of conclusion

Analysis of advertising monetization and assessment of potential profit from changes. Clockmaker

Clockmaker already had an established advertising monetization system: viewpoints, limits, cooldowns. The task was to increase revenue with little effort and without updates.

One of the solutions is to increase the limits of displaying points per hour and day at the most popular points and wait for the metrics to grow. We have come a little more difficult to assess the potential:

1. We divided the players who watch ads at each point by the number of views per day. Thus, we got the number of players who view ads up to the daily limit. It is this audience that is our target audience in the limits test.
2. Further, based on the potential of the point, we modeled:
   1. The distribution of players by the number of views, taking as a basis the existing distribution up to the maximum number.
   2. Reduction of the total eCPM of the point. The more people watch ads in a certain period, the lower the price tag in the auction.
   3. At the output, the model showed how much money we can potentially get from each point with a successful experiment.
3. At this stage, we refused to increase the limits at some points due to the small potential.

After the test, we tested the model, calculated the potential from an even greater change in limits. The metrics would have changed just as little. The next test was abandoned.

We are looking for growth points in old features. Solitaire Cruise

Products in the operation are overgrown with a mass of features. Some part of them inevitably becomes obsolete and ceases to give the required profit. We found a good growth point among them.

At some point we needed to get more purchase triggers. We decided to add new offers to the game, but previously agreed to conduct a study. Having studied the product database and delved into analytics, we found several offers from the time of the softlonch, with the right functionality and the right balance.

However, they are launched under difficult conditions, as a result of which they are seen by a small percentage of users and only in the first seven days of the game. With the change of a couple of parameters, we were able to increase coverage and get the necessary profit.

The moral here is to use your product knowledge base and a team of analysts to find clear and simple growth points.