According to statistics from DianDian Data and Sensor Tower, Glacier Network’s *X-Clash* saw explosive growth in October 2025:With 22.2 million monthly downloads and revenue of 13 million yuan, the game saw month-over-month surges of 431% and 251.4% respectively compared to August. It not only topped the iOS Free Charts in nearly 50 countries, including the U.S., Brazil, and Canada, but also ranked among the top 4 in global mobile game downloads and No. 1 in download growth.
The overseas success of this product is underpinned by Glacier Network’s systematic capabilities in game mechanics integration, project incubation, and commercial operations. Recently, Game Tea House conducted an exclusive interview with Zeng Baozhong, Deputy General Manager of Glacier Network and Head of the Incubation Center, to delve deeply into Glacier’s methodology for creating hit products—from next-generation hybrid games to incubation and market launch.
Zeng Baozhong, Deputy General Manager of Glacier Network and Head of the Incubation Center
01
Gameplay Fusion 2.0
From “Forced Integration” to “Engaging and High-Volume”
Teahouse: As the cost per user for secondary gameplay continues to rise and market competition intensifies, the industry’s interest in the concept of “secondary gameplay” has waned compared to previous years. Why has Glacier still been able to use the same strategy to create a new overseas hit like *X-Clash*?
Zeng Baozhong: People might only see the familiar side gameplay and mistakenly assume it follows the same formula as Glacier’s previous games. But in reality, this new title has undergone significant optimization in terms of content integration. The side gameplay is no longer separate from the main gameplay; instead, it is deeply integrated with core mechanics such as city-building and SLG elements, working in tandem to drive the game’s progression.
In the early "Convergence 1.0" products, the integration of secondary gameplay with the main gameplay was rather rough, resulting in a noticeable disconnect for players.Nevertheless, “Fusion 1.0” still immediately reduced user acquisition costs. This was primarily due to the “small-to-large” model—where secondary gameplay was used to drive user acquisition—which effectively helped us reach a broader audience beyond core players, bringing additional growth to the game. This is particularly crucial in an era of saturated competition where every user’s attention is fiercely contested.
After “Integration 1.0” fully validated the feasibility of the underlying logic of this model, we concluded that the integration sector holds great promise and that we could build upon the 1.0 foundation to develop a next-generation integration product that is even smoother and more engaging. Consequently, we engaged in continuous internal discussions and iterations, and this year we introduced a new concept: “Fun and High-Volume,” also known as “Integration 2.0.”Our goal is to create products that “look fun and are actually fun.”
To this end, we have also formulated the "Fusion 2.0" formula—namely, "addictive core mechanics + replayable structure + a mature monetization model"—as our internal development guideline. Combined with a game progression that moves from light to heavy and from simple to complex, this approach achieves the dual effect of enhancing the product’s appeal while ensuring playability, resulting in a product that offers greater long-term enjoyment.
Since late last year, we have been using this combination of strategies to continuously advance our efforts in integrating and developing projects, and some of our products have already delivered impressive results. For example, *I Am the Great Lord*, which we incubated and launched this year, has shown outstanding performance in key metrics such as retention and monetization. In addition, we have several other incubated projects set to enter testing in the fourth quarter.
"I Am the Great Lord"
02
The "50-50 Model" is beginning to show results
Several incubated products will be launched for testing later this year
Teahouse: We noticed that Bingchuan proposed a “50-50 Plan” earlier this year. Could you tell us how this plan is being implemented in practice and what results have been achieved so far?
Zeng Baozhong: “50-50” is the “starter kit” we provide for promising projects—projects that pass our screening process receive funding and resources to conduct testing. This allows teams to focus on refining a fun core gameplay demo without worrying about “not having the budget to experiment,” and then proceed to launch it for testing, where players ultimately decide whether it’s fun or not.We expect to have around 50 internal demos this year. It’s important to note that not all of them will meet the standards for external testing, but there are indeed some demos that have been internally evaluated as having significant potential.
Of course, money is only one factor. Many external development studios partner with us not because of the financial compensation—many top-tier studios aren’t short on funds—but because Glacier can provide support at every stage of the development process to help them create fun and successful products. That’s why they choose to work with us.
Selection of products from Glacier Network; Image source: Glacier Network 2025 Mid-Year Report
Teahouse: How exactly does Glacier manage to release new gameplay features so frequently?
Zeng Baozhong: In order to quickly identify gameplay prototypes that appeal to players, the incubation center employs a "rapid development, rapid testing" approach to validate the feasibility of projects.
Incubation is, by its very nature, a "non-focused" endeavor; after all, no one can predict in advance which product will definitely succeed. It’s like playing Texas Hold’em—you don’t know the value of your hand until you see the flop.Of course, not focusing doesn’t mean shooting in the dark. Even in Texas Hold’em, you need to have your own “hand range”—knowing which hands are worth playing and which aren’t; and for those worth playing, how to allocate investment and resources. The key to this process is to minimize the cost of failure: use rapid testing to assess a project’s potential for success or failure while investment is still low, and know when to fold.Conversely, for high-potential projects, you can gradually increase your bets once you see positive momentum. In other words, the process involves first not focusing—"checking the flop at minimal cost"—and then narrowing your focus.
03
The "Competency Centers + Creative Swarms" Organizational Model
Supporting Outstanding Creativity
Teahouse: There’s a saying in the industry that “project approval determines success or failure.” How does the Glacier Incubation Center handle the project approval process?
Zeng Baozhong: It’s widely accepted in the industry that the project initiation phase determines a project’s fate. I believe this phase accounts for 70% of a project’s success. However, a key contradiction in the industry is that most companies allocate the bulk of their budget to mid-to-late-stage production, while spending very little on the initial project initiation phase. Therefore, to focus our resources and efforts on this critical phase, my primary task upon joining the Glacier Incubation Center in early 2023 was to establish an organizational structure tailored to innovation incubation.
The Glacier Incubation Center is a networked organizational structure comprising a "Competency Center" and a "Creative Swarm." The Competency Center provides the professional expertise necessary for project initiation, including core R&D capabilities such as technology, art, and game design, as well as specialized support from the Evaluation Team and AI Lab. It also covers the full range of supporting capabilities across the entire development process, including business operations, project management, publishing, and testing.These resources are made available to internal teams as well as external content partners and producers from the very beginning of a project. By leveraging professional expertise, we reduce early-stage trial-and-error costs and maximize the success rate of project launches.
Glacier Incubator establishes a Center of Excellence, with publishing producers serving as nodes to liaise with external content partners, forming a networked "creative swarm" organization.
At the same time, members of the Competency Center can transfer to the Creative Swarm at any time. The core logic behind this is to leverage the stable support of the Competency Center to address the high uncertainty inherent in incubation work.For a project to progress from concept to implementation, it must first undergo thorough internal discussion, evaluation, and validation. After passing the conceptual phase, a small team is assigned to build a preliminary prototype and conduct testing, with resources released in stages based on phased validation. If test feedback is unfavorable, the team can make timely adjustments; if, after multiple optimizations, the need for the product is still not validated, the project members disband and return to the Competency Center to participate in new proposals. This flexible mechanism allows teams to iterate rapidly through trial and error, avoiding wasted resources.
Teahouse: Regarding the practical implementation phase, what does the general incubation process look like, from project initiation to launch testing?
Zeng Baozhong: Project initiation begins with a sound commercialization hypothesis. Every week, our internal incubation team holds regular opportunity-identification meetings—similar to brainstorming sessions—where everyone shares opportunities they’ve identified over the past week. These might include highly engaging content, a promising gameplay mechanic spotted on Steam, or insights drawn from recent market trends or leaderboards, and so on. Based on these opportunities, we discuss what the corresponding commercialization hypotheses are.After the meeting, the PM compiles a list of opportunities to either pursue further research or discard. Once the research is complete, a proposal is drafted for review.
The purpose of the review is for the team to help you conduct a conceptual analysis to verify whether the assumptions underlying this opportunity hold up in terms of business logic and gameplay design. Once the assumptions are validated, the team will quickly build a demo and make initial optimizations based on feedback from core members.
Teahouse: From "building a demo" to "going live," how many rounds of testing does a project typically go through? How do you determine if it’s worth moving forward with?
Zeng Baozhong: A four-step approach, with clear validation objectives for each step:
The first step is the “core team evaluation”: The team starts by creating a rough demo, which team members test and discuss among themselves—whether it’s fun to play, and if not, how to improve it. This process is repeated weekly… Some teams even release a new version every day until they’ve iterated to a point where the team agrees it’s fun, or until they run out of ideas and hit a wall. At that point, they can temporarily set the idea aside to work on something else; if new ideas emerge later, they can always come back to it and continue iterating.
The second step is the “In-House Trial”: This is open to all members of the incubation center to gauge how many people are willing to play and how long they stay engaged, allowing the team to gather more authentic user feedback and suggestions;
The third step is the “company-wide trial run”: With nearly 2,000 employees who all love making and playing games, the team recruits players from the company’s main chat group—the number of participants measures “user acquisition potential,” while the first- and third-day retention rates gauge “retention potential,” effectively serving as a “small-scale data test”;
Step 4 is “External Testing”: Once the internal data meets the criteria, we’ll test the game with real players to see if the day-2 retention rate meets the “50% fun threshold.”
The criteria aren’t set in stone: while a “50% retention rate” is a common benchmark for external communication, for action-adventure games, one might look at the “completion rate,” and for monetized products, the “ROI.” As long as the R&D and publishing teams reach a consensus that the product’s commercial value can be demonstrated at minimal cost, it’s worth moving forward.
Teahouse: What if it turns out to be less fun than I imagined?
Zeng Baozhong: You don’t need to try to cover every angle from the start—in fact, that’s impossible. The incubation process is designed to help you take an idea that’s initially only 20% fun and gradually polish it up to 120% fun through rapid testing and validation. If, in the end, it doesn’t turn out to be as fun as you imagined, as mentioned earlier, you can abandon it and start over—we accept that.I often say that game companies shouldn’t treat project failures as a major issue, as that would stifle the company’s innovative spirit. In this industry, the success rate is only around 10% to begin with. When it comes to incubation, you have to accept failure; what we’re competing on is the cost and efficiency of trial and error.
04
Focus on niche markets + Full-platform coverage
50 internal demo versions are ready to go
Teahouse: In the current overseas market, what other areas of integration are worth exploring? What are Glacier’s plans in this regard?
Zeng Baozhong: We will categorize tracks based on each team’s areas of expertise, rather than blindly following trends. For example, if a team excels at “super-casual” games, they can focus on “exploring deep commercial integration within the super-casual genre”—while the user base for super-casual games is large, past monetization has relied solely on advertising. Could we incorporate progression elements and light social features to attract a broader audience while creating deeper commercial opportunities?For teams or studios specializing in card games, we’ll explore “front-end traffic-driving mechanics combined with a card game core.” The key is to leverage each team’s strengths to create products that are fun and engaging for a broader user base.
Teahouse: The first-half financial report mentioned "increasing the launch rate of new products." How many products does Bingchuan currently have in its pipeline? What are your considerations regarding a cross-platform strategy?
Zeng Baozhong: Our product pipeline is based on the information disclosed in the company’s semi-annual report. As for internal playable demos, we expect to develop about 50 this year, though not all of them will necessarily be tested with players. Furthermore, even if they progress to public testing, it’s still unclear how many new products will launch next year, as we’ve set fairly high standards for what constitutes a “fun” game. However, increasing the launch rate of new products is certainly a key priority for us.
As for our cross-platform strategy, we do not impose any “mandatory requirements.” Instead, we determine the launch sequence based on the specific user demographics of each project, though we generally aim for full platform coverage in the end. For example, *X-Clash* was launched simultaneously on iOS, Android, and multiple international platforms. The key principle is to “select channels based on product characteristics,” rather than “forcing changes to the product just to launch on a particular platform.”
Conclusion
From the global success of *X-Clash* to the strategic development of 50 in-house gameplay demos, Glacier Network’s core competitiveness lies in transforming “gameplay integration” from “isolated experiments” into a “systematic capability,” and shifting “project incubation” from “relying on luck” to “relying on mechanisms.”As Zeng Baozhong stated: “Incubation isn’t about shooting in the dark; it’s about ‘first casting a wide net to identify opportunities, then narrowing the focus to bring them to fruition.’ Through intensive iterative testing, we refine the 20% of products that are fun into an exceptional 120% experience.” With more products from its pipeline set to launch, Glacier Network’s performance in overseas markets is worth keeping a close eye on.
原创文章，作者：游茶妹儿，禁止转载：https://youxichaguan.com/en/archives/194664
