CPI is 0.15$, stay 45%, why the 16-minute game is abandoned

At 3 pm on February 18th, Voodoo conducted a new Chinese live broadcast at station B (search Channel_Voodoo). The theme of this live broadcast is “Resurgence of 4 Unsuccessful Launched Products”. The following is a compilation of the live content, I hope it will be helpful to all developers.

CPI0.15$, stay 45% of “Design Master” for the second time

The first case is “Design Master” by Fabrika Studio. The gameplay is as shown in the picture. Putting the furniture from the car into the room is a gameplay that can be understood at a glance.

When the team was brainstorming, a member of the Fabrika studio suggested that it is fun to put furniture in the car. This kind of gameplay is very similar to Tetris. Another member suggested that since the furniture can be moved into the car, is it possible? Conversely, what about putting furniture from the car into the room?

This discussion gave birth to ideas for two gameplays. Fabrika studio decided to conduct a CTR test to compare two game videos at the same time. The results showed that moving the furniture from the car to the room for placement received higher clicks, so Fabrika studio decided to develop prototypes based on this gameplay.

Through this case, we can see that transplanting classic gameplay into ultra-casual games, improving from 2D to 3D, and changing the “toys” of the game can all help you get excellent game ideas.

What other great creative ideas for ultra-casual games?

After determining the development route, Fabrika studio faced the following challenges:

1. The game requires precise drag control;

2. The team realized that this product relies heavily on level design and requires a lot of content support. Moreover, in the process of content production, material production is time-consuming;

3. The game lacks a specific goal, because the player needs to put the furniture in the room and score directly after putting it out, but the goal is not clear;

Not only that, the depth of the game is limited, and you can get scores when you put furniture casually, and the scores are relatively random, and there is no real scoring system.

This is a problem they found in the creative stage, but Fabrika studio decided to give it a go.

The first test data is very good, CTR5%, CPI0.15$, 45% for the second time, 4% for the seventh time, and the game time is 16 minutes. The team analyzed and believed that the advantages of the game are: it is a relaxing game; the player can quickly enter the state of flow; and the game is completely intuitionistic. But there are also many shortcomings, such as: lack of user rewards for the game; lack of real game goals; 85% of female users, for example, during internal QA, many QA personnel in the studio are men, and they generally do not feel the fun of the game, but women It can often reach 20 minutes of game time.

After seeing the data of $0.15, Fabrika studio decided to iterate, hoping to add game goals and increase the depth of the game.

In the iteration, Fabrika studio believes that more levels and environments need to be added, because they believe that levels and environments are a good guide, and more environments and layouts can increase the long stay of the game. In addition, the game has joined the neighbor system and bidding system, and optimized the game control and game lens.

During the review, Fabrika Studio stated that some mistakes were made in this iteration. They focused on the level design and materials, such as various houses and furniture. The neighbor system means that the player needs to decorate 5 houses, different styles, etc., and a lot of time is spent in the production of these materials. The auction system means that the house can be sold after it is decorated, but the price of the auction has nothing to do with the result of the decoration, even if the furniture is piled together, it does not affect the price. Then this system is obviously not effective.

After a huge time-consuming iteration, the data and CPI of the second test did not increase significantly, and the game duration increased, but Qiliu still only had 7%. The advantages of the game are the same, but the shortcomings have not been significantly improved, because content-driven does not bring better seven retention, the game goal is still lacking.

But facing the game data of CPI 0.15$, Fabrika studio decided to push the game to the Soft Launch stage.

After entering Soft Launch, Fabrika studio also made more attempts. Such as joining the city, gaining money by decorating the house to upgrade the city, hoping to improve the long stay; adding cleaning the room, fire fighting and other games; adding a 360-degree rotating camera and so on.

In addition, a large number of AB TESTs were conducted. The Control part in the picture below is the original version. Subsequent tests have added a variety of systems such as energy system, fire extinguishing system, 360-degree camera, city system, nail art, and mission. In the end, it was found that the data of this version was lower than the original version, which meant that all iterations had failed.

With the continuous expansion, the game CPI quickly rose from 0.15$ to 0.6$. If the CPI of the game rises rapidly in the initial stage of expansion, but the retention and LTV are not significantly improved, this means that the ROI cannot be positive. As you can see from the picture, the game has also added various content such as AR elements, but they cannot save the game. Eventually “Design Master” was abandoned.

Through this case, Fabrika studio shared their experience:

First of all, they spent too much time on this project. The reason is that each iteration requires a lot of material production. Between the first test and the second test, it even reached 1.5 months. “Design Master” was jointly developed by 4 developers. All subsequent projects of Fabrika Studio only used 2 developers because the communication efficiency of 4 people was too low.

In addition, if you find that the content of the project consumes a lot in the creative stage, think twice about the project; if you find that the depth of the game is problematic in the creative stage, think twice; if you find that the project requires a lot of development time, think twice.

Failure cases in the stacking wave

The second case is from Machina Studio. There are 3 people in the team, one is a developer, one is an artist, and one is a product manager. The case to be shared today is a stacking product “Stacky Tower! “.

In the creative stage, Machina Studio saw the product “Sticky Block” and liked the core gameplay very much. Machina Studio decided to innovate based on “Sticky Block”.

In the first test, the CPI was 0.75$, 32% for the second time, and 8% for the seventh time. Because the second stay and game duration data were pretty good, Machina Studio decided to iterate. They also found some problems in the first version, such as the stacking mechanism is not smooth; the camera angle is not perfect; the visual effects of the game are somewhat outdated; the game lacks excellent level design, and the team does not know how to design the level; and they do not Determine if the game is innovative enough. The goal of their iteration is to better level design, adjust color coding, and add Fever mode.

As a result, there was a second test soon. In this test, the CPI dropped significantly. Machina Studio believes that this is because the physical effects of “destroying” in the game are cool, and they added this “destroying” effect to the advertisement. However, the retained data of this version has not changed. The most fundamental problem is the lack of sufficient depth. At the same time, users cannot get rewards from the stacked height, because the height has no effect on the game result. In order to solve these problems, Machina Studio decided to add the level ending mechanism of climbing stairs.

But the results of the third test were not satisfactory. Neither the second stay nor the 7th stay changed. The only change was that the CPI became higher. The Machina studio decided to iterate quickly. They hoped to add a “little man” and let the “little man” push the cart and stack wooden blocks on the cart. Machina Studio believes that this can give the audience a better sense of substitution and reduce CPI.

From the results, every KPI is worse. At this stage the team is very tired, and there is no better modification plan. Finally decided to abandon this product.

From this case, Machina Studio believes that they have learned:

1. The depth of the game is very important. Players of different levels must have corresponding gameplay;

2. It is necessary to define a well-defined core cycle, find a good rhythm, and try to make the player enter the “state of flow”;

3. Each version iteration needs to have clear goals and changes. For example, there is no obvious change between version 2 and version 3, which wastes time.

How to create a good game rhythm? What is the flow state?

Cube Surfer case sharing and the rhythm in pan-parkour games

CPI0.15$ gameplay innovation case

The third case comes from TARBOOSH GAMES. The team has 3 developers and 1 product manager. Usually, each developer develops his own project independently, so that three different prototypes can be developed at the same time. The case I want to see today is “Ballooneers! “.

“Ballooneers! “Was inspired by Fortnite’s balloon video. TARBOOSH GAMES likes this mechanism very much and wants to make it super casual.

Obviously, Fortnite’s original game operation is more complicated. To be super casual, you must first simplify the game control. TARBOOSH GAMES considers several operation methods, including long press & release, sliding left and right or up and down.

At the same time, how to add enough depth to a simple mechanism is also a challenge they face.

In addition, how to join AI and allow players to interact with AI? And how to make the balloon meaningful? These are the main problems they need to solve during prototype development.

The CPI of the first test is 0.15$, the second stay is 35%, the seventh stay is 6%, and the game time is 8 minutes. The advantage of this version is a refreshing gameplay mechanism. There is no similar mechanism on the market before (Voodoo: This also shows that the innovative gameplay can get a very low CPI). In addition, TARBOOSH GAMES also believes that the feeling of rising and taking off helps to obtain excellent CPI. It should be noted that the first test version was developed by TARBOOSH GAMES in only 4 days. We also hope to take this opportunity to re-emphasize the importance of rapid prototyping.

Of course, but this version also has shortcomings: the game lacks enough depth.

In order to increase the depth of the game, in the second version they added the function of aiming-aiming and shooting during takeoff. But there is no change in the retained data, but the CPI has increased. We believe that the reason for the increase in CPI is that shooting is a male game. So TARBOOSH GAMES decided to change back to the original version, emphasizing the feeling of rising and taking off, and abandoning the male shooting function.

The third test focused on level design. At that time, “Tower Run” was very popular, and “Tower Run” had a strong sense of frustration, so TARBOOSH GAMES decided to be in “Ballooneers! “Add more frustration. But the test results show that CPI has not improved significantly. TARBOOSH GAMES found that this gameplay itself lacks depth, and they cannot solve this problem. So, in the end they gave up the project.

In this case, TARBOOSH GAMES shared the following experience:

The feeling of flying and rising helps reduce CPI. We also agree with this point of view, and even all vertical gameplay can help reduce CPI.

Simple control is easier to be accepted by users than dual control. With “Ballooneers! 》For example, the first version only has one click control method, and the test data is not bad. The second version added targeting control, resulting in worse CPI and retention.

Games that make users feel satisfied are more likely to succeed. This not only affects CPI, but also affects second stay, seven stay and game duration.

If the original game concept has limitations, then his game depth may require a lot of level design to improve, or even impossible.

The last case comes from nopact’s “Human Wheel”.

The source of creativity is the fusion of “Katamari Damacy”, “Cube Surfer” and “Tower Run”.

In creative development, nopact’s idea is:

How to come up with a novel and familiar idea in the crowded stacking mechanism pan-parkour game market? Usually we will mention that creativity needs to be novel to stand out in the market. And nopact also mentioned familiarity, because the players are relatively familiar with the gameplay to reduce the player’s learning cost. So they call it a novel and familiar idea. very interesting.

In addition, how to provide depth and diversity through meaningful challenges and interactions? And how to visually convey the relationship between the size of the wheel and the height and length that it can pass? Because letting users understand is the first thing to reduce CPI.

Of course, they also considered how to use Unity Physics to give users a better sense of physics.

The picture below is the first version of the game, which was completed by a programmer in 3 days. It can be seen from the test results that this is a novel idea, and there is no similar idea on the market before this. Growth mechanism, wall climbing mechanism, and hole filling mechanism can all bring satisfaction to users. But the game also has problems, such as control problems (the game does not have a clear track), the vision is slightly monotonous, and the wheels can only accommodate up to 10 people.

Because this idea is very good, nopact decided to put all the members into this project to improve the quality of the game.

In the second test, the game control has been significantly improved, the visual effect is more intuitive, the game has added more levels and tactile feedback, and there is no upper limit on the number of human wheels. The disadvantage is that the game has no area division, no items to collect, no level to end the challenge, and only linear levels. However, it can be seen that this version has achieved a CPI of $0.22, 24% of the secondary stays and 5% of the seven stays. Nopact decided to iterate again, hoping to improve retention and duration.

In the third test, more areas were added to the game, such as forests, mountains, and cities. The game has joined the level ending challenge. In the game, not only can you collect the villains one by one, you can also collect the stacked villains. But the gold coins collected in the level are nowhere to be used, and new innovations are needed.

Judging from the test results, after adding more content, it was still unable to effectively improve the retention and duration, and finally nopact gave up the project.

In this case, nopact shared their gains:

1. Unless you can provide solid innovation, entering a highly competitive market segment is not a good idea;

2. Attractive game mechanics usually show a very low CPI in the first test, even if it is a version made by one person in 3 days;

If the indicators do not improve significantly after adding various major game mechanics, it is likely that the problem lies in the core loop. At this time, you may need to abandon the project and start a new attempt.

The above is the review of the 4 unsuccessful online products shared this time, I hope it will be helpful to everyone.

Source: Roski
Original: https://mp.weixin.qq.com/s/dzJWNALNA9i8hGhCQKAbmQ


Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *