Alexander Shtachenko, the leading product manager of iLogos, told about the most important variables – time, scarcity and frequency in his blog. With the author’s permission App2Top.ru publishes material on its pages.
Dealing with the intricacies of balancing and relying on the works of Jesse Shell, I decided to fix the knowledge I had gained about the theory of balancing free-to-play games using 3 simple graphs. Time, scarcity and frequency control all the numbers in the game – from the economy to the push notification schedule…
The article is written for novice game designers, load balancers, indie developers and analysts who use a heuristic approach in their games. It doesn’t matter what kind of game you make. This knowledge will help you understand what to start from at the beginning of work, what to go for, and what dynamics of complexity should be in the game.
Progress in the game is determined by goals
The main starting variable (located on the X—axis on all charts), we choose what can measure the user’s progress – let it be the level (level, lvl). Next, we define a list of actions that the user must do to increase their level. Depending on the genre, these actions may differ: for a casino — to make 100 spins, or spend 250 chips; for builders — to build 5 objects and produce 10 resources; for a fighting game — to defeat a new enemy; for a three-in-a—row with a journey – to pass a new level and so on. Almost every game consists of tasks — goals, both large and small.
A goal is an action or a list of actions that the user must do to increase progress. To achieve the goal, the user spends game resources, time, money.
For one micro—goal, it may be enough to spend more time, for another — more actions in the game, for the third – all at once. At a certain moment, the user is ready to spend money instead of an action.
Time to reach the goal
The first parameter of the goal is the time of achievement. Since the user is just learning its mechanics at the beginning of the game, his goals should be quickly achievable. The more the user is immersed in the game, the more time he is willing to spend on achieving goals. Accordingly, the time of reaching a new level also changes. The graph shows an example of the dynamics of changes in the time of achieving goals (consider a new level). At the beginning, the user completes the goals in 5-15 minutes, quickly opening the first 5 levels. Then he spends about an hour and so on.
It is important to stop in time, the goal limit is 24 hours — this is one goal per day. If the user is unable to achieve at least one goal per day, then he will probably reduce the number of sessions, and as a result, he will abandon the game altogether. At a long distance, I think it is optimal to set the deadline for achieving the goal — 12 hours, so in one day the user will be able to achieve 2 goals.
Example: Hay Day — Steamboat
The farm, the steamer leaves once a day at a certain time. The order on the steamer is such that your farm cannot fulfill it on its own during the day. The most efficient production is ready to fulfill only 70% of the order. For the remaining 30%, you need to pay either with money or by launching viral mechanisms.
The principle of deficit formation
The second parameter of the goal is the deficit. As can be seen from the graph, we have determined the production volumes of resources for each level and now we can form a deficit — a shortage of resources to achieve a new goal. Thus, we determine the number of actions (and their equivalent in money) and put the deficit in the goal.
For example: to reach a level, we need to complete 3 approximately identical goals. One of the goals is to harvest 17 apple harvests in the garden. But the trouble is, only 5 apple trees have been planted. The user can collect only 5 harvests in 1 session. In the third session, the user performs the other 2 of the 3 goals, and for the 3rd, with apples, he lacks 2 crops (5 x 3 = 15). It seems to him that the new level is already so close, and here are these 2 crops. “If I wait for another session for the sake of 2 harvests, then other resources will be collected aimlessly, and I will get new content later.” A good motivation is to pay 2 bucks and get my portion of emotions right now.
How to estimate the deficit in money?
Question — how much should these 2 apple crops cost? Let’s plot the time to achieve the goal under the X scale on the graph. Newzoo research shows that players in the United States are willing to spend about $ 10-15 per game per day. Thus, we allocate this amount to the deficit and get the dynamics of the growth of the cost of the goal in dollars relative to the time of reaching the goal.
If the apples ripen in 2 hours, then the deficit of the goal of 2 apple harvests should cost no more than 2 dollars.
Hunter or prey
If each time a new goal requires a lot of effort, then the user will always feel like a victim, since moments of happiness will never come. There are several statements based on behavioral science and psychology that explain why a person needs a balance swing, but I will not dwell on them now.
Let’s change the line of needs so that sometimes the deficit is zero, and sometimes it goes through the roof. Thus, the user will have moments when he will dominate the game, and moments when the game will require additional resources.
It is logical to expect a higher dump of players at levels where the deficit is off the scale. By managing the deficit, we can influence the dump without allowing a drawdown in money.
You have noticed that on the chart the deficit of the first 4 levels is zero. This is no accident! The tutorial, activation and engagement should last exactly as long as it takes to bring the right proportion of users to the first moment of difficulty in the game. If you want to entertain the user for a week, and only then ask him for money, then the deficit will be zero to 10+ levels with such a time grid.
Managing goals
We have sorted out the target completion time and the deficit. The third component of the goal is the frequency of actions. Suppose that in order to achieve the goal I need to do 3 sessions — like in the story with apples. But for a new level, I need to complete this goal three times. After each achievement of the goal, I can plant 5 more apple trees, thereby reducing the number of sessions to achieve subsequent goals. Having received the long-awaited level, the goals change, become more difficult according to the deficit schedule. I no longer need 15 harvests to achieve the goal, but 45 — and again 3 sessions…
Another example is that a goal consists of several different actions that cannot initially be performed in one session. As in Hay Day — plant wheat, harvest wheat, make food for chickens from wheat, feed chickens, chickens will lay eggs, and eggs can be made into an omelet. The goal is to cook 10 omelets.
By controlling the frequency of actions, we determine the number and length of sessions, set timers, send push notifications, and adjust the complexity of the game.
The fourth control parameter is the number of goals per level. As you can see on the graph — the number of sessions increases by one in each new level. But you can set this parameter at your discretion so as to get the most profit.
The following graphs show an example of the dependence of goals on several resources. If your game uses several types of resources, you can alternate the deficit, combine it depending on the level. You can also change the resource deficit for each goal within the level, repeat goals and conduct a lot of experiments designed to increase LTV.
The theoretical part, which I have tried to describe as simply and transparently as possible here, should help you, as it helped me in drawing up an asymmetric balance.
Examples of balance with multiple resources:
- To reach a new level, I already need to harvest 15 harvests of not only apples, but also strawberries. Everything is OK with apples, and I’m starting to invest in strawberries. After a level when the production of apples and strawberries is established, there is a new shortage — potatoes)))
- To reach the level, I need to defeat the opponent by increasing the damage of the sword. At the next level, the damage of the sword is enough, but it is necessary to improve the armor. After another level, you need not only to improve the sword and armor, you also need to pump magic…
- To win the race, you need to improve the engine, then change the brakes…
and so on indefinitely.
Total: what to remember when balancing
- Determine how the progress in the game is measured (for example, the level)?
- What goals (actions) should the user achieve to increase progress?
- How much time should the user spend on each goal in the game? What is the dynamics?
- What is the deficit of each of the goals?
- How many level goals does the user have?
- How many sessions does the user have to do before reaching the goal?
P.S. By the way, I made a mistake when balancing. Starting initially from Consume, I customized Production. On large levels, this method stumped me and I had to recalculate everything. Don’t get caught. It is easier to determine your production, and already lay a deficit from it.
Source: ProGamedev