When a game features multiple game modes – game modes that don’t directly intersect and that can be played in isolation – it can be difficult to make sense of things and measure the impact of your changes. Multiple game modes means multiple tuning tables, reward systems and daily routines for your users. It also usually means different types of users. Maybe customers gravitate towards a given mode and non-payers another; maybe your best customers gravitate towards another mode. In order to best manage your game, you need to have a strong sense of who is engaging with what part of the game, and how that engagement fluctuates.

There is both a product and an analytics challenge associated to making sense of a game that features different game modes. The difficult part isn’t having accurate data – the main challenge when dealing with a multimode game is to have relevant and actionable data. You need to ask the right questions and look at things in a way that allows you to assess changes in engagement at a glance. You need to have the right perspective to quickly produce insights that can serve as the basis for tangible decisions, changes and tweak.

Any change in the tuning or progression system of a given game mode can have an important impact on the way players engage with it. But depending on how you approach things, it can sometimes be hard to notice those changes – even when they are significant. There might be a change in the engagement with your PvP mode (for example), but the engagement with the game as a whole might not change much. So, if you only look at your game’s DAU or next-day return rate, you might not see a difference. But maybe your dedicated PvP players don’t play PvP as much as they used to. If you add on top of that the fact that customers and non-payers behave differently (and there are fewer customers), then it will be even harder to notice if your last rebalancing turned your customers away from PvP.

Having different game modes in your game provides variety and increases the overall entertainment value of your offering. However, this also increases complexity. And that means it can be difficult to assess how engaging each game mode is on its own, and also what the audience of each respective game mode looks like. You want to consider each game mode as if it were a game in its own right. What is the DAU of a given game mode, the customer concentration, the LTV of users playing each game mode, the retention/stickiness or next day return rate? You also want to get a sense of how each game mode colors the experience of your users: what percent of your installs have played a given mode within 7 days of install, what is the “engagement retention” of a given mode. If users play PvP once to try it (your tutorial will usually not give them a choice) but don’t engage with it on a regular basis, that’s a problem you need to be able to address.

If you look at each game mode as if it were a game in its own right, you can get actionable insights that can help you tune each individual game mode, assess the best pricing, and best leverage the strengths and appeal of each respective game mode.

 

Looking at multiple game modes

Determining what a game mode is can be subject to interpretation. But overall, the idea is pretty straightforward. On any given day, games like Candy Crush or Bingo Blitz only have one main game mode. Candy Crush features the linear saga map and different levels. But users use a same energy system to play match-3 puzzles with the same gameplay and rules. In Bingo Blitz users can play in a different map, but they are still always playing Bingo. The Spanish accent of the Bingo caller in the Madrid map is a nice touch, but not enough for the Madrid map to qualify as a different game mode. Bingo Blitz features one specific game mode, with similar rules throughout, and there is one same currency/energy loop.

However, not all games are “mono-mode”. Many games – especially RPG games – feature different modes (there usually is a main mode). Sometimes those different modes consume a same energy, sometimes they consume their own energy. Usually what characterizes those modes is that they feature gameplays which are very different from one another, and they constitute a relatively autonomous and self-contained game loop. Most games following the Heroes Charge model feature multiple game modes. For example, SWGoH features multiple very independent game modes. Users play with different teams, against different types of opponents with different game rules and energy system.

Even less “hardcore” games feature different game modes. Sniper 3D Assassin has a PvP mode which has its own energy system and involves a parallel rating and ranking system. Playing that game mode exclusively – or not at all – is an option for users. And since those game modes are independent – and usually have their own monetization loop – keeping track of who is engaging with those game modes is crucial. More importantly, the more game modes you have, the more difficult it becomes to paint a clear picture of what a day looks like for your users.

 

Looking at the audience of your game modes

The first way to look at multiple game modes consists in looking at what percent of your DAU is engaging with each game mode (you should also look at how many missions of each mode are played on a daily basis).

 

This is looking at game modes from the perspective of your entire game. It’s clearly important, but not enough. It’s important to know what percent of your entire DAU is playing with each game mode. But you want to go further and see who is playing each game mode and what form the engagement with each mode is taking. In order to do that, you need to consider each game mode in itself – without considering how it fits into the game as a whole. So, take the amount of users playing a game mode on a given day: that’s the group you want to be considering moving forward here.

In the example above, the DAUs of each game mode are not mutually exclusive. What that means is that a same user can play both the main mode and mode 2. And that user will be counted twice in the above graph – once in main mode and another time in mode 2. The same will happen when looking at mode specific arpdau (a user doing an IAP might be playing multiple modes on a given day). The fact that there is an overlap is not a problem here – because you are specifically interested in considering each game mode on its own. Looking at each metric from a mode-specific point of view (again: by limiting the relevant population to the group of users playing that game mode) will provide you with invaluable insights concerning how each game performs compared to one another – and also compared to the overall game performance.

Looking at things this way does help you identify player trends: what players gravitate towards what game mode. Now one thing which is certain is that users playing multiple game modes will be more engaged than users playing only the main mode. That’s because engaging with multiple facets of a game is a consequence of engagement – not a cause. So, seeing that the LTV of users playing a given game mode is higher than users playing the main mode shouldn’t come as a surprise and won’t necessarily tell you much (but seeing by how much can be relevant). But it will be interesting to compare different game modes. Knowing which game modes attract your highest payers will be critical when trying to price your items or design a monetization strategy that caters to your userbase. You want to price your offers based on what your users are willing to spend – and knowing which game modes attract customers, which game modes your best payer play – will be instrumental to determine the best price point.

In order to assess the performance of your respective game modes, you can basically look at any metric you are already looking at in your game, but apply them only to users engaging with one mode or another. One easy thing to look at is the LTV of your active users – per game mode. Looking at customer concentration is another good way to assess which one of your game modes appeals to your customers the most. Daily conversion rate (percent of users active in a mode doing an IAP) or arpdau are usually very telling.

 

Looking at the engagement of your game modes

Knowing how the next-day return rate varies per game mode will help you assess at a glance what the health of each game mode is.

Looking at a mode-specific level can be crucial. In the example above, you can see a sharp drop in next day return rate of users playing mode 2 (the % of active users playing mode 2 who come back the following day to play mode 2). But the next-day return rate of the main mode remains unchanged. So, this drop in engagement will have a marginal impact on the game’s DAU. Also, in this case, mode 2 attracts a lower percentage of DAU. So, the impact of a drop in next-day return rate in mode 2 will be very hard to assess if you only consider things from the perspective of the game as a whole and total DAU.

Looking at next day return rate from the point of view of each game mode is crucial to manage the live ops of a game. Another important way to consider the engagement qualities of a game mode is to adopt a “cohorted perspective”. In other words, to look at the engagement with each game mode from the point of view of total install. In a previous post I talked about having a lifetime engagement metric. You can apply that same logic to each game mode. You can for example map the lifetime missions in each game mode played per day since install. Or you can choose to look at the lifetime engagement rate with a feature. In this case, you would be looking at the percent of installs who have played a game mode once x days after install. This will help you get a sense of what the user experience looks like and the penetration rate of your different game modes (as well as when users start playing various missions).

 

The retention of a mode or feature is also something you want to be looking at. “Standard retention” will look at the % of installs who will return to the game the day following install, 2 days later, etc. You can adapt that same metric to each game mode. If you consider day 0 is the first day a user plays a game mode, what percent are playing that mode 1 day after the first day they played it, 3 days, 7, etc?

In this case, you can see that the retention rate of mode 2 is much higher than the retention rate of the main mode. That is partially due to the fact that only the most engaged of users ever get to try mode 2 in the first place. The cohorted engagement rate graph above did show few installs ever engaged with mode 2 – so only the most committed of users (and those with the highest propensity to engage with the game and to engage with the most features) will ever play mode 2. But looking at the feature retention – and comparing different modes and features – is something that can clearly provide some insights into the appeal of a game mode. And how that compares with other game modes.

Users will interact with multiple modes in your game. But which ones are they engaging with (and when)? Which ones are driving them back in? Which ones are attracting your customers the most? When you focus exclusively on each game mode in itself – by looking at the relevant metric only for the userbase actually playing a game mode on a given day – you are better able to identify how that game performs. You want to continue to consider each game mode from the point of view of your entire game. But in order to tune your game in the best way possible, and have clear grasp of the day to day changes in your users experience, you need to treat every game mode as if it were a game in itself.