· 11 min read

Mid-Core Success Part 3: Social

Editor’s note: This is the third instalment of a 4-part series written by Michail Katkoff, Product Manager at Supercell. In this series, he explores what it takes to create a successful “mid-core” mobile game. Make sure to check out the other instalments in the series:

Why Social?

Having started my gaming career back when Facebook was the ruling gaming platform for casual games, for a long time I saw social mechanics simply as viral mechanics-levers, which game developers could use to get new users and drive retention. But (luckily) both my perspective and the ruling platform have changed.

Forcing players to connect via Facebook and making them send dozens of invites and requests a day may still work for a few developers, but you’d be amazed how poorly these mechanics fit and work in mid-core titles. So instead of K-factors and virality, I want to write about true social mechanics. The kind of social mechanics that add to the gameplay, improve overall player experience and make the game feel more alive.


What makes a game social is that it allows users to play together – or at least it creates the illusion of it. In my mind, social mechanics should primarily be implemented to improve retention. As discussed in Part 2 of this series, retention is mainly about progress. But progress is a player-specific metric. For example, I might think of myself as being well-progressed in a game after a few weeks, but to others that have been playing the game for months, I’m still a noob.

Social mechanics can help solve the progress measurability issue. When players collaborate in a game, they are bound to compare each other’s progress. Comparing progress leads to two kinds of feelings: on the one hand, players who are clearly behind will want to catch up with those ahead of them; on the other hand, progressed players will feel good about themselves and will not want to lose their position.

The key thing to remember when designing collaboration mechanics is that collaboration between players should take place in an area of the game where players can easily show off. Another thing to keep in mind is that collaboration must benefit all players involved.

Example: Puzzle & Dragons

Puzzle and Dragons relies mostly on one social feature: Helper. There’s no chat in P&D. No player versus player mode. No direct messaging between players. No guilds. And no social network integrations. But every time players enter a dungeon they have to have another player to help them. Due to this simple mechanic, the amount of in-game friend requests is huge.

The Helper feature in Puzzle & Dragons follows the golden rule of social mechanics. It allows players to collaborate so that both parties benefit, and it occurs in an area of the game where players can show off.

Before they enter a dungeon, players have to add another monster (a helper) to their team, lent to them by another player. At the end of the match, both players get Pal Points, which can be used to get new monsters from the Pal Machine. Additionally, the player borrowing the monster also gets Leader Skills, which make monsters more powerful. The more a player logs into the game, the higher the chances of him showing up as a helper for others, and thus earning more Pal Points.

After clearing a dungeon together the player can add the helper’s owner as friend, by sending a friend request. Based on their rank, players can have only a limited amount of friends. The higher the rank, the higher the number of friends they can have. Since a helper can only be used once per owner’s login, having active friends is crucial.

In short, the social mechanics in P&D drive retention by encouraging several logins per day. They also drive players to progress, as the better helper monsters they have to offer, the more they’ll be solicited. Which in consequence results in more Pal Points, hence more monsters.

Most importantly though, the design of this feature takes place in an area of the game where players can show off. I mean, it’s all about how tough of a monster you have, and giving another player the chance to try it out is the ultimate show off. Also, both players benefit from these social mechanics: the player who lends a monster gets Pal Points, and the one borrowing it gets the much needed help to complete levels.

Example: Clash of Clans

We all know by now how Clash of Clans’ Troop donation mechanic works. Once a player joins a clan they can request and donate troops. Donated troops can either help to defend a player’s village, or can be used in attacks. Despite the simplicity of the donation system, it is one of the most powerful social features I’ve experienced.

Donation feature in Clash of Clans is one of the most powerful social mechanics I've seen.
The donation feature in Clash of Clans is one of the most powerful social mechanics I’ve seen.

From the game’s perspective, there’s no set number of troops a player should donate weekly. There’s neither a bonus for donating more, nor is there a punishment for failing to donate. And yet there’s nothing players follow more than donations. By simply enabling users to collaborate and communicate, the game enables players to create their own rules – and with the ability of kicking players off the clan, they can also enforce those rules.

So, in practice, once a player joins an active clan, they have to donate constantly. Active clans tend to set a number of donations each clan member has to make in a week. If a player falls behind without a good reason, they get kicked off the clan. Donation drives retention: as clans naturally demand the best types of units as donations, players do not only need to constantly be training troops, but they also need to progress.

By adding tracking to donation collaboration starts transforming into competition.
By adding tracking to donation, collaboration starts transforming into competition.

Because donation follows the game’s core loop and requires progression, it is a very powerful monetization feature as well. With every monthly update there are new troops and troop levels. Due to clans’ demands, players tend to buy the missing resources (and speed up research times) to finish the troop upgrades. Being the first one in the clan to donate a new unit raises that player to a social pedestal and drives everyone else to hurry up their upgrades.

As with P&D, Clash of Clans also follows the golden rule of collaboration mechanics. Donating troops to other players is – in a way – a show-off, as players can boast about the level of their troops, as well as about how generous they are. In the end, both parties benefit: those who receive them get help in battle, and those who donate improve their status inside the clan.


Creating competition between players is another excellent way to have players compare their progress. The problem with competition designs in games is that most of the developers want to get players into the competition phase too early. The best way, in my opinion, is to have players first enjoying the game, then enable social mechanics by acquiring in-game friends, have them collaborate with these friends, and only after that incentivize them to compete.

Generally speaking, there are two types of competitions: the ones where players compete against each other individually, and the ones where players form groups to compete against other groups of players. Most importantly, when designing competition features, social mechanics should be an important part of the conversation, as they intensify these features tremendously.


If you want to target your top players, the leaderboard is one of the best features for that. Listing players based on their progress, or on how they fare against other players, will affect only a very small percentage of your users. Yet those affected tend to be the most engaged ones, and will appreciate this opportunity to show off.

Group leaderboards, such as the Clan Leaderboard in Clash of Clans, are an excellent and very easy way to broaden the effect of simple player based leaderboards.

The next step is group leaderboards, such as the Clan Leaderboard in Clash of Clans. Not only does this feature affect a larger percentage of players, but putting players into teams and grading them as one will force every player in the team to do their best. By enabling direct communication, such as group chat and direct messages, you basically enable peer pressure, where clan members will force lagging players to up their game, while publicly praising the top performers.

Guild Wars

In all its simplicity, guild wars are timed events between two groups of players, triggered by the players themselves. As with leaderboards, what makes guild wars powerful is the social aspect. Again, peer pressure plays an important role, as communication inside the guild is all about who’s active and who isn’t doing their part in the war.

Holy Wars in Rage of Bahamut is a large scale Order battle where within a very short time frame, order members will engage in battle against each other. Holy War will trigger when the leader of an order declares war. Opposing Order is chosen randomly and battle notification will be sent to all members in the Order.

In addition to being a timed event, guild wars also differ from leaderboards by using nomination techniques to super-engage a few players in the battling guild. By enabling guilds to nominate specific players as leaders, vice leaders, attack and defense leaders and others, you will enable the guild to run more efficiently during the event, as these few nominated players will drive the whole guild to over-engage.


Essentially, raids are like guild wars, as they unify a group of players against a common opponent for a specific set of time. Yet raids differ from guild wars in two ways. First of all, raids are against AI. Secondly, because of this, there tends to be a story element in raids, which is lacking in guild wars. Also, raids tend to award participating players with unique items based on how active they were.

Kixeye has pretty much mastered raid mechanics. All of their live games run periodic raid events in which players are rewarded for the level of engagement they show during the raid. By participating in the raids, players gain access to special units and parts. As shown in the video above, raids are usually heavily promoted, which make them an essential part of the game and storyline.

Just Don’t Force It

I know a lot of people are against my opinions when it comes to social mechanics. For them, social games are all about measurable virality, where social features can be directly tied to the amount of returning players and new installs.

For them, an X amount of invites sent represents a Y amount of new installs. Yet from my experience these un-meaningful social mechanics just don’t work on a long-term basis. After the first couple of spikes in metrics, you’ll need to generate more and more invites for one single install. This leads to an increasing amount of slapped-on requesting and inviting features that degenerate user experience and hinder retention.

What I’m saying is that you should follow a very simple approach when it comes to social features. First start off by giving your players time to play the game by themselves. Let them learn, enjoy it and have fun, then allow them to turn social. Once they like the game and want their friends to play it as well, you can introduce social mechanics that let players collaborate. Collaboration should benefit both of the players and occur in a game area where players can show off. Once players are collaborating, you can start adding the competitive element.

In the end, it’s pretty much all about retention, and social mechanics are an amazing way to improve especially long term retention

Editor’s note: make sure to check out the next article in this series: Maximizing In-Game Monetization

Get game industry insights,
delivered to your inbox