Deconstructing Empires & Allies

Empires & Allies (E&A) is the mobile reboot of Zynga’s facebook title by the same name. The Facebook game attracted an impressive 39 million players within its first month (June 2011), but was shut down less than two years later. The mobile game is real time rather than turn based and features realistic graphics as it’s key selling point.

This is a big title for Zynga, the released the same day it announced restructuring that would make about 18% of its workforce redundant. The game has been in soft launch in various countries for 9-10 months according to the company, but after playing 3-4 sessions a day for over two weeks now, I have serious doubts whether the game will live up to expectations, despite some initially encouraging figures.

Game loop

The core game is very similar to Clash of Clans and Boom Beach, so much so that even sketching out the game loop is redundant. Players build a base, farm resources to make troops and raid other players bases for more resources and glory. Like Boom Beach troops are deployed in small groups rather than individually, and the player has a small amount of control of their troops during battles via various special abilities (A-10 bombing runs, setting waypoints, etc.). A small number of PvP opponents are placed on the world map where the PvE campaign happens, but players can equally tap a quick match button to dive straight into the action.

The mix of mechanics from various mobile RTSs feels confused and in many cases works against itself. For example the game gives the player walls to help defend their base, at contrast with the modern day setting. As well as being fiddly and annoying to place, these encourage players to group their buildings together – a poor tactic considering the area effect of the various bombing run special abilities.


The special abilities also offer attackers a considerable advantage – one that Boom Beach offsets by giving attackers only a limited space to deploy their troops (on the beach). E&A recognises the benefit of doing this in the PvE campaign, where it happily (but arbitrarily) restricts the attackers deployment zones. In PvP however, players can deploy their troops wherever they like, which is a huge advantage when combined with the ability to take out key defences remotely, especially if grouped together with special abilities. As a player my attack power vastly outperforms my potential defensive power, with the result that I have completely neglected developing defences that I know to be totally inadequate against players my own level.


Economy wise, E&A has four main resources that come into play one after the other in a similar way to Boom Beach. Whilst the player is given just one builder, also like BB, hard currency gold is given out sparingly in exchange for completing achievements, closer to Clash of Clans. This combination of a single builder and little hard currency is a serious error in my opinion, and has led to over a week (at 3-4 sessions per day) when I have had little incentive to attack, and consequentially little fun.

In Boom Beach the single builder works because they give out a lot of gems to speed up building. Where you spend your gems is part of the game, even for a non payer, and considerably accelerates the otherwise tedious period when you need to upgrade lots of buildings on short timers. In Clash of Clans having two or three builders not only provides an obvious first purchase (and then creates a shortage of resources later), but makes brief sessions satisfying as the player can collect their resources and set several upgrades in progress at the same time. In order to have the resources for this players need to start raiding regularly and so the game hooks players into its loop.

In E&A players have neither the gold they need to speed up short timers and recycle their builder more frequently, nor the lack of resources to prompt attacking. Many of my sessions are extremely short and unsatisfying, consisting of collecting resources and then setting a single upgrade in progress. This cycle has continued for 7-10 days straight without a significant increase in power for my base or my troops. Hardly inspiring, engaging stuff.



Perhaps one of the most disappointing aspects of E&A is the alliance system. At first glance it looks promising, offering private as well as communal chat, the ability to have friends as well as alliance members, and really nice feature that gives other alliance members resources whenever you successfully attack someone, giving a good incentive to be in a large, engaged alliance.

However, becoming a member of an alliance was an incredibly frustrating experience, drawn out over the course of several days. No alliances are suggested to new players, and the only way to search for alliances is by name – a daunting task for a newbie player if they have no friends playing the game. I was shocked that there was no way to find local alliances or alliances with players of a similar standard to myself. After days of randomly sending out join requests to alliances that came up when I searched for “A…” or “B…” I finally set up my own alliance and dropped the details into the design chat at Wooga – 50 odd gamers generally eager to try new games if only to check them out before they decide whether or not to commit longer to them. For me this was a modest success, and I eventually ended up with an alliance numbering 5 players, but for players that don’t have access to such a ready pool of alliance members, I wonder what they would do.

“Getting people into an alliance is more important to us than getting money from them.” – Gabe Leydon (CEO Machine Zone)

Alliances are such a vital driver of retention that driving players to join one should be of paramount importance to a game. In Game of War Machine Zone admit that players who don’t join a guild within the first two or three days have nearly zero probability of sticking with it longer. In E&A the game seems to actively conspire against players joining an alliance, in stark contrast the games that it is competing with.


In my opinion, the graphics and marketing of E&A are the make or break factor. They are far more realistic than most other games in the genre, and the use of iconic units such as A-10 Warthogs definitely makes the game stand out from the cartoony Supercell games.

Whether or not the graphics allow the game to stand out on the App Store is another matter. The added realism of the game detracts from the ability to place identifiable characters in promotional graphics. There is no Clash of Clans Barbarian or Boom Beach Heavy to be a front man for the game. The result is a look that feels both eye catching and generic at the same time.

As I’ve described above, E&A lacks the polish and synergy of features that it is up against in CoC and BB. It also has exactly the same monetization strategy and features, which can only lead me to believe that the Life Time Value (LTV) of players will struggle to match these games. Without an installed base of players, together with clan wars and events to drive long term engagement, it is likely to be significantly lower. Of course these features could be added later, but this will take time, and E&A will need to demonstrate that it is worthy of committed investment by Zynga first.


So the question becomes what is the Cost per Install (CPI) of players? Do the realistic, but generic graphics appeal to a new set of players that is currently untapped by the incumbents? If so, are there sufficient numbers of these players to bring the CPI significantly below what other games are paying? If so, then the game could be profitable on a CPI vs. LTV basis despite its various flaws. If not then other games will be able to outspend it on a per player basis, driving up the CPI to a point where it exceeds the LTV, and Empires & Allies cannot recruit new players on a profitable basis. In this scenario, even if the game attracts a number of players initially through featuring, its player base and revenue is destined to gradually decrease.

For me this reinforces the importance of product marketing in today’s mobile market. Making a fun game is not sufficient to ensure success. Games have to be able to drive their LTV up with a robust monetization strategy, and drive their CPI down with a distinct and attractive marketing message. This is the only way they can carve out space for themselves against heavily entrenched incumbents that have vast installed player bases with considerable time invested in their games, and huge amounts of money to throw at recruiting and retaining players.

Zynga could have tested its market positioning for Empires & Allies even before starting production. The question is whether they did. If the basic positioning is strong enough then the design flaws can be ironed out and new features added to further improve upon the game, giving it the prospect of sustained success. If as I suspect the graphics are not sufficient to significantly lower CPI, then whatever initial success the game enjoys due to featuring and paid marketing will soon fade away. For now it’s too early to tell, but the download and revenue figures over the coming weeks will soon make it clear which way the game will go.

This post was originally published on Adventures in Mobile Game Design.


Harnessing the Psychology of Gifting

gifting-etiquette (1)

I feel like gifting has a bad name in games. Like the term “social” it has been ascribed to Facebook games that often implement interesting features in unexciting ways. Understand the psychology behind gifting in general – the phenomenon that is ubiquitous to all human culture and almost every human interaction – and I think things get a little more interesting, as well as suggest features that would work better than the average Facebook game.

Gifting and exchange are a component of almost every social interaction. This is even apparent in the way you greet your colleagues in the morning. If you see someone for the first time in the day that you work with then you are obliged say hello. They in turn are obliged to acknowledge your greeting and respond. If you barely know each other the exchange is short and perfunctory: “hey”, “hey”. If you know each other better, it might spawn a longer conversation or at the very least require more conviction.

Even this mundane situation exhibits what anthropologists know as the three obligations of gifting:

  • The obligation to give (to say hello in the first place)
  • The obligation to receive (to acknowledge that greeting)
  • The obligation to reciprocate (to say hello back)

In such a minor social interaction the exchange is fast and low value. But even so, to omit any of these steps would be rude. If a colleague continually ignored your greetings, then you would think less of them. Let’s examine the three obligations in a little more detail.


Giving gifts or initiating exchange is an action that is required in certain situations. In western culture we buy each other gifts at Christmas and on birthdays, and we take a bottle of wine or contribution to the meal for a dinner party. Failure to give at these times can be a serious faux pas depending on how well we know the other person.


When someone offers you a gift you are obliged to accept it. To refuse a gift is to be unfriendly, if not rude. To turn down an invite to dinner for no good reason would be frowned upon, to refuse a Christmas present from a family member almost unheard of.


There is an obligation to repay all gifts. In some cases this is more immediate and calculated than in others. If you give a friend a birthday present then you might expect one back from them, and Christmas gifts are often a minefield of social obligations to find gifts that correctly reflect the value previously received and current state of the relationship between two people. The tighter the relationship between two people the more that the value and timing of gifts becomes varied – both parties expect that everything will work out in the long run.

Indeed to repay a gift too quickly or exactly and remove the implied social debt is as unfriendly as not returning it. If your friends help you move house you deliberately repay them with something where the value is hard to calculate, such as cooking them a meal, rather than paying them cash. Relationships typically start with small one-off gifts of time or effort and gradually extend into a continuing cycle of reciprocity. In extremis, almost all major religions advocate giving to charity for benefits in the afterlife or from the universe in general. We have an innate belief in karma of one kind or another that is hard to shake.

Now we’ve gone over the theory, let’s look at gifting in three games in particular to see how it’s applied:
1. Pearl’s Peril is a Facebook and mobile hidden object game with fairly typical energy and item gifting. (Disclosure: I’m currently game designer on this game)
2. Clash of Clans gives you the ability to donate troops to your clan members when they request them.
3. Animal Crossing has a sophisticated gifting system with both other players and NPCs.

Pearl’s Peril

Pearl's Peril

As in many Facebook games Pearl’s Peril allows you to send energy, soft currency and low value collection items to your friends. In each case it doesn’t cost you anything to send the gift – the gift is created out of thin air during the gifting process, and the process itself is incredibly streamlined, allowing gifts to be sent from multiple points in the UI and delivered to the recipient’s inbox where they are accepted in a click that is barely noticed coming into the game.

The result is a social feature that is rather like greeting a colleague each morning. In interviews players describe it as a habit that they fall into – almost like a pleasantry of saying hello. It’s a valuable feature because they value the energy, but the gifting act itself is less so. Players do not feel obliged to log in just to send gifts to their friends and maintain the cycle of exchange.

Clash of Clans

Clash of Clans

In Clash of Clans you can donate troops to your clan members. You have to spend resources to train these troops yourself, and you could use them in your own attacks if you didn’t give them away. In many cases the only value that gifting affords you is allowing you to take a few extra troops into battle: those troops stored in your clan castle and donated to you. It does however give lower level players the opportunity to play with much more powerful troops given by other members of their clan, foreshadowing units they will have in the future if they stick with the game. It also requires some degree of coordination so that you get troops that support your style of play, which stimulates discussion in the chat.

The system is much more powerful than energy gifting in Pearl’s Peril. Clan members can see how many troops they have donated vs. received and compared to other clan members helping them keep track of a number of different relationships and making sure that gifting doesn’t get too imbalanced. Donating troops was really the only thing that clans allowed you to do originally, but the system has been further reinforced by clan wars, which provide an obvious occasion to donate troops.

Players that fail to keep up their side of the bargain – not giving as many troops as they receive, or not giving troops when they should – face being kicked from their clan to make room for more committed players. Higher ranking clans often state these obligations and enforce them rigorously. At GDC this year, Supercell stated that the 2 year retention for Clash of Clans is 10%, and I believe that the gifting economy that they have created in the game is a key component driving this.

Animal Crossing

Animal Crossing

Anyone that has played Animal Crossing will tell you it is a magical game that creates quite a unique atmosphere. Dropped off in a village of anthropomorphic animals you can gather fruit, fossils and flowers, catch butterflies and fish, design your own clothes and furniture and develop relationships with the NPCs and other players in your village. There is no real goal, but players typically spend their time collecting things and exploring the world, which changes season by season and through a day / night cycle.

It is perhaps the game that shows the best giving mechanics that I have observed. The calendar gives you a natural context to give things due to birthdays, other occasions or simply because you found something you know they want or don’t have space for yourself. NPCs also prompt you to gift on a regular basis, by asking you for things, letting it known they are looking for particular items, and occasionally sending you gifts themselves.

Most of the activities in the game drop items on a semi random basis, with some items rarer than others, but the selection constrained by the time and place that you are collecting. This makes gifts unique items, with each one requiring time and effort to find in the game world. In contrast to the commoditised resources that are given in Pearl’s Peril, you know each gift is special.

Indeed even sending the gift requires effort as you need to find the recipient in the game world or visit the post office to send it to them. Even if players do have multiple items that they want to give, each one needs to be given individually, rather than sent off as a bulk action. Similar to sending someone a hand written note over an email, it enhances the sense that the donor really cared about giving something to you.

The result is a powerful system with real emotional weight behind each gift. Gifting is one of the key systems that runs throughout the game and gives it such a unique and magical feel. The series consistently gets superb ratings from critics and is listed as one of the best selling games ever with c.27m copies sold across 4 titles. However the real demonstration of how effective this gifting system is comes from anecdotes like this one, about a mother with multiple sclerosis’s gifts to her son. It’s impossible to imagine either Clash of Clans or Pearl’s Peril creating this kind of story.


Gifting and exchange are ubiquitous human behaviours found in all cultures and a huge variety of situations. Gifting consists of three obligations:

  1. To give in certain situations
  2. To receive gifts offered
  3. To reciprocate gifts received

Increasing the value of gifts increases the emotional engagement and social obligations that players experience. Increasing the power of gifting can be done by:

  • Making gifts unique rather than commoditized
  • Making gifts require more investment from the giver
  • Making the giving process itself require some effort
  • Prompting players to give in clear situations

Great gifting systems support strong long-term retention, player satisfaction and by extension commercial success.

This post was originally published on Adventures in Mobile Game Design.

After the Gold Rush: Competing in today’s App Store

It’s hard to believe that the App Store is only six and a half years old – it was launched in July 2008 with just 800 apps. Now there are 800 apps downloaded every second – 2 billion a month – and the number of apps available has grown more than 1,000 fold to 850,000.

Games dominate this marketplace, with more than twice the number of apps as the next biggest category, and well over 10,000 new ones being added each month. The barriers to entry remain low – you can code and release a simple game on your own – but the barriers to success continue to get higher. This is typical of all markets as they mature, but it is striking just how fast this has happened in mobile, and developers of all sizes are finding the competition fierce. Beyond the sheer quantity of games available, there are three clear indicators of this development.

Firstly, the top grossing charts are largely static. Candy Crush Saga and Clash of Clans have been in the top 3 games for over 2 years now. These games appear to have locked down their respective genres in the way that Call of Duty or World of Warcraft has on other platforms. Many of the other games in top 10 are similarly long lived, and just two developers: Supercell and King consistently account for at least half of the top 10. Games that do manage to break into the upper reaches of the charts are notable because they are so rare now, and often supported by very strong brands, such as Kim Kardashian.

Secondly, the cost of user acquisition is rising steeply. SuperData estimated that the cost of acquiring a user increased 37% between Jan 2013 and Jan 2014. Machine Zone’s recent $40m advertising campaign for Game of War illustrated just how much money the top companies can throw at marketing. SuperData also reports that CPI now stands at an average of $2.78 for mobile games, whilst average revenue per user is just $1.96 – not a good ratio for developers.

Thirdly, the production values of games have increased dramatically. Super Evil Megacorp spent two and a half years developing Vainglory, and it showed – the graphics looked closer to AAA standard than what we would normally expect from a mobile game. Smaller developers can pull off great looking games such as Monument Valley or Badlands, but only when they choose very stylized appearances that facilitate lower costs of production. Even the UI transitions in Hearthstone demonstrate a level of polish that few established studios, let along Indies could hope to pull off.

So where does this leave us? How to we compete in a market that has become this tough? It’s a question that we ask ourselves a lot at Wooga, and even with our current successes (Diamond Dash, Pearl’s Peril and Jelly Splash) something that we are still working out.

Looking at the charts it is clear that cloning games does not lead to success – the only game with similar mechanics to Clash of Clans is Boom Beach, also by Supercell. Replicates of Candy Crush Saga have performed similarly poorly. It is for this reason that we do not clone games at Wooga, and never have. It’s also creatively unsatisfying and for a combination of these reasons many people advocate the opposite end of the spectrum: radical innovation in the hope of striking it lucky.

This was my approach on my last game too. Whilst Wooga has a history of casual, single player games, I set out to make an action strategy game. I loved the genre myself, and felt there must be an audience who were likewise unsatisfied by the current offerings on the App Store. I felt that given the talent at Wooga, our understanding of game design and the amount of user testing we used in the creative process we could make a success in any genre. But in hindsight this approach seems just as misguided as cloning games, as I had failed to recognize the current state of the market and the value of building on existing company expertise, tools and audiences.

Games are complicated systems, especially action strategy games designed to give years of play. We had a good prototype and the gameplay was novel and fun. But the more we worked on the details the more problems we threw up. Exactly because the gameplay was novel, we needed novel solutions to these problems – we could get inspiration from other games, but no one had solved these exact problems before. The team did a great job of working through these, but it took a huge amount of time and emotional effort to be continually rebuilding large sections of the game.

Furthermore, as we worked through design issues, we realized that we were left with a number of risks that we could not remove before launch. By staying true to the vision of the game we had ended up with intense synchronous PvP gameplay that only Hearthstone and World of Tanks came close to. These are successful games, but it seems largely because of their existing PC audience. We also started hearing horror stories about the CPIs for the mid core audience, several times higher than the best LTV any of our existing games had. In combination we were not sure if the audience we were targeting existed, and if it did whether we could profitably reach them.

Despite the team’s enthusiasm for the game I could see that this was becoming a passion project for us rather than a legitimate chance at creating a commercial hit, and unfortunately for those of us that make games professionally, commercial realities cannot be ignored. Eventually I decided it was time to stop burning time and money and start on something afresh with greater potential. I realized that we had been arrogant enough to assume we could build something as good or better than what was already out there, despite having none of the tools, none of the design knowledge, and no audience compared to developers that were already making these sorts of games.

I remembered that Clash of Clans was an iteration on Backyard Monsters and Candy Crush Saga a refinement of Bejewelled. But even before their big successes, Supercell had experienced teams working on Hay Day and Clash of Clans, and King had its casual gaming portal and picked games out of that to develop for Facebook and then mobile. In each case the companies had existing assets that they built on, as well as refining mechanics from other games.

A company with the size and track record of Wooga has plenty of assets to build on. We have an existing audience, established IPs, game design experience earned through hundreds of user tests and millions of players on our live games, the know how to set up extensive content pipelines and so on. We hadn’t recognized their value before though, and instead choose to innovate across the board. Our core gameplay was different, our elder game novel, we were attacking a more core audience we knew only as players ourselves. The only asset of Wooga that we really used was building on technical expertise from previous games. In our post mortem we realized this was also the one area that was free from major problems over the course of the project.

Let me be clear though, that when I talk of innovation, I am not talking solely about design. I am talking about all aspects of production from the audience and genre you choose to the technical setup you have as well. Design can be innovative and successful, but innovative designs are best supported when you build on existing assets elsewhere. Hearthstone is a great example of this: the team built on a vast wealth of assets that Blizzard had to make the game a success. There was the experience of the design team, the strength of the Warcraft brand and Blizzard’s ability to recruit half a million fans whilst the game was still in beta. The gameplay was fresh, the genre basically non existent on the App Store, but even from the outset the risks that the team faced were greatly reduced by building on Blizzard’s strengths.

I also believe that this is the case whether you are a mid sized developer like Wooga with 280 employees, or an indie developer with a fraction of that scale. The key is to recognize the particular skills and assets that you have, and the niches where you can best apply these. Smaller companies may need to choose small niches to succeed, but they also need smaller successes to cover their costs. As a smaller company you can serve niches that aren’t big enough to warrant the attention of bigger players. As you serve that niche you should aim to build up tools, marketing channels such as email lists and other assets that aid your future work and thereby grow your business.

Learning from my recent experience, for my next project I will work on something that builds on a far greater number of Wooga’s assets. This is the best way for me to maximize my chances of seeing my game launched and successful – a goal that I am sure many developers can sympathize with. This might sound like a dispassionate approach to making games, but I don’t feel that it needs to be. Like most people in the industry, I work in games because I love games. Building on the assets that Wooga already has may seem constraining given the wonderful variety of games that exist. But it is an additional constraint and not the definition of an entire project, and as the saying goes, creativity loves constraints.

This article was originally published on Gamasutra