Zynga and the End of the Beginning

Where two years ago everyone was talking about 'casual' games, now they're all talking about 'social' games. Key developers have recently attracted some very big numbers. This article is not really about Zynga itself, but rather examining what underpins their business model, the likely threats to which it must adapt and how Zynga – as standard bearer of the social game community – will likely fare in the coming year.  As Zynga goes, so the rest of the social game market tends to follow. 

The first thing to say is that the people running Zynga are both very smart and competitive. They have streaked ahead of all of their competition by applying a relatively simple strategy of picking up on gaming trends, copying them quickly and then maximising every avenue of Facebook to spread their message thoroughly. Zynga currently has 4 times as many monthly active players in their games as their next closest rival. To look at the distribution of players on an Appdata.com chart, you would be forgiven for thinking that there was an error in the metric reportage, such is the disparity. 

It's also important to understand something about 'social games': Most of them are not social. They tend to be single or multi-player games that use social networks (mostly Facebook) as an easy way to drive player adoption. What the industry is calling 'social games' are more accurately described as 'viral games'. 

The focus of most viral game developers is maximising trends. Trends rise and fall quickly in response to player boredom, retention is king, and developers spend much of their time reminding players to play, to invite their friends, to post stories from the game to their profiles, and other activity designed essentially to not let the player forget to come and play. Viral gaming relies a lot on ways to grab or nudge players’ attention. Like any third party game publisher they are reliant on the benevolence of their platform holders (primarily Facebook) and the market conditions that their platform has engendered. 

This has resulted in predominantly short-term thinking. Viral game development is a battleground of very simple and usually cloned games, interruption marketing tactics, push-to-the-limit tactics to jog players into returning to play, and a lot of scrambling to be on the next trends as fast as possible. Viral game developers, such as Zynga, have little or no commitment to developing deep or rich game experiences because the market has not really rewarded that kind of activity. However that lack of depth is precisely the reason why viral gaming is showing signs of weakness typical in any runaway success. 

Zynga this week received investment of $180m from DST, a Russian venture capital firm (which also owns a small share of Facebook itself), and this signals the end of something and the beginning of something else. The big question is this: Is it the beginning of the end? Or is it the end of the beginning?

Product Differentiation (or Lack Thereof)

“Let me point out to you guys: There’s got to be some reason why FarmVille has grown to 28 million daily active players and the next one, Farm Town, has five million and is not growing, right? If all that we were doing is everyone was copying each other then why is it that some are growing and some aren’t?”

Mark Pincus in a recent television interview on MSNBC’s “Press:Here”

What’s interesting about Zynga and their competitors is that there is almost no differentiation between their games. This has happened gradually over time, with companies originally starting out with their own unique rosters of games and slowly amalgamating their catalogues to the point that they are carbon copies of each other. Whether poker, pets, fish tanks, farms, or mafia simulators, each game type is replicated almost identically. 

There are some subtle variations, but they don’t really amount to anything of significance. There are no viral game developers that have next-generation or revolutionary game designs that they alone wield. Mark Pincus’s quote above suggests that Zynga’s games must have an inherent difference that other similar games do not but they really don’t. 

Secondly, all of the main developers have a “social bar”, which is the technical term for a set of links displayed above or beside each game to cross-promote players into other games. All the major developers have such a system in place, usually at top of each game page or (in Playfish’s case) to the left.

So that means the only real differences between the viral developers are: 

So the secret to Zynga’s success, and the model that all the other developers practise to a lesser degree, is as simple as this:  

Zynga has way more players than their competitors because they spend more to get them, and when they have them they constantly remind and reward returning behaviour, and lastly kill any channels that just aren’t working. 

It’s a very un-subtle strategy but entirely appropriate for the landscape and the platform that Facebook created, and the only reason that Playdom and Playfish have not kept pace is smaller or no advertising spend and an unwillingness to exploit virality to its fullest. Zynga are simply more competitive and they use what amounts to a very successful ad-spam strategy to buy player attention in any way that they can.

But Zynga’s model can only continue to work under these conditions:

Kick enough of those legs out from under the Zynga table and they – and their competitors who have the same mindset that they do – could well be in for some tough times. It just so happens that this is exactly what’s about to happen. 

Changes: Opting In vs. Opting Out

There are two important parts to building any successful viral application: It has to spread and it has to be retained. Spreading without retention is the typical path of quiz applications and strictly casual games. They tend to fly up the charts and subsequently decline just as quickly. On the other hand, applications that don’t spread but retain well tend to be boutique games like Battle Stations, Vikings of Thule or Tennis Mania which have small and loyal audiences but are unlikely to ever be blockbusters. 

Getting an application to spread requires novelty and is fashion-driven. There’s no point making a farm game these days because everyone is sick of farm games and a new entrant will find it hard to spread. Playfish’s Country Story only has 10% of the audience size of FarmVille and is in decline, and their Poker Rivals game is performing very poorly in the engagement stakes (9% DAU to MAU compared to Zynga’s steady 24%). Both simply came way too late to the party, and I suspect their forthcoming Mafia game will not go far either. 

Getting an application to spread also requires interruption. Zynga have demonstrated with Café World, Fish Ville and now Pet Ville an uncanny ability to build millions of players in only a few days. How? They cross-promote from other games and buy large blocks of advertising to spread the word. This works best in the current environment if you have the money to do it.

Retention is then the prime opportunity to monetise, but also a major source of sustained growth. The prime techniques for achieving high retention are:

Most of these systems are based around an opt-out structure. A player can choose to block them but in the majority of cases the players are more likely to simply ignore or put up with them rather than actively deal with them. That may lead developers to think that that is fine, but when opt-in systems are over-used to remind players, they leave a sense of poor experience and a feeling of being the victim of spam. And that reflects badly on Facebook itself.

Facebook’s solution is to replace most of these opt-out systems with opt-in equivalents. They are allowing players to have much more control over whether applications gain permission to contact them at all, which means that applications cannot spam their way to success. This change, combined with another update of the Facebook home page design, is really very significant. Going back to the list above, here’s what’s changing:

Additionally, Facebook have promised to step up their compliance policing significantly to make sure that developers are not breaking the rules, and banning applications that do either until they are made compliant or permanently. 

 So for Zynga the real issue here is that they cannot really nudge players to return to applications nearly as easily as they used to. This is a big problem because they, and their competitors, are not sitting on catalogues of games which are good enough that they naturally encourage players to return and play them of their own accord. 

Viral gaming up until this point has largely been a game of distribution plays rather than content plays, which is why the developers don’t really spend a lot of time on the depth of their games. In online gaming such as massive multiplayer games or first-person shooters players do not need nudging to come back and play again and again because the experience of playing is so good that they choose to return. No Facebook game comes anywhere close to offering that. They are closer to idle distractions. 

Games of that nature simply do not register significantly with players and opting-in to receive information from any kind of product or service tends to make players much choosier about who can contact them than opting-out does. 

Population: The End of Endless Growth and the Rise of Veterans

Facebook continues to blow away the competition in country after country, adding 0.5m players a day across the world, many of them game players. Mark Pincus has called this a social revolution based on a new spirit of playing together, of finding new friends and new connections, making new shared memories, building social connections and so on. All of these are laudable statements, but most of them are fantasy. Players don't, on the whole, play Mafia Wars in a social fashion, nor do they play Restaurant City in the mould of family board games. Viral gaming’s success is much easier to explain: 

It's new.

What Facebook enables is the introduction of web gaming to normal people for the first time. This is why relatively ordinary game concepts like poker, virtual pet simulators, farming games and simple role-playing games have managed to penetrate so far. Taken in context, they're all as brand new as the equally simple Wii Sports was when Wii first launched. In any completely new game environment, unsophisticated games tend to rule the roost because players spend time being delighted by the strangeness of it all. 

The problem is that delight fades, and nothing stays new forever. Novice players start to develop expectations, and become veterans. This effect applies equally to hardcore gamers, poker players, casual gaming housewives or any segment of game players except young children. 

Viral game developers behave as though there is an endless supply of novices, but of course there isn’t. More accurately, there is due to population renewal, but the replacement rate of veterans with novices in stable markets is usually a lot lower than we’re currently seeing in the Facebook market, and that’s all because of Facebook’s meteoric growth. 

Games Workshop, Nintendo, Mattel and Hasbro are examples of companies that have figured out how to manage a stable, renewing population of players. They've done so by finding or developing great, lasting games that build reputations that spread across the generations by encouraging veterans to initiate new novices.  

Zynga’s games are not being built with renewal in mind. They are experiences built to appeal to novices and they inherently assume that novices are all there is. So they are simple to play, undemanding, lack challenge or consequence, and rely on time-oriented tasks. They may not realise it but Zynga, Playdom and Playfish (and others) are quietly educating millions of novices to expect more and then not delivering it to them. 

Facebook's growth rate is slowing down. It will likely hit 400m players by next March but may well never reach 500m players. So the number of novices who are impressed by very simple applications is going to fall and the number of expectant veterans will rise. The veteran mindset is more demanding, sophisticated and selective.

Veterans diversify. They have less interest in playing what everyone else is playing and more interest in playing the perfect game for them. This is fatal for a market in which the content is completely undifferentiated, as viral gaming is. Game developers relying on mass audiences will increasingly find it difficult to compete with identikit software. Veterans will feel that they have already played such games to death. They will want something more. 

Veterans are also choosy sharers: When people first started to use Youtube they used to share every video they discovered. Now, Youtube users and their peers have become used to the everyday Candid Camera clips that litter Youtube, and so they want video that is worth their attention. What was fun at the novice level is boring to the veteran. Novices are thus most likely to share gaming content to all of their friends where veterans will only do so selectively. 

Veterans share as a means of expression and identity. What you share says something about who you are, and so the risk of bad sharing is that of damaged reputation. Few veterans want a reputation as a spammer. So they no longer pass on every Youtube clip that comes their way to all of their friends. Instead they share selectively to groups and individuals that they think will really like the shared item. It takes a Susan Boyle moment to overcome that kind of filtering.

Advertising: The Saturation Problem

Zynga has assembled the largest player base of any viral developer partly through acquiring customers via on-platform advertising. Reportedly they have spent at least $50m doing so. In most contexts online advertising is ineffective and the Facebook environment has stayed true to form. Zynga overcome this with sheer spending power: Even the most ineffective advertising does get at least some engagement, even if it is only 1%. 

Online advertising tends to attract novices who don’t easily distinguish between platform and advert and tend to be more trusting. Veterans tend to unconsciously filter adverts away from their attention unless they are well targeted and personally relevant. As a result, entertainment (such as games) is particularly difficult to sell purely through advertising unless you plan to use mass advertising at a colossal scale (as Hollywood has proven).

Products generally appeal to players on the basis of solving a problem. This usually means a functional problem or a lifestyle problem, everything from fashionable socks to cheap flights to New York. This means that they can target and establish personal relevance based on like-for-like data. Google have made a fortune with text-matching advertisements for products and blogs by realising this, and Facebook’s own advertising is best at micro-targeting advertisements based on profile information about location, interests and etc that users have entered. So the loop of successful online products and services is finding a problem that some people need to have solved, solving it for them, and then telling them about that. 

Games (and movies, books, music, etc) don't solve problems. They may, in retrospect, help solve a problem that the player never knew they had, but players don't start looking for entertainment in a problem-solving mindset. An interest in some subjects may help you target advertisements with partial success in targeting, but for most games that really doesn’t translate into anything meaningfully useful. Like-for-like data usually translates into boredom. To be entertained, a customer needs to be surprised.

Entertainment must take us somewhere new or we quickly lose interest. This means a compelling and different story is essential, a lot of attention from media helps, and you need an advertisement or message that will punch through to people based on mass interest with enough of a budget to make it stick. It also means that an increasingly veteran audience will be less likely to pay attention. So Zynga will have to spend more to acquire them, which leads to a cycle of further saturation, lack of interest and lowering ad engagement. This is why Zynga’s recent investment haul of $180m makes sense. They need the money to market like they’ve never marketed before. 

That strategy will eventually fail, however, because the games being advertised are still shallow. Zynga may well get into more interesting marketing messaging (such as big competitions) but veterans eventually realise that these are just more tricks, and so will only interact as far as is necessary to obtain their bribe or just ignore them completely if they feel the bribes are not worth it. 

The only real way out of this saturation problem lies in smaller-scale, sexier and more remarkable development,  slower building of true fan-bases and building self-marketing software. This is very different from the trend-hopping, advertising-and-trick-sharing poor-software-quality strategy which viral developers have thus far employed.

Such an approach does not require deep marketing pockets and has been proven to work in the online game space before (EVE Online, Runescape, Puzzle Pirates, for example) but it’s a completely different way of approaching the whole market based on doing something different rather than copying everyone else. One for which I think none of the currently big developers really have the patience. 

Equivalent Quality: The Problem with Formulaic Thinking

All the excitement over Facebook and Zynga reminds me of the Atari era:

When Atari invented Pong they realised that they were onto a good thing. They quickly established a platform that players really liked and proceeded to open the doors to other developers. Initially there were some good and interesting games but, as Atari's platform matured, poor quality software became a problem. 

Novices who were impressed that they could play Pong in their living room morphed into veterans who got tired of Pong clones. The nascent game industry, high on the quick fortunes that it was making, didn't handle this at all well and continued to simply churn out product to cash in on trends. Deals were struck, games became an easy secondary revenue stream for the movie industry, and the publishers and the platform owners focused on extracting value rather than creating it. 

An over-saturation of bad product led to a crash because the companies involved in the boom were convinced that games were a digital form of fast food. During certain periods in the history of gaming platforms, some companies have stumbled onto a way to churn content toward undemanding players and reap rewards for a period of time. 

They get into the mindset that content is a “special sauce” and distribution is what really matters. And so the whole point of the games industry becomes the inventing and copying of digital Big Macs, Chickens Royale and a Fillets o Fish and shoving it down as many throats as possible before the other guy does. Competitors prefer to let each other do the heavy lifting of inventing, jump on the most promising trends with their own versions, and then really rely on distribution and marketing to win through. 

A fast food formula consists of several elements:

In viral games these correspond to:

When you have this in place you simply crank the handle. There's just one small problem: games are not fast food. 

Consumers never get bored of hamburgers because they have a biological urge to eat and – provided they don't overdo it – they will always enjoy a familiar taste of a burger. Consumers may find that they prefer a burger from one chain over another and switch brands, but they often stick to the one they know best, so the reach of the distribution chain is what really matters. 

Players do, however, get bored of Pong. Even if they utterly love the game for a time, 99% of players eventually get bored of Pong. They get bored of clones of Pong too. They don't just switch from one Pong clone another. Instead they start to look for more from Pong-type games, driving their expectations, and - when Pong has nothing else to give, they stop playing the Pong genre. 

Player boredom is not easily solved. In the short term it can be achieved by instituting a few tweaks in a successful game or genre, running competitions to attract re-engagement, or other similar behaviour. However veterans eventually get wise to that kind of behaviour. Continuous development and expansion tend to work better, which are key advantages of online games over retail games, but they too have limits if the core of the game is essentially thin. 

Players also do not have to play games. That's what Atari discovered too late: Players simply stopped buying into the platform at all. Theygot bored of being served the same few games in different packaging and moved on. As with Atari, the risk to Zynga is the fallacy of thinking that equivalent quality is just fine because it assumes that players are always looking to play. They're not. 

Novice players are currently consuming all sorts of games on Facebook because they are new. The software quality of viral games is generally low but players don't seem to care for the moment. So in response, all of the major developers are simply serving the same five or six games in different packaging and they are all scrambling like crazy playing the distribution game, much like fast food chains do with their beef, chicken and fish sandwiches. 

When developers assume that the task at hand is to simply get their version of a game in front of a player’s eyeballs before the competition and that quality doesn’t really matter, they are flirting with disaster. Viral game developers really do seem to believe that the only way for the adoption and profit graph to go is up because players are always looking to play.

Economists call that kind of thinking "irrational exuberance".

There is also no guarantee that a player will simply change and play a different game. A lot of the Zynga strategy seems to revolve around pushing players from one game into another and betting on this being an eternally repeatable process. By convincing themselves that the game business is all just a matter of winning at distribution, developers forget to create real value. They assume that equivalent value among their peers is enough because players are unable to go elsewhere.

They are, of course, utterly wrong. There really is nothing stopping players from simply going elsewhere to be entertained because the Facebook audience is not a captive audience. 

Bucking Trends: When Someone Makes a Great Game that is Hard to Clone

The other risk is the developer who creates a game that can’t be easily copied. Part of why the fast food model is so compelling is the assumption that the equivalent quality will remain more or less as-is and that any new games that come along can be fairly easily copied. In the Atari era this was true because the platform had hard physical limits on what it could achieve, both internally in the silicon and externally in the one-button joystick. The Facebook platform is far less constrained. 

Role-playing games are among the easiest of all games to copy. The role-playing game mechanic is incredibly simple at heart: A player is simply set a series of goals and a number of activities to reach that goal, and the rest of it is just time.

World of Warcraft players call this “grinding”. In games like World of Warcraft and EVE Online, grinding has a purpose: The character that you build or the spaceship that you’ve bought is intended to be used to overcome a tough challenge like a quest or a space battle. On the other hand, in FarmVille grinding is essentially all there is. You work your farm day and night assembling money to buy new stuff to put in your farm, which enables you to grind in different ways to earn more money to buy more stuff... and so on. 

The big difference between the Warcraft style of grinding and the FarmVille style of grinding is the need for game balance. Game balance in World of Warcraft is a horribly complex feat of engineering, testing and design which, even after seven years, Blizzard still obsessively correct and re-correct with every major release. On the other hand, FarmVille has no game balance at all. All that needs to be changed is a few quantities or timer lengths that players particularly complain about. It’s entirely up to the player how long they want to spend harvesting virtual wheat to buy a tractor after that. 

Grind-only role-playing games also consist of essentially one game mechanic: Acquire more stuff to unlock more stuff which, when acquired, will unlock even more stuff. Everything else is just customisation. Customisation is a bit important as it allows players to express their creative side, such as building their own unique farm, but customisation tends to become less interesting as a motivation once the novelty wears off. Anecdotally, many players in Restaurant City initially spend time laying out their dream restaurant until they realise that it is inefficient. So, eventually, all players end up creating sushi bars because they are the most efficient layout.

Games which require no real game balance combined with single-mechanic game structures are incredibly easy to study and copy. It took Zynga less than three months to ramp up Farm Ville after Farm Town appeared, less time to create Fish Ville after CrowdStar’s Happy Aquarium showed its appeal, and Mafia Wars likewise cannot have taken that long to ramp up. These games have since had further development in public but it’s the time it takes Zynga to go from a standing start to a first, reasonably playable, release that allows them to turn on the marketing machine and jump on forthcoming trends.

Even the one apparent exception to this observation, Poker, actually conforms to this rule. The only actual difference between Poker games on Facebook and other role-playing games is that they incorporate a slightly more complex resolution mechanic (the Poker playing). The goal is still much the same, i.e. acquiring levels and virtual stuff, and the game mechanic used to perform resolution is just Poker. Poker is completely generic and easy to clone. 

So the big problem that Zynga has on the development front is a developer who comes along with a game that they can’t immediately decode and replicate. For example:

With each of the four examples above, the key difference is time to market. FarmVille got to market in only 12-16 weeks, but a first-release copy of a great strategy game could easily take 28-36 weeks – probably more – to get anywhere near good. That's too late to jump on a trend, as Playfish's late entries in poker and farming show. 

The viral developers thus have a very big problem: They have little internal culture that rewards taking the time to do good game development, and so any new developers that come onto the scene that do will start to draw in veteran players in a way that the viral companies cannot easily match. Viral developers, being ultimately trend-hoppers, are no good as trend-setters. 

For example:

Short Term / Long Term

Fundamentally the problem that Zynga has is that it has spent all of its time acquiring players and no time turning them into real fans. They’re only thinking short term, and have much the same attitude toward players as amusement arcades once did – extract value quickly and move on. As a result they are failing to do the most important thing that any internet-based company must do: 

Build a following. 

Entertainment is like dating. You should always strive to be sexy.  Sexiness is all about creativity, credibility, charisma and character. Sexy people are at the forefront. Sexy people have a sense of mission and identity that they have made their own. Sexy people are personal, interested and responsive. Sexy people flirt with you. They don't rip you off. They don't make you feel used. And in return, sexy people develop followings. Smart sexy people make you feel loved. Really smart sexy people continually reinvent themselves and still bring their original followers along for the ride.  

Viral game companies are not sexy. Their brands and core values are generic brands that nobody wants to date and values that are entirely based on commercialism of a short term opportunity. The only people who find viral game companies sexy are the investment community and startup news sites like Mashable because of the earnings and valuation speculation. 

To be sexy in games means going to the edges of mainstream experience and finding something that brings interested people along for the ride. Games Workshop has been at the edge of the board and war gaming industry for a long time by cleaving to a dark style and a compelling game world, and their audience is loyal enough to spend hundreds of pounds per year on lead figurines. Blizzard has been at the edge of videogames by laying down a stamp of "done when it's done" messaging to development and forming a distinctive character to every game they do. They've been so successful that World of Warcraft by itself is often held responsible for sucking all the money out of PC retail games. 

Zynga has no sex appeal. They make generic games with generic names (Restaurant/ Cafe/ Bistro/ Hospital/ Farm/ Monster Town/Ville /City /Village /Country / Story/ Wars) and un-ambitious vision. Their games are neither inherently memorable nor compelling. For most players, they’re just something to pass the time. 

Their challenge, therefore, is to take what they have learned so far and invest in the future. If Facebook are making deep changes to their platform and handing the opt-out power back to players, then Zynga cannot survive by just looking for another way to trick and spam players back into FarmVille. They need to take their $180m investment and use it to build a real following.

They could build a portal, independent from Facebook, or even a competing social network to become their own platform. Alternatively they could invest in larger, better games or more complex and complete virtual worlds. They could broaden out to smaller, more indie and creatively-oriented developers and become an aggregator or partner network. Such a strategy would result in better and more diverse content.

Realistically though, I think they plan to do none of those things. I think what they will actually do is spend the $180m on trying to replicate their previous viral success through increased advertising spend. Maybe Zynga can figure out a way to leverage Facebook’s own ads to target to players as reminders. That would be much more in-character for a company as competitive and in-the-now as Zynga has proven itself to be.

Zynga’s coffers are deep, as are Playdom and Playfish’s, but at the heart of their model are some deep weaknesses that are going to let a lot of the air out of their Fast Food business models. The audience expectations are going to shift, the key factors enabling the business model likewise, and while it’s been a great short term success this year, viral gaming doesn’t seem to have any more easy wins left. 

Now comes the hard part. Diversification, experimentation and deep design breeding interesting ideas do not grow on trees and companies need to commit to them to see them through. Right now that’s not the Zynga way.  

Twelve months from now it will be the companies that have managed to diversify, build strong followings and create real value that will be the new darlings of the scene. Those that do not adapt will still be there but their story will be one of difficulty. As social games come to the end of their beginning, Zynga is increasingly look like an Atari-era publisher leading the charge but unlikely to capitalise in the longer term because they’re too busy thinking they’re in the burger business. 

(you can follow me @tiedtiger on Twitter)