Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Zynga Analytics at its Peak (amplitude.com)
93 points by doppp on July 17, 2015 | hide | past | favorite | 49 comments


"Based on this insight, Zynga optimized for monetization by creating buildables that had a number of parts just a bit out of reach of a users’ friend network. That way, a player would get most of the parts from their friends, but would be unlikely to gather all of the parts from their network. Since they were so close to completion, they would just pay for the last one or two parts to complete the buildable."

If there is a Gamer's Hell; once a day George and Charles Parker, Gary Gygax, and Milton Bradley will come down from Gamer's Heaven and kick these guys in the balls.


I don't understand how that works, since each user has a different number of friends. Does the number of parts for each buildable vary from person to person? Or did they just make the number of parts be slightly higher than what the average user would have available?


If the typical person has 5 friends actively playing, and there are 5 different parts, the chances that a person can get one piece from each friend is 3%? There were also limits on the amount of times that players can help friends daily; it encourages a tit-for-tat dynamic of helping only those who help you. But without built-in tools to track this (mark fav friends, etc), there was always some tension about who "deserves" your help, and players would not spend all their help options right away so they could save some to give back to helpers later in the day, etc... All of these simple mechanics (or lack thereof) interact to create additional scarcity in a very natural, subtle and organic way. And all of it guided, adjusted and optimized via metrics and A/B tests.

(different games worked differently and evolved over time, the above is not meant to be a precise and exhaustive description)


I've never played a Zynga game so I'm just guessing. I think this is about

(probability of getting an item) = (probability of getting the item for free in a drop) * (1 + (probability of a user gifting an item to a friend) * (average number of friends)).

So for the late items, they pushed down (probability of getting the item for free in a drop), so you were unlikely to get it in a drop or as a gift from a friend.


It depends on the game, but for some of the games the "cost" is indeed maximized per player for optimal "in game" purchases.


Zynga is a painter that makes every brush stroke while looking at a spreadsheet instead of the canvas. They've now been outcompeted by painters who keep one eye on the canvas and one eye on the spreadsheet. That's progress I guess.


You mean like King? Nah, I'm not seeing a whole lot of progress there.

And I think there's some risk to trying to retrodict these kinds of events. It's just too easy to pick up on coincidences and assign too much importance to them. It becomes especially dangerous in light of this big data fad in which it's become common to believe that you really can infer causation from observed correlations, and then feel confident betting entire businesses on it.


I am a bit concerned with the direction of in-game purchases. Games are being designed to push users to spend money. Whats the difference between this and gambling in casinos? Are users aware that algorithms are being gamed to maximize profits? Is this ethical? I think we really need some oversight here. I think an entire generation is being exposed to gambling and exploitation -- and I am not sure if its healthy.


> Games are being designed to push users to spend money.

Every business is designed to push their clients to spend money. Some are better at it than others.

> Whats the difference between this and gambling in casinos?

There's no possibility of winning, so no way to delude yourself that the money you spend is an "investment", or to try to make back your losses. And we haven't (yet) seen anything like the scale of social problems caused by casino gambling.

> Are users aware that algorithms are being gamed to maximize profits?

Maybe. But honestly what's the difference between an algorithm and a human salesman? (e.g. I've heard it alleged that Dumas and Dickens deliberately stretched and optimized their serialized stories, to maximize their profits) Does it matter?


An algorithm can make the game harder after it confirms you are willing to buy items. It's getting to the point where games are short circuiting your ability to reason.


I never thought about it that way. Interesting insight. One aspect is that users don't receive a cash payout. They pay money in, but never receive money in return, like you would expect from true gambling. This falls apart when you allow trading within a game though, as you can exchange real money for a virtual trade. Then it does involve a payout.


Interesting point - I would originally agree with the other posters that said Gaming is different because there is no chance of making back your money, but you are correct, in a lot of games with markets and currencies, there is a way to make money from your time spent gaming.

I guess this would only be an issue if the purchase-element manifested directly in a way to make money? For example with Team Fortress 2; you can buy keys to unlock the item crates (which deliver a player a single new random item) - but you can't buy the item crates themselves - and you can't buy an item to make the crate's randomness any more favorable.


An unregulated casino would never give payouts either. (But they would generate as much false hope as they can afford.)

It's a promise of happiness in exchange for addictive behavior.


>An unregulated casino would never give payouts either.

An unregulated casino would not see many patrons then.


Why is that? You don't think an unregulated casino could stage wins to make it look like it was giving payouts? Or just make up stories?


Do you have an unlimited number of stooges? Such a facade can only be kept up for so long.


casinos didn't birth into existence fully regulated. their illegality and regulation came about because they were fleecing their many patrons.


And not all casinos never paid out.


This sounds like the same thing credit card companies who target young teenagers and college kids to get them hooked into impulse spending and rack up large sums of debt.


I think you have your causation all messed up. We're immersed in a culture of "keeping up with the Jones'" impulse spending. Credit cards are a just tool to fulfill that desire.

Really, what do credit card companies do besides set up a booth on campus, saying, "Want a card?", and kids line up around the block?


In my eyes the type of "games" Zynga create are no different from sites like madbid.com. Except the hook is needing to have a better looking/more successful farm than your friends, rather than an iPhone for £1.50


Uuuh... and what about CPGs? Chocolate bars are designed to push consumers to spend more money. Whats the difference between this and shooting cocaine in your veins?


>"I think we really need some oversight here."

Yeah, let's get the government involved, because they don't have better things to do than make sure people aren't spending too much on their fake farms.

>and I am not sure if its healthy.

I'll bet you have a few habits or activities that people aren't sure "are healthy".


Where did I say government? If I am not mistaken, the organizations that oversee the casinos are private sector. If they are needed here then I see no reason why it cant be setup the same.

As for your other comment... jeez. Why so passive-aggressive? You are certainly right -- we all have habits that someone would consider unhealthy. But is my taste for sweets as destructive as this? I dont know, seems like they are both serious. It might even be that the addictiveness and lack of end-game of these games make them more dangerous.

Anyways, my main concern is that these games are being played by kids. I think everyone agrees that kids are still developing. What is the impact of these kinds of games on them? Do you know for sure that its harmless?

Not meaning to insult you or anyone, but dismissing these questions without a good reason is just plain ignorance and probably irresponsible. I am not equipped to answer these questions. Which is why this post and my original one are filled with questions.


> If I am not mistaken, the organizations that oversee the casinos are private sector.

You are mistaken, at least in America. Except for those Native American run, casinos in America are regulated by some form of a gaming control board.

To my knowledge, this is true of pretty much anywhere that has gambling as an industry, though I wouldn't be shocked to find out that Monte Carlo or somewhere were wholly unregulated.

https://en.wikipedia.org/wiki/Gaming_control_board


>"Why so passive-aggressive?"

I consider "people are doing things I don't approve of. We should regulate that behavior" to be passive aggressive.

>"But is my taste for sweets as destructive as this?"

I'm pretty sure obesity kills more people than social games.

>"my main concern is that these games are being played by kids."

The vast majority of these games are played by adults.

>"Do you know for sure that its harmless?"

I don't care if it's "harmless". It isn't my duty, or yours, to ensure that everything everyone else does is "harmless".


Would it bother you if casino methods were being applied to kids games?


They are. See: Disney, et al.

Please stop invoking "won't someone please think of the children!!"


Well, as a parent I'd do the responsible thing and ensure that the games I purchase for my children to play (because, you know, they don't have iTunes or Facebook accounts, or credit cards) don't employ those methods.

This is really a non-issue.


To call Zynga a pioneer is a huge insult to the vast number of game makers they stole from.


To ignore the ways Zynga was a pioneer is a huge insult to the vast number of non-game makers they employed.


I don't really think anyone is being hugely insulted by people calling or not calling Zynga a pioneer.

This is some outrage culture shit right here.


now you are saying that zynga invented analytics.


No, but they successfully translated the addictive, mindless, money-flushing dopamine rush of gambling into the gaming world, all while neatly avoiding the US's "no paid online gambling" laws.


It's hard for me to critique or analyze Zynga from an unbiased perspective.

There's no question that what they were able to create this monstrous flywheel of engagement around a strong core concept but from my vantage point (read: I know they printed money with FarmVille and CityVille and have more or less floundered in the past year), they were only really good at the "let's take this already successful concept/hook and turn it into a cash cow" side of the equation.

Has Zynga had any successes in the past twelve months? Were they able to break into mobile at all beyond paying $180M for Draw Something?


I think Empires and Allies is making a lot of money currently. It is in the top grossing charts for quite a while. The game is quite good and targets a mid-core audience. https://zynga.com/games/empires-and-allies


They have a few of the top games in the social casino category on both iOS and Android. I work there and make some of the slots games. If you're a fan of gambling they're actually really good games even though you're not playing for real money.

https://zynga.com/games/zynga-poker https://zynga.com/games/free-slots https://zynga.com/games/wizard-oz-slots


You have to take FarmVille and CityVille's so called "success" with a grain of salt. Social games companies don't measure success by revenue, they measure it by daily active users (DAU). What you don't see is how much money they spend on advertising to attract users to the game. I know at one point, they were spending much more per user than the game was bringing in. You also don't see the churn rate, which is typically high in social games.

There's a real possibility Zynga was never as successful as it appeared to be, from a cash perspective.


>"There's a real possibility Zynga was never as successful as it appeared to be, from a cash perspective."

No need for speculation, they went public in 2011. You can read their financial statements.


This is where focusing too much on data and moving things up and to the right without any overall thought goes wrong. This is how you become evil and being successful and evil is extremely hard if you don't have a monopoly on infrastructure (eg. Comcast)


“an analytics company masquerading as a games company.”

I'm wondering if that was responsible for it's downfall? I get that gaming is somewhat of a hit-driven business but not nearly like, say, movies. Incumbents have humongous distribution advantages. I'm amazed that Zynga relinquished its commanding position so thoroughly and so rapidly.


I've always wondered if Zynga would have been more successful now if they had pivoted and turned into an analytics and monetization SaaS platform with consulting services.

They are clearly masters at the mechanics and psychology of this, and the technical implementation as well. When your mine runs dry, there's still often quite a bit of money to be made selling shovels to the other miners.


I'm a former Z employee -- one of my friends worked on proposals for a spinoff of the analytics platform, but he never got anywhere.

Obviously this is very hard to to do for a number of reasons. Mixpanel employees more salespeople than they do engineers or designers, and Zynga is still 2-3x their valuation just on the core games business.


That's really interesting--thanks for sharing those insights.

Zynga is in a hits-based industry, so having a reliable recurring revenue from a b2b SaaS product could enable them more flexibility with their approach on the gaming side from a pricing an user acquisition standpoint. But of course I'm speculating without any hard numbers. Sounds like people much more informed than I am have decided it is not a valid opportunity, and given how far it is from their core business, I can easily see why.

Thanks again.


All the ways gambling takes advantage of human nature, but without slight chance of a payout. Gross!


"During its peak, Zynga was a powerhouse at using data and analytics to optimize their games for virality and revenue."

Guess they didn't optimize well enough. Or optimized only for the very-short term.


The original owners pumped it up, IPOed at the peak, and walked away with their money. They're probably very happy with the outcome.


It's good info, but I can't help think about all the creatives that these suited finance guys ripped off. So they got rid of the need for creativity by stealing. Now, all they had to do was apply metrics to their widgets and watch the money roll in.


>but I can't help think about all the creatives that these suited finance guys ripped off.

Ah yes, good guys versus bad guys. Has it ever crossed your mind that they're one and the same? Must be nice to live in a world where "sides" can be chosen so naively.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: