Yes, gaming companies are extremely data-driven and they measure everything they can. I'm in charge of dashboards that have over 300 KPI's (key performance indicators) on them. What do we measure? A whole lot of shit. ARPU, ARPPU, HC (hard currency) spend/gained, SC (soft currency) gained/spend, Retention, DAU, Spend DAU, Gained DAU, Downloads, Activations, plus funnels for every page, every user action you can think of.
Do I think measuring all this is creepy? No. Why? Because when I come in the morning, I look at my stats. I'll be able to tell within an hour if the build the engineers shipped the night before had a bug that affected 30% of our users in a certain location. I'll then report this to the Engineering team, they can create a hotfix, and users can be happy again. Engineers are happy knowing that we have actual data to validate and confirm their gameplay suggestions (made during scrum/sprint planning sessions). I also report monetization metrics to the VP's. They want to know if the boatloads of money they are spending on user acquisition are actually paying off. And the PM's want to know if users are playing the game, or how far they get into the game.
Yes, we maximize for users who spend over $10k. These users even have names. They're called "whales" in industry parlance. Why wouldn't you want more of them in your game? This is a business afterall. I feel just as bad as the next person wondering whether some of these "whales" are addicts. I don't want to be creating products for addicts. But what if these users are just rich people who have tons of money to spend off the cuff? There's no way we can know because we never get that data. All I see is what's reported in ItunesConnect.
Do we ever record any personally identifiable information? No. Do we ever sell or distribute this data to outside parties? No. Can any of this data ever be used maliciously? No.
Personally, I like the way game companies work (hence why I work in the industry). It teaches you to measure everything. You make decisions quickly. And no one can argue with the facts. You'll know real fast whether the product is shitty and you can do an about-face, gather the troops, and push out something new.
I thought that term sounded familiar.
http://www.worldfreepoker.com/poker-glossary.html#W
"Whale: A poor player with a lot of money to lose."
I don't want to be creating products for addicts. But what if these users are just rich people who have tons of money to spend off the cuff?
Sure, just like at casinos.
You're right, in a way; there are people with more money than brains who are happy to blow a good chunk of it gambling and are really no worse off for it because they're rich. But, realistically, you and I both know that's never the whole story.
You and I also know that this isn't a problem inherent to just casual mobile games. It's a chronic problem to the gaming industry as a whole, and its not going to stop people from making entertainment for the millions of users who aren't addicts.
>I don't want to be creating products for addicts. But what if these users are just rich people who have tons of money to spend off the cuff?
Ahhh, ignorance is plausible deniability bliss.
I once wanted to work at a Lithium Ion battery start-up because I wanted to help create an energy source free of geo-political and climate turmoil. Then I discovered those sources were some of the most toxic to the environment both in the harvesting and the disposing of the material.
I then went on to do graduate research in explosives detection. We created machines that would help detect bombs in Iraq and landmines left in a peaceful villages. But then TSA used the machines to look at people's private parts and there was a mass privacy scare.
My point is: you can't go life worrying who might use the things you create for harmful purposes. You may be ignorant, but the alternative will drive you insane.
Did you ever stop to consider perhaps money isn't the worst fallout? The families of addicts suffer as well–and I suspect, when we look back years from now, we'll see a pattern of spouses and children ignored & damaged because companies like Zynga analyze/optimize everything in order to maximize engagement. A single person can't compete with that.
You're not just engineering people to part with their money–you're using your skills to get them to part with their time.
Sorry, but I consider what you do (& Zynga does) unethical.
I have friends who play Starcraft II a couple hours a day. I've known people who will skip work and play 8 hours a day. I've seen a 60 minute story on someone who was addicted to World of Warcraft. This is a chronic problem of the industry as a whole, but its not a reason to stop making games because the majority of users can have fun responsibly.
That effectively means the only way to "win" is by becoming a whale; casual play is basically not rewarded. That's the problem I have with most online games; the greatest benefits go to those spending unhealthy amounts of time and money on them. IMO a mature gaming industry will be one that is able to reward a similar playing commitment to a weekly trip to the cinema; 2 hours play a week is still rewarding. That would make games attractive to a much larger audience IMO and be sustainable; you're not pushing players to burnout.
Just out of interest; do you track KPIs around hours of play per day? How many hours is a typical whale spending?
The answer may be no for now (that you know of). You can't guarantee that it will stay that way forever. Given Pincus' general shadiness to make a buck, I can totally see a slow quarter or two driving a decision to "update" the privacy policy to sell data to 3rd parties.
Similarly, it's not a big programmatic change to start tracking PII as well as cross-referencing to existing information. So while you may not have any ill intentions, I doubt you can speak for all of Zynga's actions over time.
I would emphasize the "over time" component since it tends not to be taken into account by most people which is a big reason why we are being bombarded by SOPA-like legislation. (Didn't win this one? No problem, we've got ACTA, etc. And should ACTA fail, I'm sure there's something else over the horizon)
So even though I don't believe in it (I deactivated FB for this reason), I'm not sure how much financial benefit you can gain from <UserID> <StrawberryFarmID> TSV files which most of this data is anyways.
Why aren't the engineers doing this themselves? (Honest question, not snark) Analyzing performance/results data is (should be? :) )part of all engineering.
Which isn't to say I think terribly poorly of you, but it isn't entirely non-creepy.
Out of curiosity, do you offer users a feature to prevent themselves from spending money if they self-identify that they have a problem?
Oh and that is the raw cost of the models what it woudl be if you had to by that painted is anyones guess
If you're not maximizing that above all else, that is why you are hated.
Creepy stuff:
Spying on players. Getting intimate gaming data, their habits, their networks, and how to effectively monetize given X.
This never happened. Zynga doesn't do behavioral targeting at all.
Creepy/bad stuff in my view:
All decisions are short-term. Let us build X games in Y months.
Bad code quality. There is a team which optimizes for the bad code written by engineers at high speed.
Good stuff:
Speed - They are really aggressive. And they never think twice to throw away something if there is a better alternative.
Decisions - Right or wrong, decisions are made with least and things progress.
Meritocracy - I wouldn't call it pure "meritocracy", but when they know you are a smart guy who can ship stuff, they'll love you in $ and positions. I know several young people at VP/GM positions.
Data-driven - I see this as huge plus. Lot of companies are clueless because of not enough feedback-data. A/B testing, phased rollout, metrics are mandatory for any product/feature.
Zynga is 100% data driven. Zynga is not a tech-company, it is a product-manager-company. Product managers are more statisticians than PM. All decisions are from numbers.
This statement seems fundamentally at odds with the following. Could you explain further?
Data-driven - I see this as huge plus. Lot of companies are clueless because of not enough feedback-data. A/B testing, phased rollout, metrics are mandatory for any product/feature.
Zynga is 100% data driven. Zynga is not a tech-company, it is a product-manager-company. Product managers are more statisticians than PM. All decisions are from numbers.
Behavioural Targeting: User "A" loves lot of "Strawberry crops" let me show him more of that.
Brogrammer - Brogrammers are mostly Silicon Valley / Harvard type douchebags who got into programming. YCombinator drop outs. Years of experience in managed and web languages, but who have no idea how to setup a build system nor work in native code.
Maybe he's getting Brogrammers confused with web dev scenesters. Awesome Tumblrs, lo-fi keytar side band, scruffy hair, interesting glasses, polaroid cameras. I don't know if those guys have a name yet. Hipster is too easy...it's never really been "hip" to have a desk job.
If you can't handle a pointer, you have no business being near a compiler. If you don't know why you (usually) don't want to handle a pointer, you have no business calling yourself an engineer.
I can't tell if his answers are given out of anger against Zynga or if they are a reflection of reality.
"Google and Apple teams are total cunts. Fuck em, and their hipster bullshit."
Followed by:
"I like Google as a whole, but they hire too many wanna-be-devs. And their SC2 team can suck a big herpes-infested cock. FUCK THEM."
Hard to take this AMA seriously.
> Brogrammers are mostly Silicon Valley / Harvard type douchebags who got into programming. YCombinator drop outs. Years of experience in managed and web languages, but who have no idea how to setup a build system nor work in native code.
Eh, hes just talking like every other former marine: honesty riddled with swears.
I wasn't privy to all the details, but the impression seemed to me that it was a cost-cutting measure (no benefits for comparable pay), and a hiring-cap dodger, since at this company they didn't seem to count.
The increasing use of contractors-as-full-time was part of the reason I left. I personally knew two people who were getting strung along by HR with vague promises to turning full-time, and one who was unceremoniously let go when he was getting close to the legal limits. This seemed like a regular occurrence. I hated working next to people who had no benefits, no job security, and were doing precisely the same work as full-timers.
One particularly nasty one that I ran into would arbitrarily delay payment to employees for 90 or 120 days, unless the employees agreed to pay 10% for a weekly paycheck. The bad ones are even nastier to folks on temporary work visas.
If they are a good place, they will pay people whatever the fully-loaded cost of an employee is. Contractors aren't getting rich.
As for pay, we were hiring people right out of college at $65 an hour plus overtime, which works out to over $130,000. That's more than I made as an employee.
He mentioned he quit in August (I wonder how many engineers quit in August? Single digits?) and uses curious language quirks like "[they gave] nay fucks". As a Brit he sounds potentially British to me.
I know people in Zynga's management - they'll relish the opportunity to fuck this guy for contravening some part of his NDA. I'm sure Dani is pissed.
If you are going to 'whistleblow', people, be careful.
I've often wondered, how do you balance the two? When you have that much data, wouldn't it be insane to go by what any single person thinks is "good design"?
>Brogrammers
http://www.quora.com/Brogramming/How-does-a-programmer-becom...
