Tag Archives: product strategy

8 Lessons Learned from Zynga about Virality

Posted on by
11

Editor’s Note: I’m almost embarrassed to say I actually had to play Frontierville and Farmville to write this post.  These games send notifications constantly and I wasn’t entirely comfortable with all my friends knowing how my pigs were doing. Therefore, I had to go undercover and create a new FB account under my Indian alter ego: Amit Kumar :)

Zynga has experienced explosive growth since it’s inception in 2007. They’re absolutely crushing it with:

How did Zynga become such a profit generating, user acquiring, viral monster? Here’s a few strategies that worked for Zynga which you can apply to your business:

1. Build sharing as an integral part of the product

At every point Zynga prompts users to sign up for updates and send notifications (many times crossing the line) through: fan pages/likes, invites, shares, bookmarks, stream publishing, in game messaging, and email notifications. In fact they’re so darn good at notifying ”Facebook had to change their notification policies

Before you even get started playing you’re prompted to:

- Become a Fan/Like (so they can push updates through your news stream)

- Invite your friends (pro tip: make it look like it’s required)

- Share this page on your news stream

- They again remind you right at the start of the game:

Sharing is not an afterthought limited to an invite section, it’s an integral part of the experience.  In fact the more you share and interact with others, the more you are rewarded.

2. Create ways to elevate a users status/social capital

Zynga doesn’t want you to just invite your friends to play, they want to enable you to do favors for your friends.  Mark Pincus’  (Zynga’s CEO) describes his goal’s in Wired:

One way to think about social capital is we’re all in a certain bucket with each other, and the lowest bucket is maybe you’ll accept a friend request, and the highest bucket is you’ll come over and help me move, or pick me up at the airport,” Pincus told the Wired Business Conference Tuesday. “The question is, is there something we can do to help you move buckets?

While it’s cool (sort of) that I level’d up in Frontierville, I’m not gonna tell my friends about it, but… if I can share with them some of the points I’ve accumulated then that’s a lot more compelling.  It no longer feels like spam, but instead that I’m helping them:

Applying this to a web business don’t just give a discount to the user who invites 5 of her friends to the service, let her pass on the discount to her friends (that’s something you really wanna share).

3. Assume you are gonna get it wrong at first

Don’t assume that your product is “the one.”  Zynga does a great job of experimenting and making decisions based on data.  I love this video of Pincus speaking at Stanford:

Towards the end he is asked if he could go back and share a lesson with himself when he was CEO of Tribes what would it be.

If I could do it all over again, I would have made Tribe a platform to test many ideas of social networking. We tried just one. Oh my god what the hell was I thinking? Just one? At our company we have several hundred tests going on every day and in every game. I would’ve done is made Tribe a platform to test every configuration

This is echo’d by others at Zynga:

Analysis, analysis, analysis. It’s been like that. [In traditional development] It’s just like, “Oh, it’s going to be a great experience,” and this and that. We’ll spend two years down a ship cycle, and, “Oops, I was wrong!”

So now it’s like, “If we do this, I think we can measure that, and here’s how we’re going to measure and tweak it later down the road…” We are an incredibly analytical organization, so we track just about everything. It’s the secret sauce behind all that stuff. There’s a lot of mathematics that go into it

That’s the magic behind what we do. Certain things we do will work, and others won’t. You try new ones, and A, B, C, D, E, F, G testing constant. (Source)

Several hundred A, B, C, D, E, F, G tests going on daily… :-o sweet bejeezus.

4. “Ghetto” test

You don’t actually have to build it to find out if it works.

In the last 5 minutes of the video above Pincus is asked what’s the best way to do market research. His answer – “Ghetto Test”. If someone wants to build, let’s say, a hospital simulator he creates an FB ad that says, “Ever wanted to run your own hospital?” which leads to a survey (or if it’s really ghetto a 404 page).

All Zynga has to do is track CTR and compare it to previous historical rates to get a pretty good idea of demand. I’ve heard a lot of people test demand for a product idea, and A/B test marketing copy using Adwords (you don’t care about the goal, just the intent).

The guys at Aardvark (another product I love) call this Wizard of Oz testing.

5. Kill products that aren’t performing

Focusing on products that don’t work funnels attention and energy away from ones that do.  Even if you’ve made substantial investments know when it’s time to let go.

We’ve actually made investments into some innovative games that were incredibly hardcore. If you look at Guild of Heroes, for example, we did roll that out. It was a version of Diablo built in Flash, and it wasn’t successful, and we didn’t support it any longer. (Source)

6. Create unexpected moments of delight

Surprise your users through game mechanics, humorous copy, badges, and easter eggs.  If you make them smile, they’ll tell they’re much more likely to come back and tell their friends.

One of the really fun and successful features we added is what we call the ‘Lonely Cow’ feature,” said Skaggs. “You can help find it a home, then somebody claims it. You’ll get a brown cow instead of the white cow you had before. Then you milk the brown cow and you get chocolate milk! That’s a ‘moment of delight,’ totally unexpected but cool (Source).

7. Leverage your size to cross promote like crazy

The best customer for one of your new products is an existing customer.  If you liked Farmville you’ll looooooooove Frontierville.

Only by leveraging their existing user base were they able to get 100,000 users on Frontierville’s first day.

8. Maximize Trends

Pincus says the web is about repeatable formulas and once you find something that works, it doesn’t break for a long, long time.  Think of LOLcats, rickrolling and Google.  Nothing that Zynga does is new, but they’ve executed on it tremendously well. They’ve found a formula that works and are bangin’ out hits at an astouning rate.

Certainly Zynga has it’s flaws: their games are boring, repetitive, compulsive and shallow and they’ve made shady deals to generate revenue.  But it’s undeniable that they’ve experienced enormous success due to the viral nature of their products.  Anything else I’ve missed in terms of reasons for Zyngas growth?  Lemme know in the comments.

2010 Mobile Web Research: India and the World

Posted on by
4

We’re looking to develop mobile versions of the product suite I’m working on.  I know nothing about the mobile web and had some really basic questions.  Below is some research I did in Q&A format:

Q. How big is the mobile web?
A. According to Quantcast Mobile Web Trends Report:

  • In North America about 1.3% of all pageviews come from mobile browsers.  This is expected to grow a whole percentage point by the end of 2010.
  • Globally, Quantcast predicts growth from .95% to 1.8% in 2010.

Answers to the next set of questions from Admob Feb 2010 Mobile Report (please note this data is from AdMob’s network only and may not be indicative of the entire mobile web)

Q. Do smartphones generate more traffic then feature phones globally?
A. Mobile Traffic Share: 48% smartphones, 35% feature phones, 18% other (basically iPod Touch)

Q. What smartphone OS is most popular globally?
A. 50% of all Smartphone traffic comes from iPhones (wow!), 24% Android, 18% Symbian, 4% RIM (surprisingly low)

Q. What featurephone is most popular globally?
A. 32% Samsung, 24% Nokia, 12% SonyEriccson, 10% Motorola

Q. What countries have the highest percentage of mobile web requests?
A. US (50%), India (5.9%), UK (4.2%), Indonesia (3.7%), Canada (2.9%)

Answers to the next set of questions all from AdMob SE Asia Report (please note this data is from AdMob’s network only and may not be indicative of the entire mobile web)

Q. In India are smartphones or feature phones more popular for accessing the mobile web?
A. Smartphones – 66.3%, featurephones – 33.4%

Q. What are the most popular Smartphones in India?

Q. What are the most popular featurephones in India?

Additional resources which I found useful:

Steal this Product Idea #2

Posted on by
5

So it’s 2010 and according to the 80′s movies I was obsessed with as a kid we should have hoverboards, flying cars, sexy robots, and violent but thrilling reality tv gameshows.

I can forgive scientists for failing to deliver on the important stuff, but at the very least, they should’ve come up with a really easy way for me to stay up to date on the music I love.

Here’s my problem: I love music, but downloading it is a pain in the ass.  Many times:

  • It’s not the right version (terrible sounding live album or kung-fu panda in croatian)
  • The quality sucks (camcorder rip)
  • It’s not tagged properly
  • It’s not what you are looking for (self-promoting rappers, porn or just plain weird)
  • It’s super slow (maybe that’s just cause I’m in India)
  • It’s “illegal”

So in order to satisfy my cravings for new music without the hassle, I buy albums on iTunes. The issue is most of the time I have no idea what to buy.  The only way for me to stay up to date is to manually browsing the iTunes store, explicitly ask my friends for recommendations, listen to internet radio and write down the tracks I like or browsing hype machine, pandora and other music sites and just sampling music.  That’s is a lot of work and I’m really lazy (which is why I don’t buy a lot of music).

I don’t want to go out and look for music, movies, games, books and apps I want stuff I like to find me.

Here’s what I’m thinking: create a service that lets you “follow” your favorite digital content: music, movies, games, apps, and books and receive notifications any time new related content is released:

  • Phase 1 – Music over Twitter:
    • Put in your favorite artist, band or genre (similar to iLike or Pandora)
    • Decide how and how often you’d like to receive notifications (as soon as it happens, daily, weekly) and how (tweets, @mentions, direct message)
    • Link to a summary page which shows an activity feed (new tracks, alubms, remixes, videos, etc…) for the music you have decided you like
    • Link to iTunes for affiliate sales
  • Phase 2 – Other content: Movies (Actor, Director, Genre), Games (Game, Genre, Studio), Books (Author, Publisher, Genre) and Apps
  • Phase 3 – Recommendations: Tie up with services like Netflix and Pandora to start making recommendations on content you may like
  • Phase 4 – Other Notifications: Email, SMS, Facebook, etc…
  • Phase 5 – Incentivize users to repost content, by sharing revenue

Great Scott! Music Hack Day is coming to SF in a few weeks.  Someone please build this, I’d use it.

Steal this Product Idea: Sign.al

Posted on by
16

We are overloaded with information.  During the course of my day I email, tweet, comment, post, chat, message, buzz, check in, call, sms, mms, bbm and sometimes (if I’m really lucky) actually talk to people.  We are moving towards an ever increasing flood of content (much of it automated) and it’s only gettin’ worse.  One day soon everything will tweet.

It’s not the the sheer quantity of information that’s the problem (faster flow of information will only help people achieve more), it’s how we send and receive it:

  • Sending: How do I send out information so that it reaches it’s intended audience only.  In the upcoming era of persistent, public online identity, how do I can still share my green-beer, toga party pictures with my friends and make sure potential clients don’t see it?  Additionally, I want to publicize my boring social media posts without spamming my friends who I know really, really don’t care.
  • Receiving: With all this content around how do I make sure that that important stuff gets to me FAST, while the stuff that matters stays buried (until I get really bored or have lots of time to look through it).

The idea that’s been bouncin’ around my head tries to address the second point… enter: sign.al.

I have a dream… that one day my phone will ring when my buddy is callin’ me up to go grab a beer, while calls from vodafone bill collectors stay silent.  That my blackberry will only flash in meetings only when really really important stuff happens (like the Mets scoring a winning run).  That one day, we’ll be able to ignore the tens, hundereds or thousands of messages that don’t matter, and focus our attentions on the ones that do. Here’s how sign.al would work

  • You give it all your account information (gmail, facebook, twitter, etc…)
  • It starts off like any aggregator (Seesmic for instance), showing you a timeline of emails, facebook messages, tweets, yadda yadda:

  • Aright, now’s where it starts getting cool… after a little while, it moves away from a timeline view, to a priority view.  It starts guessing knowing what you are most likely gonna want to read and respond to and starts moving those to the top.  This shift is already happening (FB news feed vs time line, Mozilla Raindrop, Xobni for Bberry)
  • Sign.al can know what’s important by:
    • Frequency – How regularly you read and respond to individuals
    • Speed – Of your read/response
    • Popularity – The number of comments, retweets, likes, and mentions
    • Proximity – Number of shared connections
    • Medium – @mentions more weightage the email cc’s?
    • Geolocation – Are messages from Mumbai and NYC more important to me?
    • Time of day – Are certain types of messages more important at a certain time?
    • Content – Am I more likely to be interested in content about the Jets regardless of where, when, how it’s getting to me? (yes)
    • Recency – Moving away from this but still a factor
  • Phase 2 – Phone app
    • Incorporating voice and SMS into the mix of content to prioritize
    • Different types of notifications instead of timeline: ring for an important sms, vibrate for a somewhat important @mention, silent for a newsletter.
  • Phase 3 – Setting status (implicitly?): In a meeting, driving, out to lunch, available – Based on this reduce/amplify notification methods.  If I’m sitting at the airport bored and normally my phone vibrates for an important tweet, now ring.
  • Phase 4 – Anticipate – Based on where I am, the type of message, who I’m connecting with, the medium, the format etc… start anticipating how I may respond to them (in a totally not creepy big brotherish way)

Ideas are a dime a dozen, execution’s what matters.  Take this idea, build it, I’ll use it and be happy (just be forewarned you need to legal operations in Albania to actually register the sign.al domain name :p ).

quick update: my buddy Aditya actually wants to make this happen… check out his blog for more info.

The Problem With Facebook

Posted on by
5

I came across an article today in the NYTimes describing how people are starting to leave Facebook.  While the author’s claims are anecdotal (certainly the numbers indicate impressive growth), I found myself silently nodding in agreement at some of her points.  The author outlines the feelings of being a stalker, concern of ownership of data, the “corporatization” of the site, the concern over FB’s tracking on 3rd party sites and the uselessness of it all, as reasons people have recently left the site.  While I’m not ready to jump ship yet, I have noticed my FB use has drastically decreased.

In the past few months I’ve found that my newsfeed has less relevant information, mostly from people I don’t care about.  I’m big on Twitter, the updates I get are personal, social and professional.  I’m able to learn new stuff about the web world, keep up with my favorite artists and (to a lesser degree) connect with friends.  The updates in my Facebook feed are largely spammy quiz results from girls I went to middle school with (Sorry Laura, I don’t care what Twilight character you most resemble).   Maybe I should only accept real friend requests, maybe not.

Now it seems like FB is trying to compete with Twitter head on, by making status updates public and adding real-time search.  Personally I think this is a terrible idea.  I use Facebook and Twitter for different reasons.  Facebook can differentiate itself by being the place I go to stay up with my “real” friends, plan events and share photos.  Create more useful communication applications (like video and group chat), give me better ways to share my photos and videos (like embedding slideshows) and cut down spam by 90%.  There are huge opportunities for Facebook in location based services as well.

I understand that Facebook needs to make money, and sees real-time as a huge opportunity, but they should focus on Facebook connect (already 10k+ sites using it).  Want profit?  Spread your tendrils across the web, become the universal ID then the defacto payment system for the web.  Create an app store (I’d pay 99 cents for Mafia Wars) and allow microtransactions (25 cents for some more poker chips).

Don’t be Twitter, that’s not why we came to you in the first place…