The Tipping Point for User Generated Content and Community Websites

Project “Tipping Point” OPIM 469: Information Strategy and Economics Ursina Beerli, Steven Gao, Juhi Heda, Grant Wilson Published with permission on  www.chubbybrain.com The blog entry featuring this analysis can be found on the  ChubbyBlog @ http://www.chubbybrain.com/blog Published with permission on www.chubbybrain.com Agenda Introduction & Motivation “Tipping Points” Student Survey Analysis Interface Insights Case Study: Yelp! versus Citysearch Fun Stuff Appendix Published with permission on www.chubbybrain.com Questions • For a two‐sided network (and user generated)  website, what factors lead to a “tipping point” after  which growth becomes exponential? • What are the implications of the way college  students use these types of websites today? • For websites that share a similar purpose, what  determines the success of one over the other(s)? Published with permission on www.chubbybrain.com Motivation A lot of literature about two‐sided networks exists but  there is not much about internet based ones and user  generated content websites, especially in regards to  this “tipping point”. Published with permission on www.chubbybrain.com The economics behind two-sided networks Definition: markets where the value of an agent in one market is related to the  number of actions taken by agents in another market   Price Product 1 Price Product 2 Externality  (2 1) Externality  (1 2 ) Quantity Quantity Network Effects:  Markets in which the value for one agent is affected by the number  of actions of other agents Externalities:  action of one agent affects the value of another Source: Economics of Two-sided Market by Prof. Lorin Hitt Published with permission on www.chubbybrain.com Who’s in your network? Profile users Video viewers Commenters Readers Application developers  Advertisers  Video creators  Advertisers Reviewers Advertisers Wikipedia users Wikipedia contributors Readers Reviewers Advertisers Published with permission on www.chubbybrain.com Website profiles URL Founded Traffic Rank Type Facebook.com Feb. 2004 5 Social networking Limited YouTube.com Feb. 2004 3 Video sharing Wikipedia.org Jan. 2001 8 Online collaborative encyclopedia None Yelp.com Oct. 2004 613 Social networking, user review, and search Ad sponsors Citysearch.com Sept. 1995 845 Online city guide Ads Integrated Pay for performance Published with permission on www.chubbybrain.com Top sites linking in: having several high traffic websites linking in may drive own traffic •Youtube.com •Wikipedia.org •Google.com •Hi5.com •Fotolog.com •Hi5.com •Google.com •Fotolog.com •Google.fr •Google.com •Fotolog.com •Metroflog.com •Wretch.cc •Flickr.com •Google.es •Google.com •Google.de •Google.jp •Digg.com •Flickr.com •Livejournal.com •Google.com •Google.co •Match.com •Cnn.com •Ask.com Source: Alexa.com Published with permission on www.chubbybrain.com Agenda Introduction & Motivation “Tipping Points” Student Survey Analysis Interface Insights Case Study: Yelp! versus Citysearch Fun Stuff Appendix Published with permission on www.chubbybrain.com Active Users (millions) Question: How can we  assess the growth of  these websites? Number of Active Users over Time 60 50 40 30 20 10 0 8/1/2004 2/17/2005 9/5/2005 3/24/2006 10/10/2006 4/28/2007 11/14/2007 Time Active users data and  Google Insights index  data for Facebook yield  similar graphs over  time. This implies that  Google Insights data can  be used as a reasonable  proxy for active user  data that was more  difficult to find for other  websites. Google Insights Index over Time 100 90 80 70 60 50 40 30 20 10 0 1/4/2004 1/4/2005 1/4/2006 1/4/2007 1/4/2008 Index Value Time Published with permission on www.chubbybrain.com 100 90 80 70 Time Between Launch and Tipping  Point: 3.4 years: (06/07 – 02/04) Index 60 50 40 30 20 10 0 01/04/04 07/04/04 01/04/05 07/04/05 01/04/06 07/04/06 01/04/07 07/04/07 01/04/08 07/04/08 Chronological Timeline: September 2005:  Facebook expands to include high school  networks September 2006:  Facebook expands registration so that  anyone can join June 2007: Facebook unveils platform for developers  of social applications (launches with 65  developer partners and over 85  applications) October 2007:  Facebook launches platform for mobile October 2007:  Microsoft invests in Facebook 06/03/2007 Date 0.4 Change in Index Value/  Week 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 10/6/2003 10/5/2004 10/5/2005 10/5/2006 10/5/2007 10/4/2008 November 2007:  Facebook unveils Facebook ads April 2008:  Facebook launches Facebook chat Date Published with permission on www.chubbybrain.com 100 90 80 70 Time Between Launch and Tipping  Point: 4.1 years: (02/05 – 01/01) Chronological Timeline: February 2004: Yahoo! includes Wikipedia in Content  Acquisition program which leads to  more  indexing. Also, new servers  are installed  after a computer crash took wiki offline for  a week in December. 08/04/04 03/04/05 10/04/05 05/04/06 12/04/06 07/04/07 02/04/08 09/04/08 Index 60 50 40 30 20 10 0 01/04/04 May 2004: Wikipedia introduces categorization. September 2004: Millionth article posted on Wikipedia. January 2005: Wikipedia.org domain changed from a  redirect to English Wikipedia  to a portal  showing all Wikipedia languages that have  over 100 articles. Date 0.12 0.1 0.08 0.06 0.04 0.02 0 10/6/2003 10/5/2004 10/5/2005 10/5/2006 10/5/2007 10/4/2008 02/20/2005 Change in Index Value/  Week Jan 2007: Esperanza  (Spanish rival) is disbanded and  former users turn to Wikipedia. July 2008:  Google launches Wikipedia rival Knol. Date Published with permission on www.chubbybrain.com 100 90 80 70 60 Time Between Launch and Tipping  Point: 2.1 years: (11/06 – 10/04) Chronological Timeline: November 2005: Yelp receives $3M VC funding. July 2006: Yelp! wins CNET award as top student site. October 2006: Yelp! Gets $10M funding to help it expand  to other metropolitan cities outside of San  Francisco. Also enters the fashion scene  with launch of Lifestyle Apparel Line. February 2008: Yelp! Gets additional $15M in funding and  opens a New York Office to speed up the  adoption on the East Coast. July 2008: Yelp! Announces the launch of “Yelp! For  Business Owners”. July 2008:  Yelp! launches native application for  iPhone. Index 50 40 30 20 10 0 07/04/04 01/04/05 07/04/05 01/04/06 07/04/06 01/04/07 07/04/07 01/04/08 07/04/08 Date Change in Index Value/  Week 0.25 0.2 0.15 0.1 11/05/2006 0.05 0 7/8/2003 7/7/2004 7/7/2005 7/7/2006 7/7/2007 7/6/2008 7/6/2009 Date Published with permission on www.chubbybrain.com Time Between Launch and Tipping  Point: 2.1 years: (03/06 – 02/04) 100 90 80 70 60 Chronological Timeline: February 2006:  SNL releases wildly popular video clip “Lazy  Sunday” in Dec. 05 which leads to great  popularity until NBC Universal asks YouTube  to take it down. Generates a lot of publicity. March 2006: First ads appear on YouTube. September 2006: Warner Music signs deal allowing YouTube  to host WM videos for portion of ad income. October 2006: YouTube bought by Google for 1.65 billion in  stock.  January 2007: Founder announces that online video service  will pay active user‐contributors a portion of  website ad revenue. March 2008: YouTube announces “YouTube Everywhere”  which allows third party develops to access  YouTube’s videos. November 2008: YouTube partners with MGM, CBS, and Lions  Gate to post full length films and T.V shows  to compete with Hulu. Index 50 40 30 20 10 0 01/04/05 07/04/05 01/04/06 07/04/06 01/04/07 07/04/07 01/04/08 07/04/08 01/04/ Date 0.2 0.18 0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 10/6/2003 10/5/2004 Change in Index Value/  Week 03/26/2006 10/5/2005 10/5/2006 10/5/2007 10/4/2008 10/4/2009 Date Published with permission on www.chubbybrain.com 100 90 80 70 60 Chronological Timeline: Mid 2006: Yelp! begins expanding to metropolitan cities  outside of San Francisco and “takes off”. October 2006: Citysearch launches mobile services. December 2006: Citysearch content becomes available  through ask.com’s AskCity. March 2007:  Citysearch acquires InsiderPages . With over  600,000 customer reviews and over 2.5  million people, InsiderPages.com was  created to help people find the best local  businesses through recommendations from  their friends and neighbors. August 2007: Citysearch adds menus to restaurant  profiles. 10/5/2004 10/5/2005 10/5/2006 10/5/2007 10/4/2008 Index 50 40 30 20 10 0 07/04/04 01/04/05 07/04/05 01/04/06 07/04/06 01/04/07 07/04/07 01/04/08 07/04/08 Date 0.1 Change in Index Value/  Week 0.05 0 10/6/2003 ‐0.05 ‐0.1 ‐0.15 04/23/2006 January 2008: “Eat It” application allows Facebook users to  share reviews with friends. February 2008: AOL and Citysearch announce strategic  content and advertising partnership. Published with permission on www.chubbybrain.com Date Takeaways • Providing incentives or access to third party developers can increase network  effects and expand user base. (Facebook) • Increasing global appeal can increase user base. (Wikipedia) • Targeting users in popular metropolitan areas may be a strategic way to propel  traffic and website use. (Yelp!) • Sometimes macro events beyond your control can greatly affect traffic. It’s up to  you to take advantage of this by finding ways to appeal to a network that might  benefit from your traffic. (YouTube) • Unfortunately, due to no “fault” of your own, another website with a similar  mission may negatively affect your own user base. Be prepared, and find ways to  add value for consumers in a way that your competition can’t, either through  content or through other features such as ease of use, sophistication of interface,  etc. (Citysearch) Published with permission on www.chubbybrain.com Agenda Introduction & Motivation “Tipping Points” Student Survey Analysis Interface Insights Case Study: Yelp! versus Citysearch Fun Stuff Appendix Published with permission on www.chubbybrain.com Demographic information of survey respondents Survey Respondents Demographics 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% Freshman 0% n University Year Major Penn Arts &  Sciences 160 Junior Other Senior Engineering Nursing  Business Other Alumni Multiple Sophomore Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Quality of content is by far the most important characteristic Importance of Website Characteristics Importance  (1‐least important, 10‐most important) 5.2 5.1 5 4.9 4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 Source: Student Survey, n=260; Quality of content, n=209; Amount/quality of content by others, n=210; Searchability, n=212; Amount of content, n=212; Update speed, n=212; Level of privacy available, n=211; Sophistication of interface, n=211; Number of friends, n=212; Level of moderation, n=208; Network, n=210 Published with permission on www.chubbybrain.com Usage and contribution is high for Facebook, YouTube, and Wikipedia How often do you use each of the  following websites? 250 # of respondents # of respondents 200 150 100 50 0 Never 1‐5 times per month 6‐10 times per month 10+ times per month Daily or almost daily 250 200 150 100 50 0 Never 1‐5 times (per month) 6‐10 times (per month) 10+ times (per month) Daily or almost daily How often do you contribute  to each of them? Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Boredom and ease are the most popular motivations for contributing What is your motivation for contributing to these  websites? 180 160 140 # of respondents 120 100 80 60 40 20 0 Fun to do when you’re bored It’s easy to do Easy platform Can reach a wide Want to for expressing audience establish and/or creativity share expertise on a topic Other Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Non-contributors would contribute more if given monetary incentives Monetary Incentive 250 200 150 100 50 0 Respondents 250 200 150 100 50 0 Respondents Status Incentive Don’t know No Yes Don’t know No Yes Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Successful websites replace others •MySpace •Xanga •Hi5 •Orkut  •Friendster •LinkedIn •University  directory •AIM  •Wayn •Lokalisten •Google Video •AOL Video •Break.com •Yahoo Music Videos  •MTV Music Videos •iTunes  •Limewire •Revver •Google & other  search engines •Ask  •Books &  encyclopedias  •Britannica •Encarta •Dictionary.com •Ask Jeeves  •Citysearch •Yellow pages  •Guidebooks Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Takeaways • Quality of content is perceived as most important, despite the fact these websites  are user‐generated. • Most people don’t perceive the network to be important, even though network  effects greatly influence the success of these websites. • Some user‐generated websites rely on a small group of dedicated fans to  contribute, though several more people use them. • A monetary incentive may be one way to grow the user base until it reaches a  “tipping point”. Published with permission on www.chubbybrain.com Agenda Introduction & Motivation “Tipping Points” Student Survey Analysis Interface Insights Case Study: Yelp! versus Citysearch Fun Stuff Appendix Published with permission on www.chubbybrain.com Facebook over three years August 06, 2005 August 13, 2005 January 15, 2006 January 5, 2007 August 3, 2007 December 3, 2008 Published with permission on www.chubbybrain.com Wikipedia over eight years July 27, 2001 September 25, 2001 September 30, 2002 August 4, 2003 April 27, 2006 December 3, 2008 Published with permission on www.chubbybrain.com YouTube over eight years April 28, 2005 June 22, 2005 November 24, 2005 January 16, 2006 November 18, 2007 December 3, 2008 Published with permission on www.chubbybrain.com Yelp! over four years October 3, 2004 October 17, 2004 August 3, 2005 October 29, 2005 August 15, 2006 December 3, 2008 Published with permission on www.chubbybrain.com Citysearch over eight years May 11, 2000 October 19, 2000 October 11, 2001 March 21, 2003 September 15, 2007 December 3, 2008 Published with permission on www.chubbybrain.com Takeaways • Create an identity! Maintaining a color scheme, logo, or layout may help users  recognize and associate with a website. Without a cohesive image, consumers may  use your website without even thinking about the “brand”. (Citysearch) • Cut the clutter if possible. (Facebook, Wikipedia) • The use of a “teaser” home page pre‐launch may be a strategic way to create  hype. (Yelp!) • Constantly displaying fresh data on the homepage may persuade users to come  back often. (Yelp!, YouTube) Published with permission on www.chubbybrain.com Agenda Introduction & Motivation “Tipping Points” Student Survey Analysis Interface Insights Case Study: Yelp! versus Citysearch Fun Stuff Appendix Published with permission on www.chubbybrain.com Yelp! Vs. Citysearch Number of unique visitors to Citysearch has been declining while number of  unique visitors to Yelp! has been increasing. Published with permission on www.chubbybrain.com Yelp! Vs. Citysearch Restaurant listings on mobile Restaurant profiles on mobile Citysearch’s mobile application launched just a few months after Yelp!’s and looks  very similar. Competition is getting fierce and each company must find a unique  competitive advantage to secure and maintain users. Published with permission on www.chubbybrain.com Yelp! Vs. Citysearch Yelp! City Search • Grouped around large  metropolitan areas (+) • 4 years old • Centered around San Francisco  and slowly expanding to major  metro areas (+) • Consistent brand image (+) • Cultish following centered by its  focus in California • Over 75,000 cities of all sizes • 13 years old • Database a result of years of  acquisitions and widespread use • Rocky history centered around  numerous acquisitions • Recently revamped data center  reorganized about  neighborhoods Published with permission on www.chubbybrain.com Agenda Introduction & Motivation “Tipping Points” Student Survey Analysis Interface Insights Case Study: Yelp! versus Citysearch Fun Stuff Appendix Published with permission on www.chubbybrain.com It Pays to have Pals • PayPal is a start‐up forming machine – YouTube – LinkedIn – Yelp! • Innovative process – Bring together smart people – Get funding from connections in PayPal • Former employees who started VC’s • Start‐up success is very circular—those who succeed  tend to succeed again. • It’s about the network! Published with permission on www.chubbybrain.com Impact of the Day of the Week Both YouTube and  Facebook see  recurring weekly  increases on  Saturday, Sunday,  and Monday (likely  as an extension of  Sunday night). Published with permission on www.chubbybrain.com Impact of the Day of the Week The weekend: Gives YouTube a boost . Gives Facebook a slight bump. Causes Wikipedia a slight decrease in users. Published with permission on www.chubbybrain.com Seasonal Effect on Websites The seasonal effect  on YouTube is  tenuous – could be  extrapolation.   Without research  papers due the next  day in school,  Wikipedia suffers. Published with permission on www.chubbybrain.com Takeaways • Do weekly and seasonal effects have implications for  advertisement cost structure or bandwidth use and  allocation? Published with permission on www.chubbybrain.com Opportunities for Further Analysis "The probability of adopting by those who have not yet adopted is a linear function  of those who had previously adopted.“ –Professor Bass p: the “coefficient of innovation”  q: “the coefficient of imitation” (reflects the influence of previous adopters) The Bass Model is the most widely applied new‐product diffusion model. It has been  tested in many industries and with many new products (including services) and  technologies. 1,200,000 1,000,000 800,000 600,000 400,000 200,000 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 M * cF M * cF_asSumf M * DE_cF_asSum_DE_f 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Published with permission on www.chubbybrain.com Appendix Published with permission on  www.chubbybrain.com Penn vs. Non-Penn ratings of characteristics Importance of Website Characteristics 6.0 5.0 4.0 3.0 2.0 1.0 0.0 Penn Non-penn Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Penn students use all websites more frequently Penn vs. Non‐Penn Usage 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 10+ times per month Daily or almost daily 6‐10 times per month Never 1‐5 times per month Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Penn students contribute less frequently to YouTube and Wikipedia Penn vs. Non-Penn Contribution 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Daily or almost daily 1-5 times (per month) 6-10 times (per month) 10+ times (per month) Never Source: Student Survey, n=260 Published with permission on www.chubbybrain.com Wiki article graph Published with permission on www.chubbybrain.com Wikipedian protester Published with permission on www.chubbybrain.com Published with permission on www.chubbybrain.com Published with permission on www.chubbybrain.com