Facebook Case Study: Offline behavior drives online usage

Facebook Case Study: Offline behavior drives online usage

written by Nisan Gabbay on November 5th, 2006

Why profiled on Startup Review

Facebook was launched in February 2004 by Harvard undergrad students as an alternative to the traditional student directory. Its popularity quickly spread to other colleges in the US by word of mouth, and the site now registers close to 15M monthly UVs and over 6B page views per month. Facebook has completed two rounds of venture financing at very high valuations, the first at a valuation of ~$100M and the second at ~$550M (valuations are unconfirmed). These valuations were driven by the multiple acquisition offers that Facebook has reportedly turned down (the latest was a rumored $750M offer). Facebook is already generating significant revenue, so despite all the valuation and web traffic metric hype, it has also established a very real business.

Interviews conducted: Noah Kagan, early product manager for Facebook. Noah will soon release an e-book on Facebook, with good insight on the social networking space. You will be able to download the book at Noah's blog, okdork.com. I have had plenty of informal conversations with people close to Facebook over the last two years, while not formal interviews, I would regard these as quality sources – employees, investors, and competitors.

I would also like to thank Nick Macey, a student at the University of Utah for helping in the research and writing of this case study, and providing the ever valuable user perspective as a current college student. Nick will be helping with some of the writing on Startup Review in the future.


Key success factors

Provide pre-existing offline community with a complementary online service

Facebook had its initial success with college students by providing an information service that was not available offline – an interactive student directory containing each student's class schedule and social network. Before Facebook added the feature sets it has today, it was simply a more complete student directory. Facebook did not create a community where one never existed before; rather they provided an important information and communication service to a pre-existing offline community.

While students already had a loose affiliation with all fellow students at a college, they didn't have an easy way to learn more about their fellow students outside their direct social network. Given the large class sizes at most universities today, students don't have the opportunity to interact with very many of their fellow classmates during class. I remember the days I spent at Berkeley in 200+ student lecture halls scanning the crowd for attractive girls or previous acquaintances. Facebook organized students by class schedule for the first time, making it possible to learn more about that classmate you might have a crush on. Although I am highlighting one particular use case, initial Facebook usage was indeed driven by dating type activity – checking people out, learning more about crushes, light stalking type of activity, etc.

The larger picture here is that Facebook created a high utility online service for enabling pre-existing social behaviors within an offline community. This makes for an interesting lesson learned: it's easier to piggyback off a pre-existing community with offline behaviors that drive online service usage.

Restrict user registration (and other behaviors) to build desired online service

Facebook made important product decisions that ensured harmony and trust between the offline community and the online service created. Facebook originally limited membership to those users who could verify they had a ".edu" e-mail address for the college they attend. Facebook also placed limits on the ability to search or browse users to the college that the user attends. These measures aim to make users feel that the site is exclusive and limited to members in their offline community (colleges and universities). In the early days of Facebook, something like 30% of users actually posted their cell phone number on their profile. I'm not sure whether this statistic is still valid, but it supports the notion that users trust who is viewing their profile.

Facebook has recently opened its doors to users outside the .edu networks. To accomplish this, they have created "networks". High schools, employers and geographic areas are, essentially, what colleges were to the original Facebook. When you join one of these "networks," you can only view others in the self-designated network. Additionally, Facebook has implemented a number of privacy controls that allow users to control exactly who gets to see the information they provide.

Aggregation of a series of deeply penetrated micro communities

Facebook is a more compelling advertising opportunity than other social networking sites because of deep penetration within a series of micro communities (college campuses). If a local advertiser wants to target a particular college campus, Facebook is the best way to get the advertiser's message to that audience. CPM rates for local advertising command a significant premium from advertisers because of their more targeted nature. With 65% of users logging in daily and 85% weekly, advertisers can run time-oriented campaigns very effectively. The large, branded advertisers, who value reach, can advertise to nearly every student in the 18-22 demographic in the US with one campaign.

Facebook will have ample opportunity to diversify its revenue streams beyond traditional banner advertising due to its deep penetration in these micro communities. Having the attention of 90% of students attending a university lends itself to online classifieds, event listings, e-commerce, and lead generation. Facebook should be well-positioned to be a major player in online classifieds given the usage patterns of its user base.

Built strong brand recognition amongst user base and advertisers

The key to an online advertising business targeting branded advertisers (advertisers looking for branding, not just clicks) is having a strong brand that advertisers want to be associated with. A perceived hot brand is what drives premium CPM rates. Two sites having similar demographics and user usage patterns may have drastically different CPM rates based solely on the perceived brand recognition and image factor. While some people I spoke with disagreed, I believe that Facebook did a masterful PR job – highlighting the impact that Facebook has made on the lives of college students and their online media consumption in nearly every story written. How often do you hear that 90% of Facebook users login to the site once per week? Clearly the PR coverage came as a result of the tremendous viral growth, but capitalizing on that PR to help build brand was a key success factor.

Founder(s) credibility with college audience

The "face" of Facebook is Mark Zuckerberg. Back in February 2004, when Facebook was founded, he was a student at Harvard. Two other students, Dustin Moskovitz and Chris Hughes were the second and third employees of the company. This added a level of credibility to the site in the minds of the student users. It was something one of them had created, not something fed to them by a "company" in the traditional sense. It was a place that they could trust because one of their own had made it.

Adding to the underground feel of Facebook was the viral spread of the site. It fanned out throughout Boston, and then the Ivy League. Students at other schools had to wait in line until Mark and friends could find time to add their school. This created even more buzz around the product.


Launch strategy

Prior to launching Facebook, Mark Zuckerberg had experimented with a number of different web products. In fact, his first attempt targeted at the Harvard student body was called FaceMash, which drew criticism from the University and some students, prompting Mark to drop the service.

Mark launched Facebook (at the time called thefacebook.com) in February 2004. Once the site was ready for users, the Facebook founders blasted e-mails to Harvard students to let people know about the site. The team had access to the e-mail addresses of Harvard students at each dorm. Thus e-mail marketing, viral feature sets, and word of mouth was how Facebook was launched. Given the immediate positive reaction that Facebook received at Harvard, Facebook began rolling out the service to other universities. Facebook did not use a targeted geographic roll-out strategy in the early days, they received registration requests from students at other schools, and then prioritized which schools to open based on the number of these requests. Interesting to note that this is how Craigslist rolls out to new cities – based on user requests.

From what I understood, Facebook did not receive any help from the schools themselves to promote the Facebook site to the student body. If anyone has evidence to the contrary, please leave a comment below.


Exit analysis

There has been much speculation in the blogosphere and mainstream press regarding who will buy Facebook and for what acquisition price. I have heard from reliable sources that Facebook did indeed turn down acquisition offers for ~$750M earlier this year. Recent reports have claimed Facebook is in acquisition talks with both Yahoo and Microsoft for ~$1B. Is such a lofty valuation for Facebook justified? It all depends on an evaluation of future growth prospects, but I think that there is a misconception in the blogosphere that Facebook is not generating much revenue. On the contrary, Facebook was generating almost $1M per week in advertising revenue in Q1 2006. It is likely that Facebook will generate ~$50M in revenue in 2006, up from ~$10M in 2005. Some reliable sources believe that Facebook will do ~$200M in revenue in 2007. Given that Facebook has been guaranteed $200M in revenue over three years by the Microsoft advertising deal, the 2006 and 2007 revenue numbers seem attainable. If the 2007 revenue goal of $200M is reasonable, a 5X forward revenue multiple does not seem to be an excessive valuation multiple.

Many people also point to the fact that Facebook is considerably smaller than MySpace from a site traffic perspective and hence should have a lower valuation than the ~$500M that MySpace was purchased for. This type of comparison based on unique visitors and page views is clearly flawed because not all page views are created equal. There are several good reasons why Facebook's page views are more valuable than those of MySpace:

1) Facebook's core user base (college students) is more desirable than MySpace's core user base (teenagers). Because college students have more disposable income and are more likely to have credit cards than teenagers, they are more desirable from an advertiser perspective.
2) Facebook represents a more compelling local advertising opportunity than MySpace because Facebook can guarantee deep penetration of college campuses, whereas MySpace cannot show the same types of local market usage patterns. The CPM rates for local advertising campaigns are typically substantially higher than national campaigns because of their more targeted nature.
3) Facebook is viewed as a safer option than MySpace for branded advertisers, as Facebook has a less racy image than MySpace. In a market where advertisers are still hesitant regarding user generated content sites, Facebook has done a better job of brand positioning.

Another noteworthy part of the Facebook story is how they masterfully handled the VC financing process, limiting the amount of equity dilution to the founders. When Facebook raised its first VC round of financing in April 2005, they negotiated a pre-money valuation of ~$85M at a time when they were generating less than $500K per month in revenue. Facebook was able to command such a high valuation by courting both VCs and potential acquirers simultaneously. With term sheets in hand to be acquired for $85M, Facebook was able to drive up the pricing on the VC round. I remember discussing with VCs who participated in the bidding for that first round how the price, which originally started at a $20M pre-money valuation, just kept climbing week after week until Accel Partners finally won the deal at ~$100M post-money. Hats off to Accel Partners for accurately assessing the potential of Facebook in those early days. The prevailing wisdom from other VCs was that Facebook would probably be capped at a $200-300M exit, and hence a 2-3X return was not high enough to justify the risk, given the youth and inexperience of the Facebook founders. Accel is likely to make a 8-10X return on its initial $13M investment in just 2 years. Facebook's most recent $25M round was rumored to have taken place at a $550M valuation after turning downing a $750M acquisition offer. Once again, the Facebook management did a great job of creating a competitive environment for their second VC round. The one piece of information I would be curious to know is how Facebook's $500K seed round of financing was structured. That investment was done at the end of 2004 by ex-PayPal exec Peter Thiel (when Facebook was available on ~30 campuses). I'm guessing that $500K bought 5-10% if it was structured as equity, but would have bought considerably less if it was structured as convertible debt. If anyone can shed some light on the seed round, please leave a comment below.


Food for thought

The Facebook success story is most interesting to me because of how daily offline social behavior drove usage of the site. There are plenty of activities in our daily life that could benefit from a complementary online product. However, if that offline behavior only occurs once every few months, you have the challenge of user recall. Namely, will users remember your service and know how to find it to fill their need. Facebook demonstrates you have a great Internet service if offline behaviors can drive nearly daily usage online. In the life of a college student, you are meeting or interacting with new people nearly every day. It is human nature to be curious to learn more about that person, hence you jump on Facebook. Facebook fills a high value need for college users on a nearly daily or weekly basis, consistently reinforcing the utility of the service, and building goodwill with users. The issue of user recall is an import one for a web entrepreneur to understand, particularly if the need they are addressing occurs infrequently in the lives of their target users.

Another lesson that Facebook reinforces is the importance of brand and PR buzz to advertising rates. The amount branded advertisers are willing to pay for online advertising is hugely subjective – it's still more art than science. To get premium CPM rates, entrepreneurs must establish a brand – not only with users, but also with advertisers. Many social services do not have high click thru rates on ads because people are not in the mind frame of looking for information when they are using a social service. All social networking sites suffer from this "lack of click thru" problem. While immersive advertising opportunities will eventually displace banner advertising on most social services over time, for the time being, traditional banner advertising is still a critical revenue stream.

Finally, we can learn a lot from Facebook by how they built initial trust between users and their service. While these days it is easy to build a consumer Internet product, establishing trust with users is not. As an entrepreneur, how quickly you can establish trust with your users can be a critical success factor. Facebook built immediate trust via the home page by showing only a select few colleges as being open to registration. Coupled with the registration process, users immediately understood that the site was exclusively for use by college students. This made them feel comfortable disclosing information that people normally wouldn't post on the Internet. Simple, but very powerful. Facebook does give users control over the information displayed on their profile and to whom it is displayed, but only a small percentage of users actually change the default settings. Thus, the key part of the trust equation is not features, but branding and messaging about the service and who uses it.


Reference Articles

"Facebook's Critical Success Factors", on Fred Stultzman's blog Unit Structures, May 17, 2006
Great blog post that discusses 5 key success factors for Facebook. Unlike my analysis, this post has some detail on features that made the product sticky.

"Facebook – The Complete Biography", written by Sid Yadav on Pete Cashmore's Mashable! Blog, August 25, 2006
A detailed overview of Facebook from many perspectives, but provides a particularly in-depth look at the product itself.

"Inside Facebook: Life, Work and Visions of Greatness"
, an e-book by former Facebook engineer Karel Baloun.
I thought the book was an interesting read from a human interest and start-up culture perspective, but could have discussed the company more from a business perspective. There is some good discussion on the vision and future direction for Facebook, how the company does and will make money, and how they've built trust with users.

"Dot-Com Bubble: Why It's So Hard to Value Social Networking Sites"
, from Knowledge@Wharton (University of Pennsylvania business school), October 4, 2006
Article has a good discussion on how to value Facebook and social networking sites. This article serves as a source for valuation and revenue numbers.

"Facebook on a roll, stay tuned"
, by Matt Marshall on SiliconBeat blog, August 29, 2005
Has some good insight and quotes on why the site has been successful

2 Comments »

  1. […] Today, Startup Review posted a great case study of Facebook, one of my favorite Web 2.0 companies. One thing that has always impressed me about facebook is the trust they have built with users. The case study talks ends its article with this paragraph- Finally, we can learn a lot from Facebook by how they built initial trust between users and their service. While these days it is easy to build a consumer Internet product, establishing trust with users is not. As an entrepreneur, how quickly you can establish trust with your users can be a critical success factor. Facebook built immediate trust via the home page by showing only a select few colleges as being open to registration. Coupled with the registration process, users immediately understood that the site was exclusively for use by college students. This made them feel comfortable disclosing information that people normally wouldn't post on the Internet. Simple, but very powerful. Facebook does give users control over the information displayed on their profile and to whom it is displayed, but only a small percentage of users actually change the default settings. Thus, the key part of the trust equation is not features, but branding and messaging about the service and who uses it. […]

    Pingback by Facebook Case Study » Personal Insights on Web 2.0, Blogging, and Business — November 7, 2006 @ 12:26 am

  2. […] A lot of entrepreneurs are secretive about their plans, particularly when operating in stealth mode. Success in life is not about ideas, it is all about execution. YouTube/Facebook/Yahoo/Delicious succeeded by executing on an  but the truth is that the reasons these companies have succeed is the execution of the initial idea. […]

    Pingback by Ideas are cheap « On medicine and technology — November 7, 2006 @ 2:41 am

http://www.startup-review.com/blog/facebook-case-study-offline-behavior-drives-online-usage.php

Leave a Reply