Why Building a “LinkedIn Killer” on Top of Facebook Will be Tough

In the spirit of full disclosure, I should preface this post by noting that a friend and I tried to build a jobs-related product on top of the Facebook platform fairly recently and ultimately decided not to pursue it. In the process, we learned a few things that I think are worth consideration for anyone broadly looking at building utility (as opposed to entertainment) applications on top of Facebook, including “LinkedIn killers” and other job discovery products such as Identified, BranchOut, and a few others that have yet to be announced. Someone will ultimately succeed here – it could be Facebook itself or an upstart.

Before going further, I’ll summarize why I think building a professional networking product or jobs product on top of Facebook is a big opportunity:

*Jobs, as a category, monetize really well – people are willing to pay to get to the right candidates
*Jobs, as a category, has supported multiple winners on the Internet (HotJobs, Indeed, SimplyHired, CareerBuilder, etc)
*LinkedIn is already a large, interesting, pre-IPO company with a clear value proposition
*Facebook is a daily use application for hundreds of millions of users; LinkedIn is not (although they’re trying to become more daily use with Twitter integration, questions, events, and other features)
*Many of the younger, up-and-comer folks spend more time on Facebook than they do on LinkedIn
*Competing with LinkedIn on the open Internet would be really hard; being on Facebook helps you address the customer acquisition problem.
*There is a growing segment of people want to use Facebook for professional purposes and FB as it stands today simply lacks the necessary functionality to do so.

Despite all of those good reasons to pursue jobs or professional networking on top of Facebook, I think there are some hard problems to solve. Having thought about this for awhile and ultimately decided not to pursue it, I’ll rank my concerns in order of decreasing importance – I think the “Facebook could do it” argument is the least compelling of all. Hat tip to my friend David King (@deekay) for helping me think through that last part earlier today.

Problem #1: Customer Acquisition – Cost and Difficulty
Once upon a time, it was fairly cheap and easy to acquire users on Facebook. There were tons of viral channels (requests, notifications, forced invitations, other forms of unsundry sneakiness, etc) – that era has passed. I would not say that virality on Facebook is dead, but it is much harder to have a fast-growing application that is not supported by advertising spend today than it was 6-12 months ago. And the applications that have seen rapid growth in the current environment tend to be entertainment and games applications. There are lots of people on Facebook who are looking to kill time and are willing to try new, fun things that their friends send their way or that they discover via the newsfeed, an ad, or some other mechanism.

Perhaps I should be a bit more granular and specific to jobs here. Let’s assume that the people who would be the targets of a professional social networking application do not “live” on Facebook (they don’t have it open all day long and are not constantly checking it). Ideally, you would not want to rely on those individuals changing their behavior and becoming more active Facebook users for your business to succeed. What’s the clever hack? You want to be able to communicate with them through other channels (email, in particular) where you can catch them when you have something to tell them. Meet them where they are, which is the email inbox. Given how buried most of the on-platform discovery options are (no more notifications, requests are buired), trying to rely on those mechanisms is unlikely to work. You can do wall posts and what not, but I don’t view that as a long term sustainable strategy.

So, in a world in which customer acquisition is difficult to achieve for free or at low cost, what’s the answer? Well, you can obviously spend money on advertising to acquire installs. So long as you know or believe that the cost of customer acquisition can be funded profitably by future revenue (or CPA < LTV), it could all work out. For example, if you are distributing jobs with $10,000 referral bonuses, you can build a sane business model along the following lines: * Focus on jobs with referral bonuses greater than a few thousand dollars * Split the referral bonus between the candidate and/or referring party and keep some margin for yourself * Invest the margin you capture in customer acquisition to grow the network For a job with a $5,000 referral bonus (not uncommon in tech), a service that kept 20% of the bonus would have $1,000 to contribute to customer acquisition and customer overhead. If the math works out, the only real risk you have is making sure you have enough money to invest in growing the network before you start to reap the benefits of the captured referrals. Problem #2: Usage Model – It’s Not Games
By far, the most successful category for Facebook applications has to be games. Just look at the revenue numbers that companies such as Zynga (disclosure – I used to work at a company that Zynga acquired), Playdom, Playfish, Kabam, and others are able to generate. One of the nice things about games companies is that they are the perfect application for people who have lots of time to spend on Facebook. For games, the challenge is not getting people to engage – the challenge is getting your engaged users to monetize.

Let’s think about the kind of people who you’d want for a professional networking application focused on white collar jobs. You want people with strong educational backgrounds, with good past employment, and other indicia of employability. Those might not be people who are spending all day long on Facebook. A lot of busy people I know don’t “live” on Facebook the way many power gamers do. They check in periodically to see what’s up, respond to friend requests, check the occassional baby photo, and they’re gone.

How do you build an application usage model that engages them? Can you do it with a Facebook-only experience? I don’t know. But I do know that there are a lot of busy professional people who are not actively engaged with Facebook – making this service work for them is going to be a challenge.

There are actually two things to consider here. For those who are actively considering new jobs, they will usually go where the jobs are. If there’s a good job board or service on Facebook, they’ll use. But that only applies to active candidates. One of the magical things about a good professional network is that it helps route jobs THROUGH passive candidates (people who are not looking or are otherwise happy in their job(s)) to those who are open to new opportunities. For a service to really work, you need to have a way to engage these passive candidates as participants. Solving that latter problem is a big deal – it would make a service super useful as I’d estimate that there are 7-8 passive folks for every active job seeker.

Problem #3: Is This Core to Facebook?
I can’t answer this one. I don’t work at Facebook and never have. But I think of Facebook as being interested in two big ideas. One big idea is having the world’s best, most complete social graph of all facets of your life (personal and professional). The second is building the world’s largest personalized, contextual advertising business powered by a vast collection of data about you, your preferences, and your friends. On the one hand, having a third party build a professional networking layer on top of the existing social graph threatens the integrity of Facebook’s stranglehold on the social graph. But is that more important than making Places and Pages successful? More important than establishing Credits as a default payment mechanism on their platform? What about distributing Facebook Connect as the default ID / login system across the web? In short, Facebook can’t do everything. No company can. They have to pick the areas where they want to concentrate their firepower. Is jobs an professional networking in the sweet spot? I simply don’t know.

Hope you enjoyed the post. Feel free to leave comments below.