Why I'm Bullish on Facebook

A article earlier this month in the NYTimes talks about how Procter & Gamble has been "dipping its big toes into the vast pool of Facebook, now the world’s largest social network." Who wouldn't want a piece of P&G's ad spend, as they spend almost as much as all companies in the online direct marketing space will earn combined. If Facebook could garner just 1% of that almost $2.5 billion budget, for example, they'd effectively double their revenues.

The Times article reference above is just one on the topic of social media and advertising. Most of the articles I've been following describe the challenges that Facebook and others a) have wooing advertisers and b) have not done a great job monetizing their inventory. Facebook's meteoric growth and valuation, in particular, naturally give rise to equally substantial expectations. Mashable has a great article that tries to answer the question "How Can Facebook Crack Its Advertising Problem," and Silicon Alley Insider looks at the combination of the ad market decline, Facebook's overhead growth, and the impact on the company's value as it looks to raise more money.

As a direct marketer, though, and one that tends to focus on how emerging inventory sources have performed and their impact especially on the online direct marketing landscape, I think we should take a step and appreciate just what Facebook has accomplished. They might not have cracked the brand dollar nut (hint: offer stock to ad agency media buyers, duh), but they've done what only a handful of other companies could do, let alone those who actually executed on it. Facebook has built a functioning (and not entirely disagreeable), robust, self-service advertising platform with reach and targeting along with the appropriate levels of customization and reporting. And, they did all this during an incredibly busy year - constant press be it about the investment from Microsoft, employees trying to cash in on their illiquid options, and constant deal making speculation such as the confirmed attempt to buy Twitter. (For more on the actual mechanics of the platfom see the TechCrunch article on MySpace vs. Facebook)

What makes Facebook's self-service "flyer" advertising platform all the more impressive is that they didn't build it with actual advertisers in mind. Given their initial focus on high dollar CPMs, they didn't look at their self-service platform as their parallel to Google AdWords, an ad marketplace providing liquidity for all page views. They thought vanity (people advertising their personal or business profile pages) would drive most of the sales. Real money would come in through the "normal" sales channel, i.e., the ad sales department, and this might bring in incremental revenue much the way their virtual gifts business does. Lucky for them, the system they built has enough flexibility so that when they realized that real advertisers wanted to spend money and promote more than just their profile page or fan page, they could accommodate.

For those who haven't tested it, and there is no reason not to (what those reasons would be is another article itself that follows the Google account evolution), the Facebook program is a combination of Google and Yahoo. It contains both automated elements and a human review process. Unlike Yahoo, their human editors respond faster and their tends to be a greater level of flexibility in what you can advertise - key to a new platform. They have guidelines and will not approve ads, but they haven't set those guidelines too strictly that it would prove a disincentive to those most likely to put down money, direct marketers.

Let's first think of all that they've done well, which in itself will unfortunately not do justify to the level of work actually involved. First and foremost they have a pleasing interface, an anything but trivial endeavor in which you must think about such questions as - which columns do you show; what is the default view; what are the other views; how do you manage the picture upload process; what is the flow for setting up the ads, reviewing them, managing campaigns, modifying bids, etc. I can guarantee you they had a product document 1000's of pages long, and that's just visually before getting into the meat of coding it to match what the product team wanted to look like. The next step, not any easier, is the adserving and optimization rules. Showing ads isn't all that tough, but when you have more than one placement on a page (they show up to three as of now), you start to run into challenging questions for even one slot. For example, you must think about how you determine the number of ad impressions to allocate for each text ad along a problem that gets more complex as the total audience size changes (such as when an advertiser adds targeting).

The optimization layer is what makes them (and almost every site the real money). It is how they squeeze out the greatest yield possible from their advertiser base. Done poorly, they will show under-performing ads too frequently, not recognize good ads, and miss out on properly motivating advertisers to increase their spend - both in total dollars and per click / impressions. Done right, you pull off financial results like Google.

The optimization layers is key, and for a first pass, Facebook has done a good job. The reason I remain bullish is thinking about all the areas where they can and will improve. Think about Google and all of the tools and levers they built over the years as they evolved their system. With Google as with Facebook, the adage, "Rome wasn't built in a day," more than applies. Off the top of my head, here are but a handful of things that Facebook has yet to tackle - no advertiser only interface (must have profile on site; limits scale), no pixel tracking, no optimization by performance (all ads deliver evenly across budget; limits yield), no API for advertisers (very manual, limits scale), no notion of quality score, i.e. room to improve their minimum bid algorithm, and no option for display ads (ala Myspace) or other animated ads. If and when they start to tackle these things, we will probably find their overall earnings double if not triples at the very least... and from the same advertiser pool. When they start to perfect their targeting (better geo targeting, contextual mapping) which allow for them to service a greater pool of advertisers, the sky's the limit so to speak. The downside to all this potential on Facebook is easy. They could do the same thing that Google has done. Earn themselves a lot of money while managing to piss off the rest of us.