This morning, I came across this post at the Reach Students blog critiquing the effectiveness of advertising on Facebook. The author states:
When we first experienced poor results earlier this year we looked carefully at creative and planning. Further experimentation saw a variety of quite different offers and creative approaches. What kept us going was the fact that others had anecdotally mentioned good returns from Facebook ads.
Yet our results did not improve.
I’m not surprised. I’ve thought about this some, previously posting an article called “Selling Social Networks,” that explored monetizing social networks. Social network pose a significant challenge for advertisers due to the unique nature of their use. Facebook, in particular, has two primary uses; the first is answering information needs, such as the quick lookup of an email address or phone number. The second is browsing, during which you’re exploring your friend networks, learning, engaging – in a state of flow.
The problem for advertisers in the first case is quite simple; with Facebook as a directory, people enter with a question and quickly leave with an answer. There’s very little opprtunity to interrupt them because they’re in the middle of a task cycle, and generally only active in the site for a short period.
The problem in the second case deals with interestingness. When a Facebook user is actively browsing/learning, advertising’s competition is immense – the user can either continue to look at pictures/profiles of her friends, or click on your lame ad :). Its a pretty simple calculus, when you think about it. Furthermore, as browsers get into the flow of transversing and learning about friend networks, there’s very little that can actively interrupt them.
The final problem is in the context shift. Adsense delivers amazing clickthrough rates because it is situationally optimized for question-askers. When does Adsense perform the best? In search results, when someone is primed because they are in a research state. Facebook simply isn’t analogous to our search environments; it is a unique, new environment. To that extent, new forms of advertising are going to be have to built for the environment.
The best bet for creating this advertising? Leveraging an adsense approach by drawing on simple, native elements for the construction of ad properties. Ads that seamlessly integrate into the profile, or even become part of the profile stand the best chance of being clicked, certainly compared to an out-of-context banner ad. Of course, Facebook must walk a very fine line with any such strategy, as its users are especially mindful of keeping the site minimally cluttered and friend-focused.
I think you’re right that the best instantiation of an advertisement is when the ad is embedded as an extension of the user’s network discovery “flow.” Therefore, the most productive use of resources in targeting a Facebook audience is figuring out a deal to position your ad within the best fitting 3rd party Facebook application.
The confound, to me, is if the best form of positioning for a music label is to buy priority within iLike, that greatly diminishes the value of Facebook’s universal network data (in this case: Music). Although, from Facebook’s standpoint, I suppose the benefit of usage generated by 3rd party apps outweigh the depreciation of universal data.
I think another very important concept to keep in mind is the usage variance amongst different sets of users. The most prominent variance to me is between colleges/networks. Fred, you assessed that roughly 10% of UNC undergrads had their profiles set to full privacy. At Carnegie Mellon, that number is much higher. So if we use your presumption that “flow” state advertising is more effective, advertising directly through Facebook using flyer’s or integrated solutions would be more effective at UNC, whereas it would be significantly more effective to advertise through a 3rd party app that allows more socialization within Carnegie Mellon’s paradigm of network usage.
As an ending gripe, I still cannot understand why Facebook does not allow targeted marketing for their most basic level of marketing solutions, flyer’s. I know I would purchase more ads if I knew my targeting would be relevant primarily to age. With no sample data, I will blindly assert that I don’t see how the economics behind this decision are justified.
Well: it’s not a bad thing click-through has got high enough so that, when ads are not needed, market drives them away. However, I don’t think this means it is impossible to “sell” Facebook: two models seam to work so far, and we can imagine a third:
1. Netvibes sells well designed “applications” to the service owners, and premium placement in the widget catalogue:
> the Facebook ecosystem could have free-lance developers paid by the service (but apparently, their service is so simple there isn’t much money there);
> premium placement in the fast growing catalogue would not have too much interference with the natural process, as most people use friends’ recommendation;
> maybe Facebook could sell ads for the apps to the most central (or any relevant diffusion statistic) members. OK: I’m offering that to get a job at Facebook ;o)
All might be considered tweaks to the “free market” of apps that Mark Z. asked for in front of the developers, but they make sense.
2. Gmail as contextual ads in a similar, very personal context. There is a moral issue to have “Big brother” do the placement, but I can tell you that restaurant ads next to a “where should we eat tonight” e-mails are great–they would be better if their where for good restaurants, though.
3. I’m not sure Facebook dwellers are always busy: don’t they sometime feel like procrastination? Then they are available; wouldn’t a sponsored Stumble-upon, based on social interest be relevant? Maybe Facebook could just try it, openly, introducing it as: “If you feel like giving us some money for our great service, just give us little eyeball time instead–no strings attached–and we’d always think you’d be better off outside having fun anyway.”
Going back to the earlier discussion of class and social networking sites, how do advertising rates on facebook compare to those on MySpace?
How does the style of advertising used by each appeal to the target demographics?
Hey Fred,
Interesting post. Heres another option for social networks:
Social networks could opt to make no money directly on their site. Instead, they could leverage the social graph to improve ad quality on search results pages.
Imagine if the next time you do a Google / Yahoo / whatever search looking to buy something – lets say a bike – you see a note saying “Your friend Bob loves his bike / gave it a 5 star review” etc. Furthermore, the demographic information people have voluntarily entered on their profile would make targeted advertisers jump up and down. You could target single 28 year olds who’ve read Harry Potter. So maybe the real money to be made with social networks isn’t actually within the social networks, but within targeted advertising elsewhere, when people are in context where they’re more like to spend money.
Interesting stuff. Thanks for the posting.
Interesting stuff.
We plan to use AdSense in our social network by making ads appear between lists of user’s friends. But they won’t appear on every page view, but only on like 1 page view out of every 5. This amounts to seeing ads 20% of the time. Hopefully our approach will be unique enough to rid users of ad blindness and provide a better return for our site.
really nice article and interesting blog, it’s very great, congratulations !!!
hmmm, I happened across your blog through blogger’s “blogs of note” and found this post very interesting as I am an avid facebooker. :)
Bertil Hatt is very correct in saying that yes, we are not always busy. In fact, he has it down to a “t” when he mentions procrastination. In my boredom, I have clicked on ads on facebook which look like from reputable companies. (I am a bit afraid to click on a lot of things as my computer is so vital to my grades, that I dare not destroying it with a rogue virus.) So, my boredom and procrastination on facebook may in fact help your concerns. :)
Thank you to share the idea of Adsense website.
Interesting thoughts. Monetizing the social networks is quite a question.