BLOG

Target Practice: Graphing Facebook User Segments

By

Social Media

It’s all the mainstream chatter & rage: “The Facebook graph this” or “social graph THAT.” Serving ads to this mystical diagram of human interactions (and the profiling-variables offered by fledgling ad platforms) are the voyeuristic dreams of heat seeking guided missile marketing freaks. “Points” on the Facebook graph, available to marketers for profiling customers, are still limited though exciting.

Facebook-1

The FB self-serve tool itself is a multi-layered lens peering into Facebook’s noisy black box village, a happy harbinger of social ad platform things-yet-to-come. I thought we’d take a look at what attributes are actually available today on a node of the first do-it-yourself social graph ad serving platform and share a tasty bit of the interface.

If you’re into the 18-34 crowd, this is little puppy is candy. Keep in mind that the FB ad platform is young and feature-impoverished. However, it offers an exciting glimpse into future of customer-level profiling. The Facebook algorithm ranks PEOPLE for CONCEPTS as categorized by KEYWORDS in nasty little focused grids which includes friends and social-actions. We’ve had a good look at ROI and we like it.
——————————————————
Make sure you’re logged into Facebook. Go to www.facebook.com/ads , Click “Get Started.”

Type the URL to promote. (If you don’t have a Website, “Facebook Pages are a free way to represent your business, product, band, etc. on Facebook-for now type in a URL)

Facebook-1

 

The next screen you come to is the heart of FB demographic research. The default age is “18-any.” Change age range to “any to any” for a glimpse of how many users are in the matrix. Today there are 22,366,340 FB users the United States. Change the drop down menu to “Canada” and learn that there are 8,075,980 Canadian players on the gird.

Check “male” or “Female” boxes in tandem with different age ranges. See how it affects the count.

Change “Everywhere” option to “States” for a look at how that grid setting affects the count.

Typing in keywords is different and touchy. Think of it as delimited Ajax keyword suggestions. Type slowly starting with the first letter until suggested FB categories auto-fill, click on the word, and then select another. Be aware that the interface is rather flakey.

Target employees of specific companies in your cross hairs. For instance, here we aim at news the feeds for users between 19 and 36 years old in Minnesota who work at Best Buy, Target, Wal Mart, or Circuit City Stores, Inc. :) Yay.

Facebook-4

Let’s reach women between 14 and 18 years old in Massachusetts, Rhode Island, New Hampshire, Vermont, and Connecticut who are in high school and like Soccer, Playing Soccer, Play Soccer, Indoor Soccer, or Intramural Soccer.

Facebook-5

Don’t Interrupt Me Please
There’s debate as to the effectiveness of “interrupt vs. search,” however this much we know: Life itself is a complicated matrix of social actions, interactions, communications, and jam-packed with rich WORDS people use to describe things. Our checkerboard patterns are reflected in online behavior.

The Future is Here (soft of)
While Facebook people write, read, and consume endless words, they’re memorialized by massive databases that “remember” each users affinities and behaviors. Facebook’s new social media based ad platform let’s advertisers identify and target very specific combinations culled from users’ profiles. . The Facebook algorithm ranks PEOPLE and their friends for KEYWORDS and actions in focused grids. It’s all the mainstream chatter & much more than a novelty. This is a glimpse of the Internet’ s evolving promise.

-->

13 Comments

  1. Todd Mintz on December 12, 2007 at 5:37 pm

    I must really suck at this because with campaigns that made big $$$ elsewhere, I can’t even get more than a few clicks on Facebook despite being able to target the right audience.

  2. Marty Weintraub on December 12, 2007 at 5:57 pm

    Todd, give me a call and we’ll take a look at it. Not every product works on FB. Some go crazy.

  3. Marty Weintraub on December 12, 2007 at 5:57 pm

    …and I highly doubt you suck :)

  4. michael streko on December 13, 2007 at 12:40 pm

    Its touchie with what works on the platform, and also know that its not going to last. I recently blogged about my first love for facebook then the abrupt end it came to when my clicks dropped and then my exposures turned almost non-existent.
    IMO 89% of the facebook users are not ones to whip out there wallets and buy something they see an add for. I have had the best of luck working with leads on the platform. But not long leads, email and zip submits work well and some occasional 5 fielders. When the campaign initially launched I was doing great, but it seems as if i had burnt through all of the scope of my users so the campaign came to a halt. Remember; its not the internet, its only facebook. So the traffic is not indefinite.

  5. Marty Weintraub on December 13, 2007 at 1:43 pm

    Michael,
    I am marketing to a pool of about 250K FB users for a career college niche’ product. The conversion of traffic to sales is insane-nearly 10%. It’s impossible to generalize because some products in the 18-24 demo work really well. I can see that many products won’t.

    FB is like any channel. Test, test, and test some more. It took us several attempts to figure it out. In our business, it feels just as good to rule channels out. I guess we got lucky on this one. :) Most importantly, it’s a harbinger of future social ad platforms.

    Each social channel will have it’s own uniqure characteristics to evaluate, rule in, or rule out. I’m sure we will not market Tony the Tiger Frosted Flake spoons on LinkedIn. The product we’re selling on FB works really well there right now at 25% of Google’s PPC cost, with higher engagement, and much higher conversion .

  6. michael streko on December 13, 2007 at 1:57 pm

    Marty,

    My point is give this campaign another month, those sales will be gone. If you ad up your daily exposures and clicks more then likely you will burn through your 250k users a bit quick. Just know when to stop or redo the campaign / change creative and so on. I have tested the hell out of FB – once you stop getting clicks your exposures drop rapidly. The best way I have found to cure this is to just make a new add on a different campaign and launch it. This will keep things fresh, it seems to me to be more in tune to fresh ads and new content on ads then long term ads.

    fb’s ad platform is a wave, ride it while you can.

  7. Marty Weintraub on December 13, 2007 at 2:47 pm

    Ah, that sounds like excellent advice. I want to point out to our readers that it was my discussions with Michael that led aimClear to FB. So the game is to watch it carefully and the minute that the impressions lag, pull the ad and change it out. I’ll report back :)

  8. michael streko on December 13, 2007 at 3:35 pm

    you got it! Change Change Change… That is the secret. ;)

  9. Tim Nash on December 13, 2007 at 3:57 pm

    We did a couple of posts on the new Facebook ads over at Payment Blogger, interesting as they were after a short while we quickly reached a saturation point where the cost per click was eating to much into the profit margins. Check out some of our Facebook Results, interestingly a similar campaign using Stumbleupon ads on a slightly modified landing page far exceeded that of the facebook ads and was a lot cheaper.

  10. Marty Weintraub on December 13, 2007 at 4:06 pm

    @ Tim: Thanks, going there now. One point, this client is used to paying $10.00 per click with 5% conversion. “Cost” is a relative thing and is sometimes less important than FINDING bodies to market to.

  11. Tim Nash on December 13, 2007 at 4:15 pm

    True, but it depends on what your selling and to whom, I love the ability to target users and in particular to specify very tight demographics but this comes at a price when your paying $15 a click even if your conversion is 100% its not going to help if what your selling is $14.99.

  12. Marty Weintraub on December 13, 2007 at 4:28 pm

    Absolutely, this client is paying .80 per click on FB max (right now), converting 25% (about 50 leads per day), of which 1% will go on to make a purchase worth 40K over 2 years.

    Most clients do not have this luxury no doubt. :)

  13. michael streko on December 13, 2007 at 4:47 pm

    The first day it launched I ran a targeted lead campaign that could generate up to $60.00 per lead. Usually I am used to PPC reaching as high as $12.00 a click – it was a relief to actually get clicks for $.35 – the conversion rate was “ehhh”. I had initially ran this before facebook had the ad system setup, it was called “Flyers Pro” – so for 5 days i had it up i spent $250.00 which turned into 86 leads which actually worked out well across the week, not as well as my clicks and exposures but still well. Since they have implemented the new system I have not been able to recreate this success. Even with a higher PPC bid and a larger budget I watched it as nothing more then a black hole for cash.

    For leads that require Email and Zip or 5 Field form, i rotate creative weekly and its been working great.

Post a Comment

*

*