Posted: March 24th, 2011 | Author: Erica | Filed under: Social Media Platforms | Tags: Edgerank, eyeballs, Facebook, likes, views | Comments Off
I tried out the “10% of your friends see your posts & 1% like ‘em” experiment, because I’m pretty curious about the way top content is selected in Facebook & wanted to see Edgerank—Facebook’s scoring system for that selection—in action.
Responses = 26% of friends.
This is a lot higher than average. Which may mean:
- Nicer, chattier friends who respond to requests like this, or
- My posts are showing up a lot because
- I interact on Facebook “more than average”, or
- The amount of action on the post kept it visible
The last comment came in 19 hours after the original post, suggesting it was dropped from ‘top news’ at that point.
The experiment is predicated on the goodwill & motivation of your friends, and is pretty much gaming ‘top content’, which is partly selected on amount of interaction a post receives. I also garnered 18 comments, which are weighed more heavily by Facebook Edgerank than likes are.
Two friends who ran the experiment got 10.3% + 4 comments and 2.2% + 6 comments respectively.
I’m not sure what this experiment proves beyond the 10% eyeballs theory isn’t always accurate, and that my friends are really nice.
Edgerank, if you’re curious (& I don’t pretend to know everything about this), is a mix of these factors:
- Recency - which is not a word; can you believe that? Consider it coined. Newer = more weight
- Interaction - more likes = more weight; more comments = even more weight
- Affinity – the more you’ve interacted in the past with the person who posted, the more likely you are to see their stuff again. Facebook took this one way too far recently, but the concept makes sense. in a please-mediate-my-world-for-me kind of way.
Posted: March 19th, 2011 | Author: Erica | Filed under: Social Media Platforms | Tags: asynchrony, button, comment, Facebook, revert, submit, UX | 1 Comment »
I’m usually all for Facebook changes, but removing the button that submits comments & using “enter” instead counters web conventions to the point of causing errors. I don’t think it’s going to be intuitive to most users.
The idea was to make the commenting experience mimick instant messaging, where you hit “enter” to post your comment (and “shift-enter” to create a new paragraph). Psychologically, this real-timieness is supposed to make you feel closer to your friends. Marketing-wise, this enforces your desire for Facebook.
But web-wide, comment fields don’t behave like that. I can just hear the angst out there as half-written, unspellchecked comments are accidentally posted.
I actually think Facebook should work to maintain the opposite feeling. Asynchrony is part of the Facebook experience that I really like. I want to mosey past the general store & chat with the folks chewing blades of wheat in rocking chairs outside. I’m not sure we want to bring the pressure of Twitter’s highway past Facebook’s front porch.
Here’s a bunch of browser extensions (let’s call ‘em fixes, hacks, ways to revert back to having a button) to slow things down a bit.
Posted: February 10th, 2011 | Author: Erica | Filed under: Social Media Platforms, Tips, Tricks, How-To's & Top 10's | Tags: Facebook, news feed | 10 Comments »
Ed. note: this tweet impressed upon me that readers may think the list of apps pictured below that I’m blocking notifications from are my apps.
I assure you, they are the apps of my elderly relatives and elementary school acquaintances!
Amid all the shouting about Facebook’s Pages updates, a bizarro but profound change happened to the news feed.
Facebook is hiding most of your friends & Pages updates, and only showing the few users/Pages you interact with the most.
This has the effect of making Facebook seem darn near abandoned in the past few days, which can’t possibly be what the aging social network is looking for. Users at this point need to be assured that Facebook is still a busy, happening place, or there’ll be no one left to sell stuff to.
The feed modification instead gives the appearance of all but the most active people up and leaving, resulting in a boring, mildly confusing, and certainly not cool vibe. This is a weird choice at a time when people are reacting with dismay to the faintest suggestion that Facebook might buy Twitter, bringing it’s (presumably) uncool culture with it.
Get reacquainted with your friends by scrolling to the bottom of the page (challenging because of the autorefresh; you’ll see) > Edit Options > All friends. Refresh.
Posted: January 17th, 2011 | Author: Erica | Filed under: Social Media Platforms | Tags: Facebook, privacy | Comments Off
Update: Facebook received some “feedback” over the weekend, and changes are afoot. All I see is PR—they don’t say it won’t be happening, just that people need to be made “more clearly aware” that they’re sharing this data. @JulesPolonetsky—Co-chair and Director of the Future of Privacy Forum, former Chief Privacy Officer at AOL & a great guy to follow if you’re watching the privacy issue—says to hang on.
When you start using a Facebook app, like games and quizzes, you typically click some sort of “allow” that lets the app access your personal information.
Facebook will now include your home address & mobile number in the information handed over to the developers of these applications. Your friend’s numbers & addresses won’t be included.
Some are questioning the Friday evening timing of this announcement, and some are encouraging people to remove this data from their profiles before bad things happen to it.
Facebook, on the other hand, is coming up with ever-easier one-click methods of squeezing more specific location & personal data from users. I spotted this “fun” quizvertising a day or 2 before I heard about the change in app permissions.
I underestimated it as merely pesterous, hamfisted data-mining before I understood just why they wanted to know.
Taken together it sounds like Facebook really wants to offer advertisers this data. Crank the dial on the privacy metre from “annoying” to “ominous”. Your social norms have been warned.
Posted: December 31st, 2010 | Author: Erica | Filed under: Social Media Platforms | Tags: Facebook, page admin | Comments Off
Remember the mid-December excitement when Facebook went down & top-secret page admin features were “accidentally” revealed? Here’s the gist of my delirium:
It was a glorious day. I was suddenly, magically, prayer-answeringly able to log in to Facebook as either myself or the pages I admin. Facebook was crashing all around us, though, so there wasn’t time to experiment, 
but I made a quick screenshot, lest these super powers turn out to be all a dream.
The chatter is that page admins will be able to comment on other pages as the brand they represent, but my New Year’s Facebook Wish is for much more than that. I hope we’ll be able to interact with our fans on our walls as either people or brands.
This would free us up to say things our brands can’t, add a human feel to brand communications, and make things not look so lonesome when there’s no conversation happening (if you have 5-10 admins in 1 organization, that’s a lot of people who can’t contribute to the conversation individually).
Fingers crossed for 2011!
If that was a little prosaic of a fantasy for your tastes, check out my New Year’s wish for all humanity. Happy New Year!
Posted: December 21st, 2010 | Author: Erica | Filed under: Social Media Marketing, Social Media Platforms | Tags: data, demographics, Facebook, stats, Twitter | Comments Off
Marketers know that more targeted efforts equal higher conversions. The more you know about your audience, the more you can appeal to them. In the age of content marketing, that means you get to create stuff people will actually like / use.
Digital is measurable. Psychographics and demographics can be aggregated or inferred. This is one of the aspects of digital marketing that draws me the most: the crisp, clean numbers attached to it. Ideas pass or fail. But extracting demographic data from social media profiles is just a little too emic.
What do I mean by “personal branding reasons”? The sort of stuff where you stretch the truth to indicate you’re not
only from, say, Winnipeg. People trade up to more glamourous locales because, hey, they can.
In gathering Winnipeg social media demographics, I noted that it relied on self-reported location data. On Facebook this isn’t as much of an issue, because to make the most of Facebook users kind of need to associate themselves with a place (and Lord knows whether Facebook is providing advertisers with public data or, you know, the other kind. With $1.2 billion in ad revenue this year, they might not rely on self-reported stats. Ooh, imagine if they read your IP?). Anyhoo, Facebook has a very vested interest in providing accurate demographic data.
Twitter, however, is a more creative space (in that you present yourself as you want to be, not necessarily who you are) and, whether for privacy or personal branding reasons, some people don’t list an accurate location. Twitter has just released their ad platform to the public, though, so they’ll be getting serious about user demographics in the name of profit.
Sysomos recently released data (gleaned from over a billion tweets) that shows 31% of Twitter users don’t have a bio, and 18% don’t list a location. While this can’t be accurately mapped to Winnipeg numbers by any stretch of the imagination, it does highlight the need to take them as guidelines, not hard numbers. There’s a lot of (frustratingly) missing information.
What marketers need in Twitter demographic tools
Twitter is rolling out its own metrics platform now, and I’d like to see it include the following capabilites (for any @name), in compliance with ToS-determined privacy, of course:
- A guess at what % are female & male
- Accurate usercount for any location
- The top hashtags for any location over day/month/year
- The most active tweeters for any location or user over day/month/year
- Trending topics over day/month/year for any location
- Generate list or word cloud of follower’s bios
- Generate list or word cloud of follower’s top hash tags
- Generate list or word cloud of follower’s top mentioned words
- Generate list of follower’s top @replied users (who they’re talking to the most)
- Who unfollowed an account
- Tweet efffect on followers (+/-)
A number of these rely on Twitter keeping tweets longer than the 4 weeks they currently do, which wold require a server farm colonizing Mars, so I’m not hopeful for this level of robustness. A few of them are a little creepy 
(though highly useful). If any app developers out there want to make my day/month/year, though, go for it
Posted: December 16th, 2010 | Author: Erica | Filed under: Social Media Marketing, Social Media Platforms, Winnipeg | Tags: demographics, Facebook, social networking, statistics, Twitter, Winnipeg | 24 Comments »
A recent eMarketer survey says 59% of Canadians are using social sites in 2010. What about here in the Peg? The numbers may surprise you / affect your marketing strategy.
Facebook use in Winnipeg
Here’s the demographic breakdown of Winnipeggers on Facebook, gathered from Facebook’s advertising platform. I’ve highlighted where I think the data is suspect [mainly due to teenage creativity]. Click the image to biggie-size.
What percentage of Winnipeg is that? A hefty 70%. 
You can make a pretty good case for your local business having a Facebook page at this point, especially with Facebook Places allowing people to broadcast the fact that they’re hanging out with you. Incentivize their endorsement with a nice coupon—Winnipeggers love that.
Twitter use in Winnipeg
And how ’bout microblogging platform Twitter? In Winnipeg, it’s not so much how many people are on Twitter as how many people aren’t.
This data is gathered from people self-identifying their location in their bios, so is subject to bullshit, but still. 6759 Winnipeggers, or 1.1% of our population, claim to be from the Peg. This is actually higher than the overall Canadian average (determined the same way) of 0.88%. [I've heard wildly different numbers for Canadian use, but this is an algorithm talking].
That said, I’ve met—virtually and IRL—lots of very cool Winnipeggers because of Twitter, and Biz Stone promised on Larry King a few weeks ago that he’s adding 300k users/day (American use is higher than Canadian at 8%). Watch for Twitter use to blow up here in the next 1-2 years, and get started buildin’ those relationships now.
Numerical caveats: Stats gathered from self-identified data are subject to inaccuracy, of course. Some people are valiantly fighting the inevitable by not providing their location data. And sadly, I’ve noticed Winnipeggers sidestepping their location in their Twitter bios as if it makes them less cool. On the contrary, we’re so cool we’re -40!
