Two Guys Talk about Bras
Written by: Bill Sherman on Sunday, 17 January 2010, 6:30 PM
On January 6, 2010, unusual status updates started appearing on Facebook. Many women began posting colors and patterns. The Facebook Bra Color meme, which promoted breast cancer awareness, became a one-week wonder across the Internet and in main-stream media. The meme was spread through an e-mail between women, letting the guys puzzle it out.
One of the nation’s most passionate puzzlers, Mike Selinker, solved the mystery of the status updates. Mike is an old friend of mine, and he’s the president of Lone Shark Games, a Seattle design studio that specializes in social network games such as puzzle events and alternate reality games.
Mike and I exchanged messages as we think through the bra-color meme: relating it to social networking, marketing, and social learning. We’re reprinting it below.
Mike Selinker: Today was… interesting. I woke up to check my Facebook feed and saw that a number of people on my flist were posting colors as their status. Instead of asking why, I tried to figure out what the people who did this had in common. They came from very different sets of friends, who had no connection to each other beyond me. Then I realized that they were all women. And I thought, “In which arena would my female friends pick wildly different colors?” It had to be something they didn’t know about each other, something intimate. And then it hit me. They were all posting the colors of their undergarments, specifically their bras.
So I posted a message of amusement, and the secret was out. I’m sure I became a vector for this meme, despite not being contacted in any way by the propagators, and not being able to participate directly. (I am not currently wearing a bra.) The virus spread through email and inboxes, and manifested in public. That’s something I don’t think I’ve seen before.
Only later did someone point out that this was started as an awareness mechanism for breast cancer. Which I find fascinating, because I’m sure that if the Susan Komen Center spent donation money on getting people to post colors on their Facebook statuses, there would be a board of directors meeting hastily convened. And yet I certainly thought more about breast cancer today than any other.
Bill Sherman: Like you, I found myself smiling and laughing during the day as I saw the meme propagate across Facebook.
I’m going to focus here on transmission vectors, misidentifications, and mutations.
By evening, I started to see a trend of women who had received the e-mail in their personal account and didn’t post until after dinner. So, there was a second wave of color postings that rippled east-to-west in the evening.
The bra color meme, in some way, is a mutant strain of the pink campaign. For many years, breast cancer awareness (as an advocacy issue) has been associated with pink. We have pink ribbons, turn buildings pink, and redecorate grocery stores each October. We’re conditioned to think about [color] –> breast cancer awareness. In many ways, we’re primed psychologically to connect the concept of color with breast cancer awareness.
Throughout the day, the virus followed two separate transmission patterns. E-mail was the initial vector. It included an explanation of what to do, why it was important, and a note of encouragement. It also turned the activity into a gender-divided game.
However, the Facebook status updates quickly produced an array of responses (from men and women) who hadn’t received the e-mail. Many women and men had their first experience with the meme outside of the e-mail.
In these cases, the meme was misidentified or and in some cases even mutated.
For example, early yesterday morning a male friend posted the following cryptic status update “is in on the joke. It’s not hard to deduce, guys . . . ” He had deduced the game component of the meme, and he assumed that it was merely a semi-risqué meme that had propagated.
Also, it became possible to track mutations created by people (men and women) who had not received the e-mail but wanted to participate. Here are two of my favorites status updates:
- a list of fifteen colors in one status update (by a man)
- “infrared”
While these posts were delightful (and caused me to laugh), they were mutant strains of the initial meme which had been divorced of the breast cancer awareness element. These actions reinforced the self-replicating social game, but it created “noise” that competed with the issue-advocacy element.
So, we have two transmission vectors for the same meme. And it’s a really exciting way to use social media technology. The primary vector (e-mail) reinforced the social advocacy, and the secondary vector (seeing the meme first on Facebook) often produced mutations and misapprehensions.
Splitting the meme this way created some interesting ripples. It’s like throwing two rocks into a pool of water. We saw a single meme’s ripples create interference patterns with itself. That was incredibly cool (and compelling to watch).
Here’s my question for you, Mike. We both found this meme fascinating. Yet, when it comes to issue advocacy (or even marketing), awareness is good but moving people to action is far better. Was this a five-minute diversion or did it make a difference?
Mike Selinker: A reasonable question, and one I get asked about games all the time. I’m a proponent of buried messages, important points that bubble up when you think back on the experience you had. For example, for our alternate reality game Citizens of Virtue, left-wing evangelical preacher Rob Bell asked us to make people think about hypocrisy in church messaging, a very complex issue. So we invented a fictional Focus on the Family-style organization called Citizens of Virtue, which sounded very plausible to many. Their “Virtual Virtues” campaign—”virtual” in the sense of “electronic,” but also in the sense of “not real”—was a weekly dispensation of tasks that purportedly would inspire one of the Seven Cardinal Virtues, but instead propagated the corresponding Deadly Sin. So for example, “Humility” was encouraged by purchasing an expensive “LORD shackle bracelet” a la Cartier’s LOVE bracelet, to show how humble you were. This of course was the height of Pride. These were complex messages to pass on, but as people were having fun exploring the game and undermining the CoV from the inside, they walked away with some important thoughts about important issues.
So too, I think, with the bra meme. There were three things going on that I think directly benefited the cause of breast cancer awareness:
1) The repetition of the rationale for the color posts. People would say “Why are you doing that?” and others would respond “It’s my bra color. It’s a breast cancer awareness thing.” That alone could lead to quite a few reminders to schedule mammograms.
2) The minor backlash that concerned the female-only delivery method. There were several posts that said, “Hey, men get breast cancer too!” Now, we do so only at a rate of 1% of the female population’s incidence, and it’s a very strange sense of male entitlement to demand equal treatment for diseases. But there are probably men who thought, “I better get myself checked out.” They may never have considered themselves at risk before.
3) After I posted that I was tempted all day to write “Pics or it didn’t happen,” the call to action you’re looking for came from a friend who pledged $5 to breast cancer research for every “proof” picture he received in his inbox. Quite a few women took him up on that, apparently willing to sacrifice that amount of dignity for a good cause. So voyeurism gets subjugated to good works. Creepy, but awesome nonetheless.
So I’d say yes, it was a five-minute diversion, and yes, it made a difference. Is it more effective for cancer awareness than the TheTruth.com ads? Probably not. But which would you rather watch?
Bill Sherman: Although this meme launched on Facebook, the social media ripples have extended to Twitter and blogs. This quickly became a meme that people wanted to discuss. And the discussions have been fascinating:
1. A woman who had a double mastectomy blogged her thoughts to the meme.
2. Mary Carmichael at Newsweek blogged her response to what she called the “pointless underwear protest.”
3. Users created and joined new FB groups: “Your Bra Can’t Fight Cancer, But Your Wallet Can.”
4. Individuals became fans of Susan G. Komen for the Cure on Facebook.
We saw two distinct phases in this game. First, many women chose to play the game, and that created the many color “status updates.” I honestly didn’t expect so many people would want to discuss this meme.
I think that people inherently sensed many of the flaws of the original meme, and social media facilitated the conversation about those flaws.
A friend of mine wrote a status update where she: “is fairly certain that everyone knows that cancer exists. How exactly is this game helping? Instead of posting @%#& telling people to post more @%#&, why don’t you go donate a dollar?”
In this way, we saw a social media game begin to evolve as people identified the flaws and want to improve them.
The color game has spread to Twitter, but the text mutated into Tweets like the following one:
“White! Which color of bra do you wear? Add a Twibbon now to support breast cancer awareness! ”
This new mutation contains the original game, a viral question, and a call to action. Is this new strain better than the previous one? Well, it depends on the goals of the meme. I’m sure the people who provide twibbon icons are happy.
People care about this game (because of the issue advocacy), but they also recognize the original game’s flaws. They have taken time to discuss and improve the game in a form of collaborative game design and development.
We’re looking at the intersection of social games and social learning. Fascinating stuff.
Mike Selinker: I quibble with the word “flaws.” There’s nothing necessarily valuable about hemming something toward the ordinary. When a friend turned the meme toward underwear color, she added a sentence “For ovarian cancer awareness” and a link explaining symptoms of that horrible disease. While this was laudable, turning it into a link (as opposed to a status update) automatically blunted the spread of the mutation. We’ve all seen friends post charity links before, so this was just a variation on that. If it’s ordinary, it’s probably not going to spread virally. That’s not an improvement of the game.
Without any question, this was a chain letter, spread through a whisper campaign. Those are negative terms; nobody I know is going to admit to sending a chain letter, or engaging in a whisper campaign. This got them to do it. It’s a good game as is.
It’s also important to view it as a limited-time game. As people continue to post bra colors, people are already in the mindset of “Honey, that was so yesterday.” That’s okay. If you were part of the spread, you talk about it later. If you weren’t, you’re made to feel like an outsider. Even a nationwide in-group enforces clique rules.
Now comes the question of whether this is a marketing lesson. I am sure that I soon will get a call from one of my advertising clients saying, “Mike, can you make the color thing happen for us, but this time with varieties of yogurt?” Even if I say no, someone will say yes.
Bill Sherman: As you said, “Can you make the color thing happen for us, but this time with varieties of yogurt?”
If we’re going to think about marketing, then we need to pull in some ideas from the social psychologists. First, the behaviors of people around us influence our behaviors. We saw this behavior manifest with the bra-color posts, but it resonates with Milgram’s crowd-behavior experiments.
Stanley Milgram, who performed many different (and notable experiments), ran one experiment on the streets of NYC in 1968. He positioned research assistants at strategic points on the street and had them look at the window of a nearby building for one minute.
If only one person stared at the window, then 4% of passersby would stop; however if 15 research assistants stared at the window then 40% of the passersby also stopped.
It’s the old “made you look” technique. We can create buzz and direct eyeballs and behavior (whether fully or partially) by planting conspirators within a social network. Then passersby mirror (wholly or incompletely) the behavior.
So, while stunts can attract attention, will they be significant enough to get people to notice the product (our yogurt) or cause them to act (refer to a friend, make a purchase, etc.)? That’s what matters.
For that, Robert Cialdini makes a good point about descriptive norming and behaviors. He’s done a lot of research around environmental messaging (such as recycling or towel reuse at hotels) and finds that people respond when people believe they are acting in concert with people like themselves. Descriptive norming creates a positive social pressure to spread behaviors virally. These ties of “people like me” are powerful, so that even knowing that people who have shared the same physical hotel room typically reuse their towel influences your behavior.
In Connected, Christakis and Fowler argue that we’re influenced by the behaviors of our friends’ friends’ friends (3rd degree connections) for behaviors such as smoking cessation, depression, suicidality, etc.
If the average Facebook user has 130 friends (which approximately correlates with Dunbar’s number of 150), then a single person launching this meme would have had access to 130^3 (2.179 million) users. However, that number would have been reduced by many duplicated connections. So, just like radio broadcasts, retransmitters boost the signal.
I think our strategy for replication would depend on a number of different factors:
1. Yoplait (known brand) vs. Mike’s Yummy Organic Yogurt (unknown brand)
2. Definition of success . . . do you want people to become Facebook fans, want them to buy yogurt or become aware you exist?
3. Is this a flash-in-the-pan strategy or does the client want to create a relationship with these people?
I think we could craft many “made you look strategies.” There are also many “build relationship strategies;” however, it’s often difficult to achieve both goals simultaneously.
Mike Selinker: Amusingly, it took less than a week for a client at a marketing firm to say to me, “Did you see people posting their bra colors on Facebook? Well, like that, but….” (But not on a yogurt account, sadly.)
Facebook seems a direct challenge to Dunbar’s number. Dunbar believed that settlements broke at 150 people, because humans couldn’t maintain social connections with more than that number of people. But Dunbar was talking about groups with intense external pressure to stay together. An army unit in wartime has such pressure, because if it fails to move effectively, its members will be killed. So too with an office or a school, where social disorganization is problematic to authority. I think what we’re seeing is that without that pressure, without that need for authority, we are capable of monitoring and casually interacting with much larger numbers. I’ve had a running joke of every time I get another 100 friends, I make a stupid status report announcing it (e.g., “Mike has 20 square friends” for when I hit 400). I didn’t expect I would have to keep doing it past 1300.
And here, I think the army travels at the speed of its FASTEST member. I don’t need all my friends to be aware of a meme to respond to it, I only need one. I wasn’t the first of my friends to hear of it, but I was certainly the first male I knew to figure it out. That made a whole lot of people spread it outward from me. The meme was designed to treat me as a scalar, but I became a vector.
That’s where marketing can learn. We ask all the time, “who is this message aimed at?” What we also need to ask is, “what is the effect of this message on people it is NOT aimed at?” There was a horrible example in the game industry of my former employer Hasbro marketing the new version of the boardgame Risk to boys with a site featuring a game where you try to sleep with as many women as possible, and another where you fling poo.
Though I doubt it, let’s assume this hit square with the boys it was aimed at. The resultant backlash among women—specifically mothers, who buy almost every boardgame—likely destroyed any positive effect Hasbro could have gained. As it spread among Facebook friends, the female gamers I knew couldn’t see what any of this had to do with Risk itself. All they knew is that they weren’t going to buy it. (Not that I dislike things aimed at males, of course. There’s an amazing promotion for the Dodge Ram. That’s all male, all the time. But it’s also awesome, which excuses a lot of sins.)
When you set something out where everyone can see, you have to consider the impact on everyone who can see it. Facebook has the potential to reach everyone, but you have to reach them with something they want. Otherwise, you may wish you hadn’t reached them.
Bill Sherman: According to the current Facebook press page, the average user has 130 friends. According to a research study published in 2008, that number was 110 friends, so we have seen an upward trend.
Also, check out the 2009 analysis of maintained relationships on Facebook’s site. It’s well worth a read.
I’d argue that your 1300 friends is probably several standard deviations from the mean. Quite simply, you’re a network hub.
When we look at a social graph, your Facebook friends’ network has a high degree of centrality. You’re closer to the center of a network than to an edge.
When you cracked the code of the colors in the status lines, you became the equivalent of a high-powered radio station broadcasting louder and farther than people with smaller networks. If we accept the 3 degrees of influence benchmark, then 1300^3 leads to a network reach of 2.3 billion (but that number will certainly be lower in practice due to mutually-shared connections).
If you hadn’t touched the meme, then it would have evolved and spread in an entirely different fashion.
Technology has changed how messages are transmitted through networks. Instead of closed, geographically-bound relationships within villages, we have been able to maintain strong, geographically-distributed relationships. Yet, I’d say that Facebook allows us to maintain many more weak-tie connections than we could previously (even when compared with Granovetter’s day).
Interestingly, the Risk example and the Ram microsite are both examples of a traditional marketing message where each visitor interacts with a company-approved site. Memes work differently, because they spread person-to-person rather than through an approved central hub. That opens the opportunity for mutation.
So, here is what I see as our final takeaways:
1. You point out that you have to consider the potential impact on everyone who can see it, and that message may produce undesired reactions.
2. I emphasize that the message itself transforms when people within the network interacts with the meme and rebroadcast it. Therefore, once you set a social meme out to the world without requiring it to “phone home,” it takes on a life of its own.
All in all, these two very powerful trends reshape how marketing messages and learning ripple through a large-scale social network.
Mike Selinker: That all makes sense, Bill. This has been without doubt the most interesting discussion about bras I will care to admit to having.
Something to say?
You must be logged in to post a comment.