Social Fame, Social Shame, and the Accounting of a Game: Fixing Fast Company’s Influence Project

[#1]

Michael Wu, Ph.D. is Lithium's Principal Scientist of Analytics, digging into the complex dynamics of social interaction and online communities.

He's a regular blogger on the Lithosphere and previously wrote in the Analytic Science blog. You can follow him on Twitter at mich8elwu.

---

Last week, Fast Company launched the Influence Project, which claims to identify the most influential person online. I’ve written an article on this project last week: The Fast Influencer Myth. Since this topic garnered much discussion, I wanted to take an opportunity to address some of the more interesting topics that have arisen from the subject. I was going to start my mini-series on the business implications of what we’ve learned from the social anthropological roles of communities and social networks (loosely referred to as cyber anthropology) but I will do that later instead.

What’s Wrong with the Influence Project (IMHO)?

I think I was pretty clear on this in my last post, but I wanted to start this article by saying that in my opinion the Influence Project is not measuring influence at all. Influence is not just about affecting the behavior of those you interacted with, as Fast Company defined it. It matters a great deal how you affect their behavior. If you spam someone and frustrate him into clicking your unique link, that is not influence. Moreover, since Fast Company encourages participants to use any means to spread their link, there is no defense against users who may abuse and game the system. Technically, you can lie and trick someone into clicking your link. That is definitely not influence!

What I dislike about this project is that Fast Company is putting an attractive label, “the most influential person,” on these potentially misbehaving spammers, tricksters and liars. By featuring these people on their magazine, Fast Company essentially glorifies this misbehavior. This just doesn’t pass my “gut check” as being something good and worthy of promoting.

How Can I Fix the Influence Project?

Several people have asked me, how Fast Company can improve the methodology of this poorly designed experiment? Being a social analytics scientist who has investigated the topic of social media influence, I thought perhaps I could offer some constructive thoughts on the subject.

There are probably many things that I would have done differently if I was devising this experiment to make it scientifically rigorous, but that would probably take away a lot of the entertainment value too. So, to make the problem more well-defined, I gave myself a more challenging problem: If there was ONE thing I can change about the experiment, what would I do to make the experiment better?

At first I thought to myself, I couldn’t possibly turn this project into a scientific rigorous experiment under this constraint. However, after the World Cup final, when I was a little lubricated from cheap wine, I was daring enough to let my mind wander for a few hours over this problem. I’m pleased to say, that the wine did the trick, and I did come up with couple of changes that Fast Company could do to improve their Influence Project! And here they are.

Social Shame: A Defense against Deception

Although I believe that most people are good at heart and well intentioned, we all know that there will always be a small group of trouble makers who will misbehave, especially under fierce competition. One effective defense against deception is social shame. So how can Fast Company make one simple change to their experiment to put this into effect? Fast Company simply needs to store the tweets/updates from which a participant’s unique link has been clicked, and state that they will be displayed along with the photo of the winner in their November issue. That is it!

This simple alteration – the addition of a public audit trail – can dramatically change the psychology of the game, because not only will people see the winning contestant (the most influential person), they will also see exactly the means of how he won. Knowing this, participants who are truly influential and care about their reputation would likely refrain from using any deceptions, because any misconduct that earns points will be revealed to the public with complete transparency. For every desirable reward there need to be an associated cost. Otherwise the system cannot sustain itself economically. Likewise, when there is potential for social fame, complete transparency provides the mechanism of social shame to keep the misbehaved in check.

Normalized Accounting: A Defense against Spamming

Although social shame is a powerful deterrent against social misconduct, such as deception and manipulation, it does not however, guard against spammers. A simple way to defend against a spammer is to change the influence accounting. Basically, we need to normalize Fast Company’s influence score by the amount of tweets/updates.

Mark Borden (Fast Company editor writing about this project) said in his post “Someone with 100,000 followers who only gets 100 people to join the project is less influential than someone with 150 followers who gets 100 people to join.” My question is which of the following is more influential?

1. Someone who has shared his unique link 100,000 times over the course of this experiment and get 100 people to join.
2. Someone who has only shared his link 150 times and got 100 people to join.

If I use Fast Company’s overly simplistic definition of influence, then by the same logic the latter should be more influential. Why did they only address followers, but not frequency of sharing? Do they encourage spamming? Why? I’ll leave that for you to ponder.

To implement this simple change, Fast Company would have to keep a count of how many times each participants shared their unique link. Since each link is unique to a participant, this shouldn’t be difficult. Then divide their influence score by the shared-count to come up with a spam-corrected version of influence score. This completely changes the dynamics of the game. Now, spamming others with no effect can actually reduce your influence. So if people want to increase their influence, they better share less and share more effectively.

Conclusion

Although Fast Company’s Influence Project (or experiment) is fraught with errors, I was able to cook up a couple of small changes that could help to address the problems with deception and spamming. These changes will not correct Fast Company’s simplistic definition of influence and their flawed experimental design. Consequently, the experiment is still far from perfect, and it does not really measure influence in any rigorous sense, but it will be better than its current state. So what is the moral of today’s post?

1. Clever use of social shame can serve as a defense against deception
2. Measuring influence with normalized social accounting is less prone to spamming
3. A little bit of alcohol can help opening your mind. Don’t abuse it though!

Alright, I will see if there is anything that I can write about the Influence Project next week. Maybe I will have another spark of inspiration. Otherwise, I will resume our journey on cyber anthropology. In the mean time, let me know what you think. Was this the wine talking or was I enlightened? Kudos, comments, questions, discussions, and criticisms are all welcome as always. See you next time.

[#2]

Great write-up, Michael. You've offered a couple simple ways this project could be improved immensely. I'm reminded of the saying that criticism is often valuable, but it's far more valuable to got beyond that and offer specific, actionable improvements.

"Why did they only address followers, but not frequency of sharing? Do they encourage spamming? Why? I’ll leave that for you to ponder." - It's simple: they don't have an Analytics Scientist/Ph.D. to throw at this problem. I've got an idea for a Lithium Social Analytics marketing line - "Got Wu?"

[#3]

Hello MikeTD,

Thank you for the kudo and the nice comment.

Yeah, I think their goal is simply to drive a viral campaign into positive feedback. Calling it  "The Influence Project" is just a way to get their initial participants started so they reach critical mass quickly for self propagation. And finally understanding the science and priciple comes after that.

There are still time until their experiment ends in August 15. I hope they would try to fix (at least improve) it. If they change the rules now, at least they will have couple months of good data. But that might not be what they want. 

Not sure about that marketing line... But thanks for thinking so highly of me. See you next time.

[#4]

Hi Mike,

Awesome. What you have suggested would definitely provide the right picture of who can get his followers to vote for him in a popularity contest that has absolutely no bearing on any particular subject. I completely agree with ‘Social Shame’.

While I still have a few questions about Normalized Accounting. Let us assume, I sent my unique link to one friend of mine. He clicked on the unique link and did what was desired. My friend then resent that unique link to 100 more people who have absolutely no relation to me. And they did not respond. In that case, my influence score will be 1/101. Is that correct?

My second question will be more in terms of how it is being currently calculated at Fastcompany, which we do not know.

Do they allow the same person to submit for more than 1 candidate? In other words we may end up with say 10 participants scoring 10 points each with just 10 unique people having cast their votes.

[#5]

Michael, This is brilliant. I wish I had read this before posting "Is it Worth Getting Your Hands Dirty in Fast Company's Influence Project." You make some incredible points and the Social Shame transparency idea would definitely fix the spamming and tricking aspect. As I just responded to a reader: If I ever get my photo on the cover of Fast Company, I want it to be for something I achieve that resonates with the heart of the Fast Company brand. Not because I knew how to rickroll and spam. Oh and because of the "influence" of this post, I'm now following you on Twitter. Michelle @michelletripp

[#6]

I guess I have to comment on everything you write about the Influence Project It's becoming a trend

I love love love the idea of social shame. Brilliant. Public accountability is something the social media space could use more of. I bet it would drastically reduce the number of gurus and ninjas around if there was more insight into how they got to their guru-dom. But I digress. I think public accountability can and should be present in more things. And kudos to you for suggesting solutions! We are all expressing views and trashing it but there hasn't been much in the way of solutions yet. 

Cheers!

- Maria Ogneva, Attensity

@themaria

[#7]

Michael, this is before your time at Lithium (difficult to imagine Lithium without you), but when we took away the old ratings system and implemented Kudos, one of the drivers was transparency. We wanted it to be really obvious to everyone who had given Kudos to what (and whom) to introduce public accountability for content rating.

I think one of the central aspects of "game mechanics" in weak-tie collaboration systems like ours is that there needs to be a high degree of accountability in the rules.

[#8]

Example of the above - is that your wife giving you kudos, Michael? :-) 20 kudos, in fact... Good to know you have that kind of influence. My wife couldn't find the Lithosphere with a map.

[#9]

Hello PhilS,

Thank you for filling in the history that I missed.

I wonder why transparency wasn't used more often to have a better social accounting (whether it is just for accountability or just to keep the misbehaved in check). Nevertheless, I'm glad we can say that our Lithium Platform is leveraging some of these ideas.

And yes. One of the Kudos on this post is from my lovely wife. She's a published writer in Taiwan, so it is good to have her eyes on my work too. This transparency is definitely a good example of social accountability. Not only you can see that she kudoed me, you can also see which post she kudoed. Unfortunately, not all of my blog post have gain her kudoes.... I got to work harder. 

[#10]

Hello Maria,

Thanks for coming back again with such a nice comment.

Social shame is actually not a new idea. People have experiment with them. Finding the right time to implement it is crucial. If you do it too early, it may scare people away and get low adoption rate. If you do it too late, people who misbehaved would have already gotten all the fame, glory, credit, etc. out of it. In pratice it is best to implement a transparent (social shame) strategy soon after the system goes viral (able to self propagate).

Having a better social accounting is always a good thing. As the industry matures, I think more of them will be put in place. I've once said in a sCRM panel that

     "The only way that any organization can scale with social is via social."

This social shame strategy is one way of using the social's pair of eyes to watch out for each other's misbehavior for the common good. It's like a digital version of your good old local crime watch program. Right now companies probably think that they can handle the volume, but when they find themselve cannot scale up with the social, I would expect to see more of these type of large scale social solutions.

Anyway, thanks for coming back. Cheers.

[#11]

Wow -- your wife can write in addition to sing? Kudos!

[#12]

Hello Michelle,

Welcome to Lithosphere. Thank you for stopping by and commenting on my blog.

I just read your post. I believe that a lot of people probably feel the same way you feel. If they have to use whatever means they can to get their friends and followers (which took a long time to build up) just to click a link, I believe that most people would have the conscience to sense something is wrong there.

However, as you mentioned, many people simply uses scheduling and automation tools to auto tweet and mass DM their followers and friends. These tools often remove us from actually repeating the act of spamming (and/or deception) over and over again ourselves. This is how people can become blind to the righteous calls deep within their heart.

Anyway, I admire you for realizing there is something icky about the project and have the courage to stop and write a post about it. Kudos to you for sharing your experience.

And thanks for the follow. Just follow you back too.

[#13]

I personally do not think it's Fast Company's fault or issue if people spam/trick people to click their link.  If that does happen people can close the window before the page actually loads and warn others by sending out a post/tweet/etc.. 

I did join the project but I expalined to all the people that either read my blog post (http://wfapm.com/aU82Ql), tweets (http://twitter.com/jayphilips) and emails what the link was that they were clicking.  I even explained in the emails I sent how they could help promote my link (http://fcinf.com/v/cnw8) or theirs.

[#14]

I am disappointed. I really expected Fast Company to flip this on its head and turn it into a study of social competition and the community reactions. I'd love to see numbers related to helping behavior over self-promotion. Alas, the study of the results appear to be shallow so far. Thank you for this thoughtful piece.

[#15]

Michael,   good perspective.    I know the intent of FC was to use ego to build some form of a viral graph to

determine influence and drive traffic.   The implementation was whack for sure.   As soon as this hit I saw the share flaw you pointed out.      In the past I created an affiliate tracking platform that was based on who shared to whom how many times their degree of separation and the frequency of invites etc.   The implementation was really good for calculating payouts to the down line. 

I believe if FC would have implemented this would have provided better ranking these inverted pyramids.   BUT this is not influence.    Now with Klout do you agree they have a good way of determine influence ?     As you know, many out there are so called influencers but what do they influence …. re baking of content mashups, regurgitation and others RT as fast as possible on who can be first to build up their own social currency.     

Again is the behavior we are seeing now on what we call influence real and how far off is FC with what they did?   Maybe there implementation is exactly what’s being classed as the -Hype Influence-.    “Famous for being Famous” I’m sure a few people come into mind.    

@chasemcmichael
@infinigraph    

[#16]

Hi Michael,

Good points you make Dr.Wu, implemented it should be. Just watched Star Wars again with my 7yr old and so mind is filled with Yoda speak :-).

No question about transparency - definitely a good idea and I would think based on the learnings from social media these past couple of years that it would be a given that this be a standard criteria for everyone doing anything in the social space.

On the normalized accounting, I agree with your suggestion - but things are very tricky here. One of course is the issue of spamming - the person with only 150 followers might have sent the link 50 times to get his 100 clicks whereas the one with 100,000 followers might have sent it only once.  I also think because we are mere mortals, emotions like jealousy, ego etc. will bias the score ("A" might not click on "B"'s forwarded link but then register and pass out his own link etc.). There is also the factor of 2nd tier 'influence' (the sum of the followers of all your direct followers) and how it interplays with the influence score ...and the like. In addition, there is a very strong liklihood of the "American Idol" syndrome creeping in with this study too (people calling their friends and families to "click" - not exactly spamming -- but this is not influence either).

But all in all definitely an interesting problem to water down with some wine :-). Btw, not since my college days studying Cook & Campbell have I heard anyone critiquing the experimental design -- you are right again about all types of biases creeping into this study.  

Regards,

Ned

[#17]

Hello Whatsay,

Glad to see you back again. Sorry for the delayed posting of this reply, because it took some digging to get the answers.

I will answer the 2nd part of your question first. Fast Company provides the following guideline and explains how they allegedly measure influence. They also have an FAQ with added explanations. I’ve copied the relevant info on scoring below.

GUIDELINES:

The Influence Project is designed to visualize your online influence in action. Everyone who participates will have his or her photo appear in the November issue of Fast Company magazine as part of an amazing photo spread.

1. Sign up and get your own personalized, unique URL link to the site.
2. Your goal is to use any and all of your online social networks to spread your unique link and influence people to click on it.
3. Each person who clicks on your link for the first time helps increase your influence and the size of your picture.
4. You'll also receive partial credit for clicks generated by new members who register through your link. The credit will progressively diminish (1/2, 1/4, 1/8, etc.) the further out these clicks spread, but you can earn fractional points up to six degrees away from your original link.
5. The "most influential person" will be the participant who demonstrates the greatest overall influence by the deadline.
6. The competitive portion of the project begins July 6, 2010 and ends August 15, 2010 at midnight EST.

HOW WE MEASURE INFLUENCE:

The scale of your influence, and therefore the size of your photo, is based on two measures.

1. The number of people who directly click on your unique URL link. This is the primary measure of your influence, pure and simple.
2. You will receive partial "credit" for subsequent clicks generated by those who sign-up as a result of your URL. In other words, anyone who comes to the site through your link and sign-ups for their own account will be spreading your influence while they spread theirs. That way, you get some benefit from influencing people who are influential themselves. We will give a diminishing, fractional credit (1/2, 1/4, 1/8 etc ) for clicks generated up to six degrees away from your original link.

FAQ:

1. Does someone need to register in order for me to get credit for influencing them?
    No. You get credit simply when they click on your personal URL the first time.

2. Do I get more credit if someone who clicks on my URL registers for their own?
    You will receive partial credit for subsequent clicks on their unique link, as well as clicks for those who sign up through them. See HOW WE MEASURE INFLUENCE

3. What if the same person clicks on my link over and over again?
    Only unique clicks will count toward your influence level (and picture size).

4. Do I get credit for clicking on my own link?
     No, sorry, you get nothing for influencing yourself.

Now back to your hypothetical scenario. Under the normalized accounting, what you said would be true. If they are smart about it, they can do some sender IP hashing in their shrunk URL, so if the link is not sent by you, then they are not effective. This will prevent your friend from sabotaging you by mass distribution of your unique links random people who most likely will not click the link.

Now, under the normalized accounting, if your friend registered and send HIS unique link to 100 friends, who all ignored him, then you will still get (1 click/1 tweet)=1 influence point; and your friend will get (0 click/100 tweet)=0 influence point. But if one of his friends respond to his 100 tweets, then you will get 1+1/2=1.5 influence point; and your friend will get (1/100)=0.01 influence point.

[#18]

Hello Jay,

Thank you for the comment.

First of all, kudo to you for being honest: explain everything with blogs and tweets etc. Good for you! I believe that majority of the people are good at heart, and they want to play the game fairly. But one thing we learned about the social system is that there will always be a minority group that will abuse the system and use inappropriate methods to get ahead.

Suppose in a hypothetical country, there is no police. I can argue that it is not their government's fault for not having any police, because it is the individual's fault for breaking any laws. If people don't break laws, then we wouldn't need any police. But the problem is that SOME people do break law, albiet small number of them. And those who broke the law will affect the rest of the community if they are not punished properly. So I think the government has a responsibility for institutionalizing police to keep the society in order.

By the same argument, although I would agree that it is not completely Fast Company's fault that people cheat or spam, I think they share some responsibility for designing a system that allow for such kind of misbehavior to surface. Moreover, by allowing misbehavior and glorifying them at the end, they essentially encourage these misbehaviors. That is why I dislike this project.

But, I appreciate your honesty, and be outspoken about your point of view. So thank you, and I hope to see you around next time.

[#19]

Hello Shannon,

I totally get how you feel. I was so disappointed that I wrote my first post on this topic last week: The Fast Influencer Myth. I totally agree that they should turn this experiment into a study of socical competition or viral chain reaction. Besides that, I can only see some entertainment value coming out of this project now.

What you said about helping other is very good. If Fast Company or Mekanism is smart, they might be able to turn this into a game of Pay It Forward and make some lasting good impact. In the mean time we can only hope for the best.

Thank you for stopping by and commenting. Hope to see you again on Lithosphere.

[#20]

Hello Chase,

Glad to hear that you saw the same flaw in their experimental design as me. That tells me that at least I am not crazy, or if I am crazy, I’m not alone. I’m quite interested in the affiliate tracking platform that you’ve created. Do you have some white paper or academic paper written on it?

The methodology that Klout currently employed is still far from perfect. With respect to the “The 6 Factors of Social Media Influence,” they did not address temporal relevance, nor did they address target confidence at all. Despite that, they are certainly one of the leaders in this relatively immature industry.

I actually went and talk to Joe Fernandez (Klout’s CEO) last week. They are in need of scientists and mathematicians who can help them improve their algorithm. Too bad there isn’t a clone of me, because I think they have a very challenging problem there due to the scale and size of their problem. And I would love to take a stab at it myself if I have the opportunity. The good news is that they have just hired a research scientist. She is very good. So I think they will significantly improve the accuracy of their influence score in the near future.

I totally agree that the Fast Company’s measure of influence is really not influence at all. Even popularity is stretching it a bit. But I like how you coin the term “Hype-Influence.”

Anyway, thanks for commenting, and I hope to talk to you again.

[#21]

Hello Ned,

Welcome back Master Yoda, and thanks for the comment.

As in my reply to Maria, I think transparency (social shame, social accountability) will be more prevalent as the industry matures. It took thousands of years for humans to learn to behave and live harmoniously in the world. I certainly hope that we learn faster on the online world.

The normalized accounting that I proposed was under the constraint that I gave myself of changing only ONE thing about the experiment. There are a lot of fancy things that we can do with Renormalization Theory that can make the system essentially hack proof. Well, nothing is 100% immune to abuse and gaming. But we can take care of 99% of the cases with some fairly simple implementation of normalized accounting. The basic idea is that we need to take into account of 99% of the things that a participant can potentially abuse by gaming the system to increase his score, and implement mechanism for automatic punishment when those abusive actions are observed.

I used the frequency of sharing as an example because that is so obvious and easy to understand. And it is ONE simple thing that Fast Company can change easily without altering other parts of the Influence Project.

Finally, thank you for the compliment and support. And see you again next time.

[#22]

Michael,
I am with you on changing "ONE" thing to make this experiment better is a pragmatic approach and agree we don't have to apply the Renormalization constraints to it :-) [though studying the recursive calculation for link clicks might be an interesting experiment].

Truth be told, I am inherently against instigating people to "demonstrate" their influence -- this very fact (imho) introduces bias. Influence should be measured organically in the sense that we should measure what is there not what somone creates just for the purpose of measuring influence.

Also, was reading your response to Whatsay and am having a tough time accepting their explanations on measure of influence. In the section 'How we measure influence', it is mentioned:

1. The number of people who directly click on your unique URL link. This is the primary measure of your influence, pure and simple.

2. You will receive partial "credit" for subsequent clicks generated by those who sign-up as a result of your URL. In other words, anyone who comes to the site through your link and sign-ups for their own account will be spreading your influence while they spread theirs. That way, you get some benefit from influencing people who are influential themselves. We will give a diminishing, fractional credit (1/2, 1/4, 1/8 etc ) for clicks generated up to six degrees away from your original link

The problem I have here is that person "A" might only have 2,000 followers and person "B" 20,000. However, the 2,000 followers in the first case might be the creme de la creme of the society or within a domain; and the 20,000 followers in the second case might be just high-school chums, relatives, and such (and this is not a totally hypothetical scenario either). Anyway, based on the explanation above it is a given that "B" will most likely get a higher influencer score than "A" - even though personally I would rate "A" as having more influence.

(Of course, one can argue back that it would depend on what defintion of influence is being used).

Anyway, we both definitely agree that the study can be improved.

Regards,
Ned

[#23]

 Hello Ned,

I totally agree with all your point here.

Just to show how absurd this measurement scheme is: Fast Company's measurement scheme is like measure someone’s wealth by asking them to donate money to you, and then declare the one who donated most as the wealthiest person? For the average person, this may look kind of right, even though it feels wrong some where. But they cannot really pick out why. This measurement scheme is one-side biased. It seem kind of right because if you are not wealthy, you cannot win, because you cannot donate much. However, how much a person donates has nothing to do with how wealthy the person is. The wealthiest person can donate a lot or donate nothing at all. The winner is simply the one who donates the most money, and has nothing to do with how wealthy the person is.

Similarly for Fast Company, the most influential person may get many clicks or few clicks (for whatever reason). The winner is simply the person who get the most click (for whatever reason), which has nothing to do with influence.

Fast Company’s way of computing influence is clearly based on a pyramid scheme, which is a fraudulent business model in many places. Even if someone is much more influential, he would lose out simply by getting into the game later. The whole scheme just doesn't make sense if you look at it more carefully.