Science of Social blog

The Relativity and Economics of Relationship

By MikeW

The Relativity and Economics of Relationship

by Lithium Guru ‎10-06-2010 12:38 AM - edited ‎09-15-2012 12:03 AM

Dr Michael WuMichael Wu, Ph.D. is 927iC9C1FD6224627807Lithium's Principal Scientist of Analytics, digging into the complex dynamics of social interaction and group behavior in online communities and social networks.


Michael was voted a 2010 Influential Leader by CRM Magazine for his work on predictive social analytics and its application to Social CRM.He's a regular blogger on the Lithosphere's Building Community blog and previously wrote in the Analytic Science blog. You can follow him on Twitter at mich8elwu.



Konnichiwa! Greetings from Japan. I’ve just returned from a two-week vacation in Japan, so I want to write about a travel related topic today. I want to share with you a social/relationship phenomenon that I’ve observed from my previous travels. I will be using the concept of tie strength from my last post, so if you haven’t read last week’s post on “Figuring Out the Relationship Puzzle,” I recommend taking a quick skim through before proceeding.


Have you ever gone backpacking? I was an avid backpacker during my grad school days. I would save up for the whole year and then backpack for three to four weeks during off-peak seasons to a country that I’ve never been to before. It was a great experience that I now miss very much since I don’t have the luxury of taking three to four weeks off anymore.


When I backpack, I mainly travel alone; but occasionally, I would travel with a friend. What I find interesting is the different interaction I had with the locals when I travel alone vs. when I travel with a buddy. When I travel alone, I interact with the local residents much more. Duh! There is no one else. If I interact at all, it would have to be with the local people. But I also interact with them at a much deeper level. With some of them, I’ve kept in contact for years. But, when I travel with a friend, my interactions with the locals are much shallower. Even when I spend a lot of time with them, I tend to identify with my travel buddy and vice versa. Consequently, our conversations with the native residents are often superficial at best. Let’s try to analyze this situation and see what we can learn from it in terms of social CRM.


The Relativity of Tie Strength

Escher Relativity2s.jpgLast week we explored the topic of relationship from a sociologist’s perspective, and we discussed the four components (i.e. time, intensity, trust, and reciprocity) that contribute directly to the strength of a relationship. So we can now quantify and compare the strength between relationships. However, humans usually have difficult time thinking in absolute terms, especially with abstract concepts that are intangible, such as relationships. Therefore when we speak of weak ties and strong ties, they are in relative terms; and they are relative to our engagement circle.


Your engagement circle consists of the people that you can engage with easily (i.e. people whom you can focus your attention on with little to no effort). In the real world, your engagement circle would be the people in your immediate surroundings that you can simply open your mouth and talk to. If you are on the phone then, your engagement circle is the one person on the other side of the phone (assuming it is not a conference call with many people). In the online world, however, your engagement circle consists of people that you can access easily without switching platforms. So your engagement circle on Facebook consists of your FB friends, and if you are in a community, then it consists of other members of the same community.


When I was traveling with my buddy, the relationship between my buddy and I would be considered a strong tie compare to the relatively weaker ties between me and the local residents. That is because my friend and the locals are the extent of the people that I can engage with when I am backpacking. However, back home when I am with some of my lifelong friends, immediate families, and close relatives, my travel buddy may only be a weak tie compare to these much stronger relationships that I can engage with when I am at home.


The Attention Economics

A strong tie between two persons means that they will have a stronger affinity for each other. By definition it means there is more trust, more reciprocity, greater emotional intensity (greater sense of closeness) between them, and they will want to spend more time with each other. Therefore stronger relationships will draw more attention than weaker one. I guess that is why the relationship became strong in the first place.


However, attention, like time and money, is a scarce and limited resource for everyone. You can’t be everyone’s best friend. Humans simply don’t have the time and cognitive capacity to do that, at least not all at once in the same place at the same time. By spending more time with one person, you will have less time to spare for another. If you try to jam everyone into the same time slot, then the level of engagement with everyone will simply be very shallow, and you will never get very close to anyone in particular. So within your engagement circle, individuals with different tie strength are constantly competing for your limited attention, creating an attention economy.


When all things are equal, the strongest tie will win over the others. Therefore in the presence of strong relationships, weak ties are harder to develop due to the reduction of attention. However, weaker ties may occasionally win some attention due to contextual and extrinsic factors, such as urgency, easiness, arousal, interest, etc.


So when I was on a trip with my travel buddy, our strong relationship inadvertently drew my attention away from engaging with the local residents. Not only did it draw away time, it also drew away the emotional intensity that is often required to carry on a deeper conversation. But when I was traveling alone, there are no strong ties in my engagement circle that would distract my attention. And since my tie strengths with the native residents are virtually all zero (I don’t know anyone there), I can devote all of my attention towards building relationships with them. As a result, stronger ties can naturally develop.


The principle of attention economy explains why it is often much easier to make new friends when you move to a new town or switch to a new jobs, but harder to make friends in a group of already very close friends. However, it is definitely possible to break into a group of already well-connected friends. After all, I was able to make some friends with the local residents during my lonesome backpacking trips. You just have to try harder to find a time when they are not actively engaged, or else you will have to overcome the attentional demands from their close friends.



So what have we learned today? Two very important concepts: one about relativity, and the other about economics.


  1. The perception of tie strength is relative to our immediate engagement circle.
  2. During an active engagement within our engagement circle, there is an attention economy that governs who we are likely to engage with and at what level we will interact.
  3. Strong relationships demands greater attention, so that in the presence of strong relationships, weak ties are harder to develop.
  4. Weak ties can win some small amount of attention though contextual and extrinsic factors that is beyond the tie strength.


Alright, I hope this post will lay the foundation for some interesting investigation about how relationships work later. Like relativity in physics and economics, the relativity of tie strength and the attention economics can explain many of the observed phenomena in social media today. I will try to cover some of these topics for the posts in the next few weeks. In the mean time, comments and critiques are always welcome. See you next time.



by Skip Shuda(anon) on ‎10-15-2010 09:38 PM

Michael -  Welcome back from Japan.  I hope your travels were rewarding!


Thanks for grounding your narrative in a travel story.  Its a sound illustration of weak and strong ties - and the fundamental concept which has come to our forefront in the last 10-20 years, the attention economy.


I just returned from watching The Social Network - the amazing story about the founding of Facebook.  On our way home from the movie, my wife asked how the company could be worth $25 Billion when its a free platform.   I tried to explain that FB had captured something that was very valuable to people even though it was free.  Your post reminds me that Facebook is very much valued by is market-share in the attention economy.  The 500 Million active Facebook users represent 1 in 14 people in the world... with the strong and weak ties that follow them.


Thank you again for making this conversation happen!

by Lithium Guru on ‎10-17-2010 04:20 PM

Hello Skip,


Thank for the comment.


My travel is great. It's so exciting that it's exhausting at the same time :smileyhappy: But all is good.


Yes, definitely Facebook is valued by the share of attention economy. And part of the reason is because that we have a lot of strong ties on facebook that draws much attention. Most of the strong ties are however formed off-line and were simply maintained on facebook (see Maintaining the Strong Ties). If we have only weak ties on FB, I'm certain that we wouldn't spend as much time on it as we do now.


I watched Social Network too. It is indeed an amazing story. But I have to wonder how much of it is truth and how much of it is fictionalized. Well, not sure if we will ever find out.


Thank you for being the first to leave a comment on this blog post. I've already written a sequel to this post that puts the concept of attention economy to use. If you are interested about whether we can REALLY have more than 150 friends, check out "Where is the New Dunbar Limit?"


Thanks again for the comment and see you around next time.


by Robert Shaw(anon) on ‎11-09-2010 11:03 AM

Great column, interesting thoughts. Congratulations.

One of the reasons the cyber-world changes so fluidly is precisely because the ties are weak and do not represent community or relationships but are random walks in cyberspace, mandelbrot sets, fractals, Rorschach patterns that briefly appear significant but then dissolve.  Seeking out patterns in this world you will find some but many are boojums not snarks.

Repeat visit>>>Click>>>Accident>>>Opt in>>>Repeat Transactions>>> Relationships>>> Communities>>> Ties>>>Mandelbrot>>>fractal>>>Rorschach

The distinction between these is important if we are to accurately portray what is happening in social networks.

Repeat visits to cyber-locations can happen by accident, or due to habit, by clicking, but do not mean that visitor has any kind of "relationship" or belongs to a community.

People have relationships with others they like and also relationships with people they dislike.  Work is an example of a place where there can be negative as well as positive relationships.

Transactions follow patterns, repeat transactions are common but do not imply relationships.  I drink milk with my coffee every day, but have no relationship with my milk.  I do so out of habit.  My habits often have little or no psychological or anthropological meaning.  OK I'm a member of a generational group that drinks beer, I drink several brands of beer, and my choice is mostly random a result of chance events in the past.  The test of randomness is can you predict my choice of brands from your knowledge of me?

Community is a word widely used in literature about cyber-locations but patterns described as community are often not and many patterns are random pseudo-patterns.

An interesting question is how many of these patterns are real, significant and how many randomly generated fractals.

If you look for groups of 150, I guarantee you will find some, they are there and many are random.  I suggest you also look for groups of 15, and see also how many groups of 1500 are out there, and 15000 and so on.  Bet you will find these size groups too if you look.

That said, I do think there will be significant patterns and some will be relevant to businesses that want customers and want repeat business from people who interact with the social media.  Social CRM is going to be big, and the prize will go to those who can tell a snark from a boojum.

Mr Wu, your column is interesting and I do think you may be the person to tell a snark from a boojum.

by Lithium Guru on ‎11-11-2010 10:05 PM

Hello Robert,


Welcome back and thanks for the comment. Sorry for the late reply. We've been really busy at Lithium with the Virtual Summit and everything that goes on.


Although weak ties in our social network are not precisely a community, they are not really a random walk either. There are definitely structures and dynamics that governs how we connect and to whom we connect. Otherwise, any of the standard random graph model (Erdos-Renyi, scale-free model, small world model, social circle model) would be a decent model of our social network.


I must also say that models can predict with 100% confidence. Everything has probability associated with them. Although repeated visit doesn't imply anything definite. The fact that there are repeated visits means something, and business can leverage that. That is all. Whatever the reason they comeback is not important. They can say it is due to relationship or due to habit. Probably both are right, there are people come back b/c they have a relationship with a brand and some probably come back b/c of habbit. The reason is just for intellectual satisfaction and for understanding future scenarios. But the fact is still the same: some people come back repeatedly, so people's visit patterns, purchase patterns, etc. are not completely random. This fact alone is sufficiently beneficial to most business if they can leverage it well.


Likewise, the fact that you are able to find groups of 15 or 1500 doesn't mean anything. If you look at the distribution of groups size, they probably have near infinite support (only bounded by the population of our planet). And there is 1 group that has that size (6.6 billion) namely earthlings. But if you look at the mode of the size distribution, you'll be surprise to find that they probably do have some peaks around 150.


So definitely there will be patterns that are important to business and some that might seem less useful. But the so-call "less useful" are definitely NOT useless. It just means that business don't know how to leverage them yet. But when they do, they will have a competitive advantage. :-)


Finally, thanks for the nice comment. Let's chat more about Social CRM soon.