In a study published in the Journal of Marketing Research, titled What Makes Online Content Viral?, Jonah Berger and Katherine L. Milkman analysed about 7000 New York Times articles to examine which articles made it to “the most e-mailed” list. The conclusion of the research was the emotional intensity evoked by the articles was a major factor in determining their virality. Thus a change in the article characteristic by one standard deviation causes the following changes in the probability of making it to the most e-mailed list:
The articles have higher virality (more shared) if they stir up emotions. Some emotions have higher impact than others, but the one outstanding aspect is that anger and awe have a higher influence than the practical value or the interest of the article. The research shows that the low-arousal emotions (like sadness) dampen our enthusiasm for sharing. This is also interesting. My reading into this is that sadness has lower social value and it is not looked upon favourably by our peers. I dare say this is in line with behaviour linked to depression when people tend to suffer in silence.
Do we have the same appetite for sharing in other social situations, such as Facebook or Twitter?
Ben Elowitz wrote an interesting post on his blog, where he shared (!) a few thoughts about emotions that drive sharing. He analysed the Most Shared Articles on Facebook in 2011 and discovered that the emotional drivers have a different pattern of influence on. Number one factor is emotionality at 23%, followed by negativity (18%), interest (15%), surprise (13%) and positivity (13%). The practical value is way down to 3%!
Sure, the emotional patterns differ from one sharing environment to another, but clearly sharing is an emotional act. As Berger and Milkman noted in What Makes Online Content Viral?, “Virality is partially driven by physiological arousal”. I guess spiritual/intellectual emotions would count as low-arousal – there is not one word about that in that study.
Harvard Business Review published research on the same topic. In The New Science of Viral Ads, Tales Teixeira describes five ways of increasing the chances of people sharing video ads:
- Utilise brand pulsing: don’t put your logo straight in the middle, but show it only from time to time. This increases the chance of sharing by 20%
- Open with joy: right at the opening do something that causes positive, joyful reaction. This prevents boredom.
- Create a roller coaster: keep coming up with surprises and joyful experiences: this prevents the drop in interest
- Surprise but not shock: if the shock is negative, embarrassing or venturing into taboo territories, people may feel inhibited about sharing
- Use all of the above: joyful opening, show the brand here and there, keep people hooked, don’t shock.
The methods of measuring the emotion, the type of sharing medium and the type of content are different, but the role of the emotion in the decision to share something is undeniable
This may explain why sharing in formal circumstances doesn’t work. When you get the emotion out of the content, sharing is not that attractive, lacking the virality we see in informal social situations. By contrary, as the Berger and Milkman’s research shows, the low-arousal emotions have an inhibiting factor.
We share more when we feel that emotional impulse in the content. However, emotion is involved not only before but also in the act of sharing. What do people feel when they share? Unfortunately the research is mostly focused on the correlation between the emotion triggered by the content and the decision to share ignoring the emotion of sharing.
Emotions such as feeling appreciated, feeling good about being funny, intellectual, practical or artistic, getting rewarding feedback, feeling social as opposed to isolated, feeling contributing to an important cause, being the best in some ranking system, all of these contribute to making people want to share the content. These emotions have to do with our character, and our character and experience have to do with our choice of environment for communication.
Our sharing behaviour must be influenced also by the kind of environment we use and the type of human network we are part of. For example, people are never going to share funny videos on LinkedIn. Why because there is an expectation you need to look “professional” there. You can’t take the risk of broadcasting on LinkedIn a prank video of the Pope dancing with Madonna at Mardi Gras festival even if you think is really funny. You will be frowned upon by all the HR departments.
This is why I think we should be looking on not only what makes the content viral, which is a marketing problem, but also about what makes people share, which is a social communication problem. Formal network environments miss the opportunity to engage and get people become more creative, curious and motivated because they lack of emotion, inhibiting the social communication.
As an example, SharePoint is a very good product, but it is too formal, too boring to capture the imagination of people making the information sharing less effective. The SharePoint 2010 My Profile feature is a typical corporate “search to find information” strategy. Even the voice describing the features is emotionless. The product design is all about information, fields, tags, skills and projects. There is very little about socio-emotional features, a comprehensive anti how to make information viral strategy. I dare challenge you to listen to that video describing My Profile feature until the end of it. If, as research shows, people are far more likely to share in a state of arousal, this sleep inducing video, has close to zero chance of being share. Oh, except for me. I just did it! I guess we can put this in the category “interest” or “practical value” basket. Anyway, it needs to add a socio-emotional dimension to it to really become an effective information sharing tool; let’s say a combination of Facebook/Google Plus and SharePoint would do much better.
Social games are derided by some as being too superficial, something for de-stress therapy in the weekend. But modern games are really getting better at engaging our internal emotions which defines us who we are. Gamifying the sharing experience works because it adds emotional value to the act of sharing. If the sharing environment adds features that tap into these emotions such as giving rewards, showing your approval/disapproval for shared content, follow and be followed, all these increase the chances content will be shared. This could mean that specialised social networks will spread and gradually replace the formal sharing environments and put more colour into our virtual office lives.
While trying to explore the principles of creation of knowledge, I wanted to run an overview of the epistemology as a way of understanding knowledge from a philosophical perspective. As result of this mini-exercise am writing a brief overview with a few comments, followed by my brief personal critique of epistemology.
What Is Epistemology?
Philosophy as a thinking system likes to explore the universe through “rational investigation of the truths and principles of being, knowledge or conduct” (www.dictionary.com). Where does epistemology fit?
According to Peter D. Klein “epistemology is concerned with the nature, sources and limits of knowledge”. This is where the trouble starts. What is knowledge? According to some knowledge should include objective forms, others think that in the context of epistemology, knowledge is only about beliefs that something is true as opposed to knowledge about how to do things.
Epistemology ventures into areas where precise measurement is impossible. This is why there are many definitions of epistemology and heated debates have been going on for centuries.
Propositional Knowledge Epistemology
The focus of epistemology on knowledge analysed on the basis of beliefs and truth takes this philosophy out of the natural philosophy branch, where I would have preferred it to be. In my view this limits the influence of the advances of science on the development of epistemology simply because subjective knowledge is impossible to measure. The claim of the traditional epistemology is that the quality of the reasons for our beliefs determines the conversion of beliefs into knowledge. This approach is called the normative epistemology, and is supported by theories of justification. Another tradition, the naturalized epistemology, claims that the conditions in which the beliefs are acquired determine the truthfulness of the beliefs.
This tradition has two views about the structure of reasons: foundationalism and coherentism. The foundationalism reminds me of the Anglo-Saxon common law. Rulings can be made based on precedent rulings which have been gathered throughout the history of application of the law in territories under the crown for many, many centuries. Some of the rulings are unique and they can form the basis of a new ruling for a case that could occur in the future. When they occur and the reference to the precedent is accepted as being similar, the court can rule without having to repeat the previous process. Thus, according to this view, beliefs can be based on other beliefs which have been proven true in the past, therefore they don’t have to be justified and thus together they form the basis of the aggregating belief and deem it true. In other words if the new belief X is based on A deemed true, then X is true. Of course, it gets complicated when deciding if a belief is true when more than one hypothesises are available.
The basic beliefs can be of several types: empiric (Hume and Locke), rational intuition (Descartes, Leibniz and Spinoza), innate (Kant, Plato), or conversational contextual.
Coherentism by contrary, states that a belief is true if multiple beliefs are inferred for its justification. But this is not very helpful either. Gettier formulated a scenario (Gettier’s problem) where the assumptions might be true, but the inferred belief is not necessarily true. Gettier’s example of Jones and Smith when they apply for a job and Smith is making a deduction in which he concludes “the person who has ten cents in his pocket will get the job” which proves in the end to be false, although the hypothesis are true, seems to be focused on semantics rather than facts. When Smith said “the person with 10cents in his pocket will get the job” he was thinking of Jones. Thus, the actual belief was that Jones will get the job because he knew he has 10 cents in his pocket. Gettier tries to prove the point by solely focusing on the last sentence that went through Smith’s mind, not on the actual belief. The experiment clearly makes no connection whatsoever between the 10 cents and the job allocation, hence the hypothesis is false anyway.
Also called naturalistic epistemology, this tradition describes the knowledge as produced in natural circumstances and beliefs are considered true based on conditions verified using methods, results and theories specific to empirical sciences. This type of epistemology tends to rely on cognitive psychology and its empirical methods to determine the quality of conditions in which the knowledge is acquired. Quine, a naturalistic epistemologist, considers epistemology as part of psychology, while Thomas Kuhn thinks the social sciences should be applied to epistemology. This approach would solve the Gettier’s problem by qualifying the source of knowledge as not entirely reliable. Mind you, this is not bullet proof because the method cannot be applied to what you don’t know. Smith didn’t know he does not have 10 cents in his pocket, so his statement sounds true.
The fundamental issue I have with the proposition offered by epistemology, that knowledge is about beliefs and justification as an indication of truth, is that it is entirely subjective (even the empirical methods ultimately attempt to “guess” the quality of the subjective thought) and limited to human interpretation and mental storage of knowledge.
With the development of computers and large network systems the idea that knowledge is limited to the human brain and defined by individual beliefs is unsatisfactory. There are two major weak points in the traditional epistemology: knowledge can be stored outside the human brain and used as a repository which is accessible on a need by need basis or through gradual discovery and that knowledge could be distributed across large number of people and shared as common source of knowledge.
The first issue is a bit surprising. Epistemology seems to be stuck in a debate that has only marginally changed since Plato, based around a discourse focused on beliefs as mysterious forms of reflection of the external environment or as outcomes that result from internal mental processes. At a time when information was an inexistent concept and everything was mechanical, far more obvious and easier to recognise than thoughts, the fascination with the mind’s perceptions and deductions was understandable. But know this approach is outdated in my view because it does not recognise the possibility of knowledge created by and with computer systems in vast networks.
The second objection has to do with the lack of recognising the socially created knowledge as something that is acquired by large social group through an iterative process of sharing, collaboration and collective action. The role of social networks is ignored completely, thus missing the opportunity to explore the creation of knowledge at a higher order and implications of availability of knowledge across large populations and geographical areas, including the whole planet.
Social skills are matters of ethereal domain. It is all in the mind of people. Apparently, there are some physical signs that could be linked to the invisible art of social engagement.
In an interview run by Harvard Business Review (HBR) for this January edition, Amy Shelton, associate professor at the Department of Psychological and Brain Sciences, John Hopkins University, says that people with good spatial skills actually have good social skills.
OK, count me in, because I always thought I have good spatial skills!
Not so fast, says Amy. When HBR mentioned that engineers have good spatial skills, but, ahem, they don’t have a reputation for people with a knack for social subtleties, Amy explained that the research found that only a certain type of spatial skills are relevant. It is all about being able to view the perspective of another person. The research used dolls placed at various locations in relation to objects and it required participants to describe what they think the dolls see. This simple skill, the ability to imagine the physical world seeing by someone else, seems to be strongly correlated to social affinity.
Also, it seems that ability to perceive navigation from a different point of view strongly correlates with social skills.
Interesting, isn’t it? Improve your spatial IQ to get higher social IQ. Maybe that works better than using Twitter.
In a recent article (How Amazon Ate Sears’ Lunch) Elizabeth Olson writes an obituary for a company that has been an icon of retail for almost a century. In the past decade or so, Amazon has gradually eroded Sears’ market share to the point where the old superstore had to take the drastic measure of closing over 100 of its outlets.
How is it possible? Sears launched a web site to the public trying to replicate Amazon’s success, but to no avail. Unfortunately, the article I mentioned earlier doesn’t really explain why Sears has lost so much business. I venture to point out one factor that contributed substantially to the failure of the online enterprise, and in fact of the many other retailers who find themselves in a similar position.
The biggest visible difference between the two is in the quality of the cognitive structures presented to their visiting customers. These structures have a tremendous role in helping the customers find information, learn about new products and visualise a purchasing experience before they actually execute it. It is what makes a web site “sticky”.
If you go to Sears’ website you will find a simple interface – that is good – and a sense that not much is happening there – that is bad. The navigation is linear, you have to click through a tree-like structure to get your item or use the search function. The problem with this web site is not only that is not very helpful, but is not working well either which points to a serious problem of the overall design of the information architecture. I searched for TV and it took me to SanDisk SD card, which is shown under the heading Appliances->Sewing & Garmet Care -> Memory Card & Sticks. At the bottom of the screen under “People Who Viewed This Item Also Viewed” there is a list of refrigerators, air conditioning units and washing machines. Yeah, right! It just shows that Sears doesn’t take the importance of the information architecture seriously, and they are paying for that.
But this is not the main problem. The killer here is the lack of evolved cognitive structures that are needed to help customers make sense of the web site. Sears has a design form that is typical to the first generation of ecommerce solutions: use the tree-like structures. In a time when people don’t have time to “learn” the web site, the host must invest in developing sophisticated cognitive structures.
Amazon has a plethora of cognitive helping tools to assist the shoppers: on the left you have a permanent toolbar that shows you exactly where you are at any point in time, with up-to-date number of categories available. Then there is the real-time inventory, shipping information, a long list of filtering criteria, Today’s Deals, Gifts, Listmania, your recently viewed items, browsing history, shipping options, customer reviews, certifications, conditions and the list goes on. When you look at the item individually, the page is filled with helpful and practical information including Frequently Bought Together, What Other Items Do Customers Buy after Viewing This Item, Add to Wish List, More Buying Choices, Technical Details, Product Description, External Web Sites, Customer Reviews, etc.
Amazon is a living beast with a very well organised mind. One of the best executed strategies of Amazon was to get its customers to contribute with content which transformed Amazon into one of the best product information websites. Statistically the reviews are accurate because they are so many and because of the self-regulating system Jeff Bezos put in place from the beginning.
Meanwhile, Sears and others implemented the web site as an afterthought, ignoring the value of the information architecture. The irony is that in essence this is not something new in terms of a selling system. Cognitive structures were used successfully in the past to improve shopping experience, but in a different form. A hundred years ago the successful retailers were those who understood how to assist the customer once they stepped inside their shop: placing carefully useful items visibly near the door, other items near the counter, organising the aisles, hiring helpful shop assistance who would answer any question the visiting shoppers might have, use shelves in an innovative arrangements, place extra helping information for new products, arrange for special stands, etc.
This is exactly what Amazon does in the virtual space. Somehow, the traditional retailers have lost their way and they could not get themselves to apply their experience online. This may be due to the fact that IT departments don’t work well with the sales and marketing. Or maybe there is no one to translate the message from one to another.
In Australia, Harvey Norman struggles to come to term with the new world. Gerry Harvey, the retail chain’s CEO, is trying desperately to setup web sites overseas to win back its customers who are deserting him in droves. But he does not get it. He thinks you just need a web site with buttons to click, the equivalent of a robot made out of pots and springs, wheels and cogs. It is mystifying how he has forgotten how hard he worked to make his business work from its humble beginning.
Information is king and the use of cognitive structures is essential in getting into the mind and soul of the customer. In this sense, the IT is critical to the success of the business, but I should point out that it is the application in a human context that makes it work in the end.
It is easy to think that online search consists of entering keywords on Bing or Google web pages followed by hitting the “search” button. In reality, there is a lot of search that takes place in the mind of the searcher. Part of the information behaviour, better called human information behaviour (HIB), searching plays at multiple levels: cognitive, emotional and situational. Psychology has a great role in the information behaviour. The design of information system user interface has only recently started to take into consideration the human aspects of searching and the fact that seeking information is far from being a linear process limited to the use of sequential keywords.
The quality of search is perceived as the degree in which the search engine responds to the explicit query, but this is not that simple. The user doesn’t always know which keywords to use for searching. Very often the user wants to broaden the search scope, delegate control and navigate through responses offered by the search engine to discover new information. This is a learning journey that many users prefer when the problem is not clearly defined in their mind and they feel the need to know more before they actually decide where to focus their information search moving forward.
The opposite of direct response, which is an area where traditional engines have tried to excel, is a situation where the search tends to broaden rather than focus. This divergent search is most appropriate for creative situations when users acquire new information outside their domain of expertise. Convergent search focuses on what the user knows, while the divergent search is about discovering new things, and adopt a multi-disciplinarian attitude.
Facebook is clearly in the camp of divergent search despite the obvious lack of search capabilities. In fact FB friends find a lot of information on Facebook, but not by way of convergent search and by seeking specific information. On Facebook users encounter information. With each visit, through sharing information is presented to the user in the Newsfeed in form of links and commentaries. This is not an accidental feature. This year Mark Zuckerberg announce the addition of a new and powerful feature called serendipity. It may seem innocent, but this feature is very powerful because it brings friends closer and because it has a subtle attraction that keeps users coming back to the web site. And of course, it is a key element in the Mark’s strategy to gather as much information about our habits and wants.
Of the two ways of finding information, search and encounter, the latter is one that requires the least amount of effort on behalf of the user. Couple that with the element of surprise and reward that comes in form of viewing a funny video, learning something interesting, listening to a great song or reading a captivating article, when none of this could have cross your mind without at least putting in some effort, and you get hooked. Serendipity is one of the most powerful ways of finding information you did not know you needed or wanted.
The job of search engines is difficult because it requires the user to think and make an effort to research their own information need. It is like work. That is Google and Bing. Facebook on the other hand is strong on serendipity which is fun and enjoyable and very, very social. Not only you find information, but you get to leave comments, “like” and be “liked”. On search engines, no one likes you.