What makes social media content go viral?

Capptain ObviousUsers of social media are most likely to be motivated to share online content if it is interesting, important or funny, according to a new global study of social media usage.
In a poll of 12,420 global internet users conducted by Ipsos Open Thinking Exchange, the global innovation centre of market research firm Ipsos, 61% of those who indicated they had shared some content online over the past month said their main reason was because they wanted “to share interesting things”. (source)

Read this in the news today and was wondering how insightful such research could be.

There was a good case from PolicyMe though, that used behavioral analysis to get their content viral:

Play to emotions

A lot of behavioral science boils down to seeing humans as emotional creatures, Plank says. “Although we like to think we are rational beings, we are in fact driven by our emotions.” That means we respond to bold language in headlines, and content that makes either delighted or outraged. Use a clear voice and stir up those emotions, Plank says, and your stories are bound to be attention-grabbers.

Deploy the “curiosity gap”

This is an idea PolicyMic copied from Upworthy – it dictates that you should reveal some tantalizing information in a headline (or tweet or status update, or whatever), but not give away too much of a story. The headline should beg to be clicked on in order to reveal its mystery. It should stoke, but not completely satisfy, a reader’s curiosity. Example: “The libertarian answer to the minimum wage debate.”

The underdog-overcomes-bully narrative is always a winner

Readers love a “David and Goliath” story, so play it up. Example: This piece about a “hero mom” who stood up to an officer for sexual assault in front of her infant daughter got almost no attention on PolicyMic, until Plank changed the headline to demonstrate the bullied-confronts-bully dynamics.

People are suckers for protest stories

Just as the Brazil protests attracted a lot of interest on the day highlighted in Horowitz and Plank’s post on the Knight Foundation’s blog, PolicyMic saw a similar spike in interest around stories about the Occupy Gezi protests in Turkey.

Use a lot of images

Posts with lots of pictures are quick and easy to read and eminently shareable, apparently. BuzzFeed with its famous “listicles” is obviously more than aware of that.

Recognize the difference between optimizing for search and optimizing for social

This story about the Turkey protests, “Turkey protests: Has the Arab Spring finally reached Turkey?”, has a headline written for search-engine optimization purposes. You’ll notice it has all the key terms “Arab Spring,” “Turkey protests,” to catch the attention of Google’s crawlers. However, the piece has no tweets and just five Facebook shares.

This story about the Turkey protests, on the other hand, is optimized for social. It carries the headline “Taksim Square protest: 11 images from Turkey that will give you the warm fuzzies.” The combination of high-arousing emotion, pictures, and the important event makes it a social winner. It got 48,000 shares on Facebook and 1,400 shares via Twitter. Pandodaily

Negative buzz affects sales

McKinsey found out that social media buzz does affect sales. They, however, were looking at negative buzz and its correlation with drop in sales.

The consulting firm found bad buzz for an unnamed telecom client hurt signups by 8%, “offsetting their entire TV spend,” McKinsey principal Jonathan Gordan said in a presentation Monday at the Advertising Research Foundation’s Audience Measurement 8.0 conference in New York. (AdAge Digital)

And here the insights come not from sentiment alone. Consultants did drill into the data and found out that the negativity was mainly about the bad quality of the sign up process.

Complaints about the sign-up process and call-center workers at the telecom provider drove the negative sentiment, Mr. Gordon said. The experience, he said, convinced him, “When we get into this stuff and analyze it properly, it will drive insights.”

I would actually prefer not to link social media buzz to the drop in sales. We do not have data proving that social media comments were directly affecting the decision to choose a different provider. What is more obvious is that negative social media buzz will point at the problem that can directly affect the sales. And this is one possibility to link your social media listening to the sales.

The same way we can discover a positive effect that social media has on sales. The funnel here is the following:


Here is one of the examples for such model taken from Nokia-Brandtology case:

Through Brandtology we discovered that a lot of the consumers were excited about the new Nokia Lumia 920’s device bold and colorful designs but were unsure about its Windows Phone software and exclusive Nokia Services experience which is new to the market. Hence we have created a dedicated Nokia Lumia experience channel on our Youtube page to educate the consumers about the superior experiences offered in Nokia Lumia. This experience showcasing has successfully guided our consumers from Awareness to Consideration and ultimate towards purchase at our Nokia Retail stores, a win for both Nokia and Brandtology for using Social Media insights to ultimately drive sales. (Brandtology)


“All right,” said Deep Thought. “The Answer to the Great Question…”
“Of Life, the Universe and Everything…” said Deep Thought.
“Is…” said Deep Thought, and paused.
“Forty-two,” said Deep Thought, with infinite majesty and calm.”
― Douglas AdamsThe Hitchhiker’s Guide to the Galaxy

“Likes and followers do not mean anything”. This is the buzz phrase and the tagline of many speeches on measurement and analytics.

What question we ask determines what we count at the end. So is it that surprising that after being presented with a question from a client “I want to see community’s growth”, marketers would count likes?

So, quite recently Coca Cola mentioned that social media had no impact on short-term sales (which she later clarified that social media is still important, but the “damage” was done).  I found Coca Cola’s three year old social media strategy -a very comprehensive presentation (can find the full ppt on slideshare) and did some analysis on their objectives trying to find how they planned to link the social media to sales.


The presented achievements are very qualitative, and not a single link to the sales is presented.

One more slide is basically presenting us with the objectives of Coke for social media: brand spirit and intent. Intent is closer to sales but is still quite far on the purchase funnel.


Having this objective in mind, this is what the approach to measurement the success:


I could not find a more recent presentation of Coke’s social media strategy, so maybe the sales got into the picture. But for an average brand sales are usually very vaguely present in the social media strategy. It is hard but possible to link social media to sales and I am going to continue searching for good cases of such strategies.

The key to success is always – asking the right  question. As a client if you want sales, do not ask about followers. This is the task of social media strategist to link the followers to sales and prove that the link is valid.

Social media for e-commerce

New research from Google … shows that social media is very much an awareness-building vehicle, and less so a purchase-driving tactic. In fact, of all the online behaviors studied by Google—which are almost identical to those examined by Monetate—social media is the second “earliest” (after only clicks on display advertising) in the average U.S. consumer’s purchase path. (from the recent E-Commerce Quarterly Research by Monetate)


More “news” about social media not living up to great expectations.

Given the extreme stickiness of social networks (especially Facebook and Pinterest), it is indeed possible that part of the issue with social commerce is one of session interruption, whereby consumers don’t want to leave the visceral comforts of their social network by clicking a link and going shopping online, but would rather store the awareness of the deal in their noggin and visit the website later, when they are less ensconced in social media bliss. This, of course, would culminate in a website visit from the direct source, with no credit accrued to the ignition source of social.

Yes, this.

The Best Product on the Market of Social Media Research

So guess it’s time to start blogging again. Almost 7 years ago in 2006 I started a blog on PR 2.0 — not even a buzz word in Moscow, Russia then. I was working in corporate communications and was mainly busy with media monitoring. For a corpcomms person, the main revelation was that blogs share and distribute information very freely. And after a few incidents when we found info about the company on social media — we started monitoring blogs on a regular basis. So when I started a blog,  the most thrilling topic was what place will PR people take in the new world when journalists can share information directly with the bloggers and then with the wide audience.

Very short time passed and the blog turned into my own social media research company “NetMind” (Russian only). I left corporate world and started my own agency, where we focused on social media monitoring and research. These were the most exciting years and it is getting only better with time.

Now I joined a global social media intelligence company and I found out that I almost stopped blogging. Time to go back to blogging. With all the exciting things happening with business intelligence and data analytics and 2013 being a year of statistics – there are more than enough reasons for that.

One of the most recent surprises was the shut down of almost all NM Incite/Buzzmetrics office. This was even more unexpected since they were hiring for quite a while here in Singapore.

“NM Incite is winding down its social listening platform to focus on areas of social media analytics where it has a competitive advantage and can deliver truly unique, differentiated solutions – like social TV,” a spokesperson told TechCrunch. link

The market is crowded indeed, however there is no strong leader. One could say Radian6 is one, and that would be partly true, but only for the market of automated platforms. Out of all competitors, companies like Nielsen do have the potential to provide the unique product that is in the greatest demand: consultancy and analysis on top of the social media data. But the company sees its unique product in social TV analytics, building on their “exclusive partnership” with Twitter.

Interesting topic here is: what is the competitive advantage now in the social media research market and who will get it and become the leader in the industry? The common choice now is between good data and good analysis. So most likely the industry leader will have both. The trick is to determine what is “good” both for data and analysis.