New Book: Creating Value With Social Media Analytics

cover (10)

Often termed as the ‘new gold,’ the vast amount of social media data can be employed to identify which customer behavior and actions create more value. Nevertheless, many brands find it extremely hard to define what the value of social media is and how to capture and create value with social media data. In Creating Value with Social Media Analytics, we draw on developments in social media analytics theories and tools to develop a comprehensive social media value creation framework that allows readers to define, align, capture, and sustain value through social media data. The book offers concepts, strategies, tools, tutorials, and case studies that brands need to align, extract, and analyze a variety of social media data, including text, actions, networks, multimedia, apps, hyperlinks, search engines, and location data. By the end of this book, the readers will have mastered the theories, concepts, strategies, techniques, and tools necessary to extract business value from big social media that help increase brand loyalty, generate leads, drive traffic, and ultimately make sound business decisions. Here is how the book is organized.

Chapter 1: Creating Value with Social Media Analytics
Chapter 2: Understanding Social Media
Chapter 3: Understanding Social Media Analytics
Chapter 4: Analytics-Business Alignment
Chapter 5: Capturing Value with Network Analytics
Chapter 6: Capturing Value with Text Analytics
Chapter 7: Capturing Value with Actions Analytics
Chapter 8: Capturing Value with Search Engine Analytics
Chapter 9: Capturing Value with Location Analytics
Chapter 10: Capturing Value with Hyperlinks Analytics
Chapter 11: Capturing Value with Mobile Analytics
Chapter 12: Capturing Value with Multimedia Analytics
Chapter 13: Social Media Analytics Capabilities Chapter 14: Social Media Security, Privacy, & Ethics

The book has a companion site (, which offers useful instructor resources.

Available on:

Posted in Uncategorized | Leave a comment

Our New Book: Digital Analytics for Marketing

I am pleased to announced that our new book on Digital Analytics for Marketing is now published:

new book

Posted in Uncategorized | Leave a comment

4 Types of Social Media Analytics Explained

Social media analytics can take four different forms, namely, descriptive analytics, diagnostic analytics, predictive analytics, and prescriptive analytics. Read more here.

Posted in Uncategorized | Tagged , , , , , , | Leave a comment

New Book on Social Media for Government


I am pleased to announced that my new book, “Social Media for Government: A Practical Guide to Understanding, Implementing, and Managing Social Media Tools in the Public Sphere” has been published by Springer.

The offers practical knowhow on understanding, implementing, and managing main stream social media tools (e.g., blogs and microblogs, social network sites, and content communities) from a public sector’s perspective. The book also provides practical guidance in developing a social media policy and strategy, and addressing issues such as those related to social media analytics, security, and privacy.

Posted in Uncategorized | Leave a comment

A Social Media Analytics Perspective on Australian Elections

Gohar F. Khan, PhD

On July 2 2016, Aussies will be going to the ballet box to choose their next premier. In an effort to be chosen, the candidates have launched highly competitive campaigns through a variety of media including social media. At the forefront of this is Twitter where candidates have developed social media campaigns that directly engage with their audiences.  While the two leading Aussie candidates are active on Twitter, it is not clear who is comparatively more popular, interactive, or influential. Social media analytics can help us find out. Thus, I analyzed Malcolm Turnbull and Bill Shorten’s Twitter presence to contrast and compare them in terms of interactivity, influence, and popularity.

Interactivity—it is the overall interaction and responsiveness of a Twitter user and can be measured with:

  • Re-tweet %: Percentage of re-tweets in the total tweets of a candidate. The higher the percentage the more a user interacts with others.
  • Replies %: Percentage of replies in the total tweets of a candidate. The higher the percentage the more a user interacts with others.
  • User mentions: the average number of people mentioned per tweet by a candidate. The higher the percentage of mentions the more interaction is happening.

Let us begin our discussion by looking at the interactivity statistics extracted from more than 3,000 tweets per candidate tweeted over the last 4 years (Table 1).  The last four years Twitter history graphs (figure 1) shows that both Turnbull and Shorten are consistently active on Twitter averaging more than two tweets per day. This analysis also highlights some seasonable trends, for example, Shorten’s tweets spiked on March 15, 2016 when he appeared in the ABC news channel. It is interesting to note that in terms of interactivity (i.e., re-tweets, replies, user mentions), Turnbull is leading the way. However, when looking at the contents of the re-tweets and the nature of the replies, we found that most of the interactivity is occurs within political circles and not with general public. For example, Turnbull mostly re-tweeted Christopher Pyne (@cpyne) the Minister for Industry, Innovation & Science, the Dept. of Communication, and Senator Cash. And Shorten mostly re-tweeted his own party (@AustralianLabor) and the Labor Herald (@LaborHalard).

Table 1: Candidates Interactivity over Twitter as of June 4, 2016.

Candidate Avg. Tweets/day Re-tweets % Replies % User Mentioned/tweet
Turnbull 2.83 21% 25% 0.59
Shorten  2.48 19% 9% 0.42

Figure 1: Turnbull’s (left) and Shorten’s (right) Tweet history

Influence—the capacity to have an effect on others which can be measured as:

  • Tweets favorited: the proportion of a user’s tweets favorited by others.
  • Re-tweets: the proportion of a user’s tweets re-tweeted (or transmitted) by others.
  • Mentions: number of times a candidate is mentioned by name. The higher is better.

The influence statistics, however, paint at a different picture (see Table 2 and Figure 2 respectively). Interestingly, Shorten  emerged as a more influential social Twitter user. In order to understand the influence statistics, let’s first understand the network figures 2 where nodes represent candidates and their followers and links represents mentions i.e., mentioning a person by name (or the so called Twitter handler). The size of the node is based on in-degree i.e., the number of mentions received by person; the size is bigger when a person receives more mentions. For clarity reasons, only the most influential users are shown. Both the candidates are prominently visible in their networks. Shorten’s network density and clustering co-efficient, however, are higher than Turnbull’s which indicates that Shorten’s network is more cohesive. Close to 70% of Shorten’s tweets were re-tweeted (or transmitted) and 77.78% of his tweets were favorited by his followers which are much higher than Turnbull’s percentages. Based on this, a conclusion can be reached that Shorten is comparatively more influential and his tweets (messages) resonate more with the audience.


Figure 2: Turnbull’s (left) and Shorten’s (right) Twitter mentions network

One fun thing to note is that Shorten mentioned Turnbull 169 times in his network; more than any other member of his network. While Turnbull mentioned Shorten only  46 times in his network. This may indicate that Shorten is taking a more aggressive stance towards his political counterpart.

Table 2: Candidates’ Influence over Twitter as of June 4, 2016.

Candidate Nodes/links Clustering coefficient Mentions tweets retweeted % Tweet Favorited %
Turnbull 675/2,436 0.152 489 62.80% 64.52%
Shorten 718/2613 0.275 545        79.9% 77.78%

Popularity—it is the overall fame of a Twitter user and can be measured with:

  • Followers: the number of fans/supporters of a user. The higher is the better.
  • Follower/following ratio: it is the ratio of followers per person followed. A ratio more than 1 means a user is more popular.
  • Listed: the number of people who added a user to their public list.

And finally, in terms of popularity, with more than 9 thousand followers, 3541 listed, and a follower/following ratio of 137 (i.e., for each person followed by Malcolm receives 173 followers), Turnbull beats Shorten. Shorten had 3,647 followers, 1,244 listed, and follower/following ratio of only 13. A much higher followers/following ratio with a huge number of followers indicates that Turnbull is comparatively more popular than his counterpart.

As an overall conclusion, we can say that over Twitter, Turnbull is more interactive and popular, whereas, Shorten is more influential. However, the real question is whether this type of analysis can be used to predict the future of election outcomes? I will leave this question to future research.

Posted in Uncategorized | Leave a comment

CFP: Social Media Analytics Track at the PCAIS 2016

We are chairing a track on “Social media Analytics & Related Issues” at the 20th Pacific Asia Conference on Information Systems (PACIS 2016), June 27 – July 1, Chiayi, Taiwan. Please consider submitting your original research.  For more details please refer to the  conference website:

Posted in Uncategorized | Leave a comment

New book on social media analytics

book cover2I am pleased to announce my new book on social media analytics. The book offers concepts, tools, tutorials, and cases studies to understand and analyze the seven layers of social media data, including text, actions, networks, apps, hyperlinks, search engine, and location layers. 

It is non-technical in nature best suited for business (and information systems) students, professors, and manager. Here is how the book is structured:

  • Chapter 1: The Seven Layers of Social Media Analytics
  • Chapter 2: Understanding Social Media
  • Chapter 3: Social Media Text Analytics
  • Chapter 4: Social Media Network Analytics
  • Chapter 5: Social Media Actions Analytics
  • Chapter 6: Social Media Apps Analytics
  • Chapter 7: Social Media Hyperlinks Analytics
  • Chapter 8: Social Media Location Analytics
  • Chapter 9: Social Media Search Engine Analytics

Chapter 10: Aligning Social Media Analytics with Business Goals

The book also comes with a companion site ( which offers Updated Tutorials, Power-Point Slide, Case Studies, Sample Data, and Syllabus.

It is available through Amazon Store and CreateSpace Store.

Thank you,

Posted in News | Tagged , , , , , , , | Leave a comment

Virality over YouTube: an Empirical Analysis

Why some content go viral on YouTube? In order to answer this questions, in a recent study (accepted for publication in Internet Research journal),  we found that popularity of the videos was not only the function of YouTube system per se, but that network dynamics (e.g., in-links and hits counts) and offline social capital (e.g., fan base and fame) play crucial roles in the viral phenomenon, particularly view count.  

For more details on the study design an  other  findings, you can read the authors’ version of the article here


Khan, G. F., Sokha, V., (2014), Virality over YouTube: an Empirical Analysis, Internet Research, accepted for publication. Download Authors’ version.

Posted in Publications | Tagged , , , | Leave a comment

What is Government 2.0?

Government 2.0 is “a governance culture of transparency, openness, and collaboration facilitated by social media” (Khan, 2013, p. 8)[1].

In other words,  Government 2.0 is not just “likes,” “tweets,” and mere establishing social media presence (e.g., creating a Facebook fan page or a government Twitter account), but it requires (or should be complemented with) a governance culture of transparency, openness, and collaboration. Without a mindset and culture of transparency, openness, and collaboration, establishing social media presence is useless and ineffective.

These and other Government 2.0 fundamental concepts (such as, government 2.0 implementation scenarios, utilization models, and relationships) are discussed, in a systematic way, in my latest publication entitled, “Government 2.0 utilization models and implementation scenarios,” published in the Information development journal. You can download author’s version of the paper here.


[1] Khan, G. F. (2013). “The government 2.0 utilization model and implementation scenarios.” Information Development


Posted in concepts | Tagged , , | Leave a comment

Social Media for government

I am helping  the United Nations Asia and Pacific Training Centre for ICT (UN-APCICT) to develop a module on Social Media for GovernmentThe main purpose of the guide is to help governments—particularly those in the developing world—to implement a transparent, open, and collaborative government by means of social media tools. For example, after reading this guide, you will be able to

  • Understand, configure, and manage main stream social media tools (e.g., blogs and micro-blogs, social network sites, and content communities) to socialized government information;
  • Understand, configure, and manage collaborative social media tools (e.g., wikis, cloud-based tools, social tagging) to establish inter agency mass collaboration;
  • Understand, configure, and manage analytic tools to monitor and measure social media.

Last week, I presented content of the module at a gathering of representatives from 32 member States across Asia-Pacific gathered in Bali. You can see the content of the guide at Slideshare. Once finalized, the module will be available at the UN-APCICT academy for users to read and download (free of charge).


Posted in Uncategorized | Tagged , , , | Leave a comment