Gauging Consumer Behaviour via Social Analytics

28
Gauging Consumers Through Their Online Behaviour Hareesh Tibrewala [email protected]

description

 

Transcript of Gauging Consumer Behaviour via Social Analytics

Page 1: Gauging Consumer Behaviour via Social Analytics

Gauging Consumers Through

Their Online Behaviour

Hareesh [email protected]

Gauging Consumers Through

Their Online Behaviour

Hareesh [email protected]

Page 2: Gauging Consumer Behaviour via Social Analytics

Consumer Insights

• Object of marketing / advertising / PR…is to finally influence consumer behavior in a way that it is beneficial for your brand

• For the first time in the history of brand-kind, there is an opportunity to track consumer behavior– In real time– On a large scale– Access to authentic information

• Social Analytics derived from Social Media Listening

Page 3: Gauging Consumer Behaviour via Social Analytics
Page 4: Gauging Consumer Behaviour via Social Analytics

Listening to these consumer conversations has become increasingly important

Page 5: Gauging Consumer Behaviour via Social Analytics
Page 6: Gauging Consumer Behaviour via Social Analytics

Customer Relationship Management (CRM)

• Consumers are using social media platforms to share their opinions about brands

• In case a consumer puts up a complaint about a brand, it is important for the brand to engage with the consumer and to be seen as a responsive brand

• In case a consumer puts up a positive review about the brand, brand should engage with the consumer and use this opportunity to generate favorable brand advocacy

Negative

Negative

Positive

Positive

Complaint

Brand Respons

e

Customer

Response

Page 7: Gauging Consumer Behaviour via Social Analytics

Positive Neutral Negative SIM Score Positive Neutral Negative SIM Score

Booking and Customer Care 0% 20% 80% -0.60 0% 29% 71% -0.41

General Feedback 40% 26% 34% 0.32 13% 37% 63% -0.13

On-ground Services 33% 8% 58% -0.17 0% 55% 45% 0.09

On-board Services 26% 26% 48% 0.04 8% 79% 21% 0.67

Punctuality 38% 7% 55% -0.10 0% 3% 97% -0.94

Corporate 25% 73% 3% 0.95 3% 27% 73% -0.44

Kingfisher AirlinesIndigo Airlines

Vertical

Understanding Brand Sentiment

• Listening to what consumers are talking about a brand, helps understand and map consumer sentiment

• Not only can one map the sentiment for one’s own brand, one can do it for competitors brand as well

• Understanding consumer sentiment in the marketplace can help create actionable product / communication / customer service strategies

Sentiment Analysis for two airline carriers

Aspects (verticals) of the

air carriers

Ranking

Metric

Share of each sentiment

Area of Concern

Area of Concern

Page 8: Gauging Consumer Behaviour via Social Analytics

PR Crisis Alert

• PR Crises now-a-days generally tend to start from social media and then at some point of time hit mainstream media

• Monitoring social media platforms on an ongoing basis can help identify an emerging crises

• A timely response management system can help prepare for the crises and ensure that a major negative PR event gets averted

0 hours 5 hours 8 hours 15 hours 24 hours

Extremely High

High

Moderate

Low

Harmless

First appearanc

e

Initial Conversations

Rapid Sharing

Mainstream Media

Internet Publications

Page 9: Gauging Consumer Behaviour via Social Analytics

Identifying Sales Opportunities

• Just as brands are looking for customers, the customers are also looking for products

• Social Media helps identify situations where a potential customer may be looking for your brand

• One can then guide the conversation with that customer into a sales opportunity

Page 10: Gauging Consumer Behaviour via Social Analytics
Page 11: Gauging Consumer Behaviour via Social Analytics
Page 12: Gauging Consumer Behaviour via Social Analytics
Page 13: Gauging Consumer Behaviour via Social Analytics

Generating Business Intelligence

• Listening to conversations on Social Media allows brands to capture data which they would have otherwise missed

• This could have been data about their product, brand, service, category or industry

• This data – which is conversations among people, can be scrutinized to extract business intelligence

• This could be predictive information about sales, a perception matrix about your brand or product, among other types of intelligence

• This is actionable intelligence, which you can use to take more informed decisions

Page 14: Gauging Consumer Behaviour via Social Analytics
Page 15: Gauging Consumer Behaviour via Social Analytics

• Great Pedigree– A part of Salesforce.com, a $20bn market cap

company– Global presence

• Robust Technology– Fetches conversations from depth of digital

universe– Real-time data discovery– Relationships with Twitter, Facebook etc for live

feeds• Market Leader in Monitoring Tools

– Clients include Pepsi, Dell, L'Oreal, Fuji Film, UPS, 3M, Commonwealth Bank, KLM, Queen’s University, Mayo Clinic, Edelman, Golin Harris, Bissel, Crocs, Intuit, Durex, Airwick, Clearasil, Nurufen, Reckit Benckiser, Bell Aliant, Southwest Airlines, Microban, Dettol, and many more

• Excellent User Interface and Reports– Customisable UI– Real time analytics

Radian6 – The Enterprise Monitoring Tool

Page 16: Gauging Consumer Behaviour via Social Analytics

Monitoring Tool

To extract relevant conversation

Human Intelligence

To convert data into actionable

intelligence

A mix of social media conversations both relevant and irrelevant to our

search

A mix of social media conversations both relevant and irrelevant to our

search

Purely relevant conversationsPurely relevant conversations

Page 17: Gauging Consumer Behaviour via Social Analytics

Predicting Personality Traits

• Paper published by Kaggle.com (2012)

• Carried out an experiment that involved analyzing 2927 Twitter user handles

• Profile attributes of the handle as well historic Twitter data was analysed

• 586 different features were studied– Friends, Followers, Number of tweets, Number of RTs – Average number of followers of my followers, use of predefined

words– Use of pronouns (“I”, “We”)

Page 18: Gauging Consumer Behaviour via Social Analytics

Predicting Personality Traits

• More attributes– Use of swear words– Use of numerals in the Tweet– References to family and friends– Emotions expressed

• Were able to predict and correlate behaviour of a person with the words the person uses

• In spite of the fact that a person may be very careful about what he Tweets, it is the choice of words that he uses to communicate that gives away his personality

Page 19: Gauging Consumer Behaviour via Social Analytics

Predicting Outbreak of Diseases

• Project by the US Centre for Disease Control and Prevention

• By looking at spike in search times in Google results is able to predict outbreak of Flu or Dengue epidemic (Google.com/ FluTrends)

• By looking at Twitter and other conversations on social networks is able to track diseases and natural disasters (Mappyhealth.com)

Page 20: Gauging Consumer Behaviour via Social Analytics
Page 21: Gauging Consumer Behaviour via Social Analytics

Screen Shot 2013-12-05 at 10.01.51 PM.png

Page 22: Gauging Consumer Behaviour via Social Analytics
Page 23: Gauging Consumer Behaviour via Social Analytics
Page 24: Gauging Consumer Behaviour via Social Analytics
Page 25: Gauging Consumer Behaviour via Social Analytics

Twitter Mood Predicts Stock Market

• Research done by Johan Bullen and 3 other researchers are University of Cornell

• “Using tools like OpinionFinder and GPOMS, which measures mood in terms of 6 dimenions (Calm, Alert, Sure, Vital, Kind, Happy), we cross validated market swings with mood swings

We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the DJIA and a reduction of the Mean Average Percentage Error by more than 6%.”

• New Business Model : Zingfin.com

Page 26: Gauging Consumer Behaviour via Social Analytics

BlueFins.com : Social TV

Page 27: Gauging Consumer Behaviour via Social Analytics

Online Market Research

• Client profile– Global pharmaceutical brand

• Challenge– Wanted to some insights

pertaining to factors that influence the buying pattern for patients with diabetes

• Solution– More than 100 communities of

diabetes patients / care givers were identified

– More than 3000 conversations over a 90 day period were mined, classified and analyzed

– This analysis was used to help the brand gain insights into factors that influence the buying pattern

• Outcome– Research done on social

platforms corroborated findings from a traditional market research exercise which was also commissioned by the brand

Page 28: Gauging Consumer Behaviour via Social Analytics

Questions?

If you need a copy of this presentation, please leave your business card. We will

email it to you.

Hareesh TibrewalaJt. CEO, Social [email protected]

Company blog: blog.socialwavelength.comLinkedIn: linkedin.com/in/hareeshtibrewala