SOCIAL MEDIA MONITORING & ANALYTIC TOOLS Case Study
Jul 15, 2015
SOCIAL MEDIA MONITORING & ANALYTIC TOOLS
Case Study
02
SOCIAL MEDIA
SOCIAL MEDIA
03
SOCIAL MEDIA
04
HISTORY
Follower Engagement Response ?
05
HISTORY
Social Media
Data Metric Pattern Knowledge
06
SOCIAL MEDIA SOLUTION
Tools Expert Solution
07
HOW IT WORKS
Brainstorming
Objective Identification
Research & Assessment
Strategy Building
Execution
Measurement & Evaluation
08
OBJECTIVE
• Performance Assessment
• Customer Service
• Competition Landscape
• Customer Understanding
• Campaign Assisting
• Industry Trend
• Channel Optimization
• Reputation Management
• Product Decision Journey
• Content Marketing Optimization
STUDY CASE
10
INDEX
• Banking
• Pharmaceutical and Healthcare
• Entertainment
BANKING
12
BANKING
Type
Pre-Campaign Assisting
Background
Bank A is one of the biggest Bank Indonesia. In order to develop a product that suits people need, Bank A need to understand the people’s behavior
13
BANKING
Challenge
Indonesia demography currently is dominated by young people who become the focus in the product development
Objective
Understanding and comparing people’s financial behavior in terms of spending and saving
Time frame: 13-20 March 2014
14
BANKING
Word Analysis
• Defining Twitter keywords that suit with saving and spending behavior
Buzz Monitoring
• Crawling all tweets which contain defined keywords
Peak Time Analysis
• Looking into incoming mention number enhancement in some period of time
Method
15
KEYWORD ANALYS
SAVING
Deposito DebitReksaDana
Tabungan Investasi
SPENDING
Dugem Nongkrong Belanja
Liburan Nonton
16
SAVING VS SPENDING BUZZ
30393
16240
9179
3495
875
130
798406
167622
148025
61978
8740
0 100000 200000 300000 400000 500000 600000 700000 800000 900000
Nonton
Liburan
Belanja
Nongkrong
Dugem
Investasi
Asuransi
Tabungan
Reksa Dana
Debit
Deposito
Spending Saving
17
SPENDING TRENDS
• Play
• Go to vacation to Bali and EuropeLiburan
• Movies
• TelevisionNonton
• Clothes
• FoodBelanja
• Drinking Coffee
• Surfing InternetNongkrong
18
SPENDING TRENDS
12.00-16.00 16.00-20.00 20.00-24.00
Sunday Belanja Liburan
Monday Nonton Nonton
Tuesday Nongkrong
Thursday Liburan
Liburan
Nongkrong
19
BANKING
Result
• People more interested on spending than on saving, proved by total conversation and tweet share of spending much higher than saving. Conversation about saving is only 5% of spending conversation.
• People like to talk about spending, mostly about shopping, going holiday, watching movies and hanging out, after 12 p.m. on some days.
LESSON LEARNED
20
Determining the right keyword is vital as it will greatly affect the monitoring result
• The result of wrong keyword will not related to the research’ topic. Thus, It has no value and useless
PHARMACEUTICAL & HEALTHCARE
22
PHARMACEUTICAL AND HEALTHCARE
Type
Customer Understanding
Background
A pharmacy company Y has a facebook fanpage Y Health Page and Twitter @yealthwhich are used to build a community focus on healthcare. Y didn’t use the social neither to promote nor selling their product
23
PHARMACEUTICAL AND HEALTHCARE
Challenge
Increase customer loyalty by giving information which suit to community’s interest
Objective
Understanding the behavior of the active member of X Health community
Time frame: 15-26 September 2014
24
PHARMACEUTICAL AND HEALTHCARE
Method
Media Monitoring
• Measuring every social media metric of Y Health
Media Comparison
• Comparing community performance of each Y Health social media which is Facebook and Twitter
Social Media Ethnography
• Analyzing community members characteristic by implementing ethnography method on social media
25
MEDIA MONITORING AND COMPARISON
The Facebook account drives more conversation than Twitter account
3.38%
0.50%
0.00%
0.50%
1.00%
1.50%
2.00%
2.50%
3.00%
3.50%
4.00%
Engagement Rate
Facebook (106.232) Twitter (2.936)
ETNOGRAPHY
26
Ethnography (n) : sociological method
that explores how people live and
make sense of their lives with one
another in particular places
(Columbia University)
ETHNOGRAPHY FRAMEWORK
27
Space Objects Acts
Actors Time Activities
Feelings Events Goals
28
CASE STUDY: TIME, GOALS, FEELING
Morning(6-10)
Noon(10-14)
Afternoon(14-18)
Night(18-22)
Participating in engagement activities
50% 0% 100%
Sharing Experience -20% 55.6% 14%
Asking Further Question -66.7% -100% -100%
Praising and Saying Thank You
100% 100% 100%
29
PHARMACEUTICAL AND HEALTHCARE
Result
• Building community engagement on Facebook is giving better result than
on Twitter.
• The best time of publishing post to get better sentiment is at night,
meanwhile the bad time is at noon. The best sentiment score is gotten by
praising activities. On the other hand the worst sentiment score is gotten
by asking further question activities.
LESSON LEARNED
30
• Social Media could be used as a tool to determine the ethnography of a particular group
• The advantage of using Social Media compared to field monitoring is data trackback ability that could speed up the monitoring process
ENTERTAINMENT
32
ENTERTAINMENT
Type
Channel Optimization
Background
Singing competition X gives audience the feeling of being judge by giving them a mobile application that can be used to give vote which account for 70% of the contestant final score. The mobile app is integrated with Twitter and will automatically tweet and mention @AjangX when a vote is given.
33
ENTERTAINMENT
Challenge
Activation of Audience who mostly are young people with smartphone and a social media avid user
Objective
Identify the effect of mobile app usage towards the social media account of @ajangX
Time frame: 19 Oktober-8 November 2014
34
ENTERTAINMENT
Metode yang digunakan
Buzz Monitoring
• Getting all mentions to @ajangX
Peak Time Analysis
• Looking into incoming mention number enhancement in some period of time
Mention Categorization
• Categorizing incoming mention into organic mention and mention from official app
35
PEAK TIME
0
5000
10000
15000
20000
25000
30000
35000
40000
Mention
The number of mentioner increase rapidly (40-60x) on Friday
36
PARTICIPANT MENTION CATEGORIZATION
• App : Tweet that comes from Ajang X’s official app. Marked by “I Voted YES/NO” phrase on the tweet
• Organic: General tweet from people using other type of application
37
ADOPTION RATE
33% of mentioner is using AjangX official app to vote for their idol
33%
67%
App User Non-User
38
MENTION DISTRIBUTION
The tweet from app account for 25% of all mentions to @ajangX during
the performance day
7737 8002 7351
27644 23729 28481
0
5000
10000
15000
20000
25000
30000
35000
40000
1st week 2nd week 3rd week
App Organic
39
VOTING PERFORMANCE
Most people use the app to support their idol instead of bringing down
the competitors
73457633
8322392
365
357
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
1st week 2nd week 3rd week
YES NO
40
VOTING PEAK TIME
Most people use their vote between 22.00-23.00. The number of votes is very dependent on the contestant that was performing
0
500
1000
1500
2000
2500
3000
3500
4000
4500
21.00-22.00 22.00-23.00 23.00-24.00 24.00-01.00
1st week 2nd week 3rd week
41
ENTERTAINMENT
Hasil
• Enhancement of total mention to @ajangX account is 60 times bigger on the show day than usual and 50% of it are from mobile application used for voting
LESSON LEARNED
42
Co-creation can be implemented by integrating mobile app with social media.
• Social media give reason for people to download and use the app because they like to show off their experience
• People also have a chance to appear on national television when they give their vote
• Mobile app is one of ways in conducting co-creation. Besides, AjangX also conducting social media campaign and meet & greet with idol to build the co-creation ecosystem
43
In the end, social media monitoring is just a tool.
Sometimes we don’t really need a sophisticated
and complicated metric. What we really need
though, is a great methodology
There is NoLimit in the future
Have any question?Contact to : [email protected]