Operationalizing Customer Analytics with Azure and Power BI
Ā
dabblr_report_final
1. To draw a co-relation between new app
users and various factors influencing the
influx of new users in the cases of addition
of a new feature, application updates and
the subsequent impact of new users on
session count and session duration.
Business
Analytics for
dabblr
product
Aditya
Saikiran
Manish
2. Page1
0. In-side the Report
Introductionā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦..2
Hypothesisā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦..9
Research Methodologyā¦ā¦.10
Data Setā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦.11
Statisticsā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦16
Findingsā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦.19
Recommendationsā¦ā¦ā¦ā¦ā¦.20
Referencesā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦ā¦21
3. Page2
1. INTRODUCTION
1.1 Analytics
Analytics is the discovery and communication of meaningful patterns in data. Especially
valuable in areas rich with recorded information, analytics relies on the simultaneous application
of statistics, computer programming and operations research to quantify performance. Analytics
often favors data visualization to communicate insight.
Firms may commonly apply analytics to business data, to describe, predict, and improve business
performance. Specifically, arenas within analytics include Predictive analytics, enterprise decision
management, retail analytics, store assortment and stock-keeping unit optimization, marketing
optimization and marketing mix modeling, web analytics, sales force sizing and optimization, price
and promotion modeling, predictive science, credit risk analysis, and fraud analytics. Since
analytics can require extensive computation , the algorithms and software used for analytics
harness the most current methods in computer science, statistics, and mathematics.
1.2 Analytics for Marketing Optimization
Marketing has evolved from a creative process into a highly data-driven process. Marketing
organizations use analytics to determine the outcomes of campaigns or efforts and to guide
decisions for investment and consumer targeting. Demographic studies, customer segmentation,
conjoint analysis and other techniques allow marketers to use large amounts of consumer
purchase, survey and panel data to understand and communicate marketing strategy.
Web analytics allows marketers to collect session-level information about interactions on a
website using an operation called sessionization. Those interactions provide the web analytics
information systems with the information to track the referrer, search keywords, IP address, and
activities of the visitor. With this information, a marketer can improve the marketing campaigns,
site creative content, and information architecture.
Analysis techniques frequently used in marketing include marketing mix modeling, pricing and
promotion analyses, sales force optimization, customer analytics e.g.: segmentation. Web
analytics and optimization of web sites and online campaigns now frequently work hand in hand
with the more traditional marketing analysis techniques. A focus on digital media has slightly
4. Page3 changed the vocabulary so that marketing mix modeling is commonly referred to as attribution
modeling in the digital or Marketing mix modeling context.
These tools and techniques support both strategic marketing decisions (such as how much
overall to spend on marketing and how to allocate budgets across a portfolio of brands and the
marketing mix) and more tactical campaign support in terms of targeting the best potential
customer with the optimal message in the most cost effective medium at the ideal time.
1.3 Analytics Spectrum and Analytics Tools
1.3.1 Reporting
The domain of Analytics starts from answering a simple question ā What happened? This
activity is typically known as reporting. These are typically the MIS which people want to receive
first thing before analyzing. It is a snapshot of what has happened. Following is an example of
how a typical report might look like:
Tools used in reporting:
Majority of elementary reporting happens on MS Excel. More evolved Organizations might pull
the data through databases using tools like SQL, MS Access or Oracle. But typically, the
dissemination of reports happens through Excel. Therefore, the tools used in reporting in a
snapshot are:
ā¢ MS excel
ā¢ MS Access
5. Page4 ā¢ SQL
ā¢ Oracle reporting tools
1.3.2 Detective Analysis
Detective Analysis starts where reporting ends. You start looking for reasons for unexpected
changes. Typical problems are āWhy did the Sales drop in last 2 months?ā or āWhy did the latest
campaign under-perform or over-perform?ā. In order to find out answers to these questions, you
look at past trends or you look at distribution changes to find out the reasons for the changes.
However, all of this is backward looking.
Some of these insights, which you find out after looking at backward analysis can be used for
business planning, but the purpose of analysis is typically to find out what has worked and what
has not.
Tools used in detective analysis:
Typically used tools are MS excel, MS Access, Minitab, R (basic regression). You tend to use
advanced Excel and Pivot tables while dealing with these problems and typically creating time
series graphs helps a lot.
1.3.3 Dashboards
Dashboard is an Organized and well presented summary of key business metrics. They are
usually interactive so that the user can find out the exact information he is looking for. Dashboard,
in ideal state should provide real time information about performance. Following is an example of
how a dashboard might look like:
The whole science of creating data model, dashboards and reports based on this data is also
known as āBusiness Intelligenceā.
6. Page5
Tools used for creating dashboards:
For limited size of data, dashboards can be made using Advanced excel. But, typically
Organizations use more advanced tools for creation and dissemination of tools. Business
Objects, Qlikview, Hyperion are names of some such softwares.
1.3.4 Predictive Modelling
This is where you take all your historical trends and information and apply it to predict the future.
You try and predict customer behaviour based on past information. Please note that there is a
fine difference in forecasting and predictive modeling. Forecasting is typically done at aggregate
level, where as predictive modeling is typically done at a customer / instance level
Tools used for Predictive modeling:
SAS has the highest market share among tools used for predictive modeling followed by SPSS,
R, Matlab.
1.3.5 Big Data
Imagine applying predictive modeling with a microscope in hand. What if you can store, analyze
and make sense out of every information about the customer. What kind of social media
community he is attached to? What kind of searches is he performing? Big data problems arise
when data has grown on all three Vs (Volume, Velocity and Variety). You need data scientists to
mine this data.
Tools used in Big data:
This is a very dynamic domain right now. Tools include Splunk, Skytree etc.These tools typically
work on Hadoop to store the data.
7. Page6
1.4 Google Analytics
Google Analytics is a service offered by Google that generates detailed statistics about
a website's traffic and traffic sources and measures conversions and sales. It's the most widely
used website statistics service.
The basic service is free of charge and a premium version is available for a fee.
Google Analytics can track visitors from all referrers, including search engines and social
networks, direct visits and referring sites. It also tracks display advertising, pay-per-
click networks, email marketing and digital collateral such as links within PDF documents.
Integrated with AdWords, users can now review online campaigns by tracking landing page
quality and conversions (goals). Goals might include sales, lead generation, viewing a specific
page, or downloading a particular file.
8. Page7 Google Analytics' approach is to show high-level, dashboard-type data for the casual user, and
more in-depth data further into the report set. Google Analytics analysis can identify poorly
performing pages with techniques such as funnel visualization, where visitors came from
(referrers), how long they stayed and their geographical position. It also provides more advanced
features, including custom visitor segmentation.
Google Analytics e-commerce reporting can track sales activity and performance. The e-
commerce reports shows a site's transactions, revenue, and many other commerce-related
metrics.
Google Analytics launched Real Time analytics as of September 29, 2011.
A user can have 50 site profiles. Each profile generally corresponds to one website. It is limited to
sites which have a traffic of fewer than 5 million pageviews per month (roughly 2 pageviews per
second), unless the site is linked to an AdWords campaign.
Google Analytics includes Google Website Optimizer, rebranded as Google Analytics Content
Experiments.
1.4.1 Why Google Analytics
ā¢ Google Analytics can be used to build audience by analyzing the userās activity in and
outside your pages. Google Analytics offers features to identify your audience through
features like App-Specific Metrics and Dimensions, Audience Data and Reporting and
also provides scope for Custom Metrics, aiding in scalable analytics.
ā¢ Google Analytics can be used to trace the customer path through features traffic sources
and visitor flow, which are key to identifying routes taken by the users, which help in
improving user experience.
ā¢ Google Analytics also provides features like Event Tracking, Flow Virtualization, In-page
analytics, Real-Time reporting and Site Search which aid in identifying what the users are
looking for and what they like, then tailor your content and marketing activities for
maximum impact.
9. Page8 1.4.2 How to Implement Google Analytics
Google Analytics is implemented with "page tags". A page tag, in this case called the Google
Analytics Tracking Code is a snippet of JavaScript code that the website owner adds to every
page of the website. The tracking code runs in the client browser when the client browses the
page (if JavaScript is enabled in the browser) and collects visitor data and sends it to a Google
data collection server as part of a request for a web beacon.
To set up the web tracking code:
ā¢ Find the tracking code snippet for your property. Sign in to your Google Analytics account, and
click Admin in the top menu bar. ...
ā¢ Find your tracking code snippet. ...
ā¢ Copy the snippet. ...
ā¢ Paste your snippet (unaltered, in it's entirety) into every web page you want to track.
The tracking code loads a larger JavaScript file from the Google webserver and then sets
variables with the user's account number. The larger file (currently known as ga.js) is typically
18 KB. The file does not usually have to be loaded, though, because of browser caching.
Assuming caching is enabled in the browser, it downloads ga.js only once at the start of the visit.
Furthermore, as all websites that implement Google Analytics with the ga.js code use the same
master file from Google, a browser that has previously visited any other website running Google
Analytics will already have the file cached on their machine.
10. Page9
2. Hypothesis
āMobile Application gathers more new users as its version
maturesā
āMobile Application has a sudden influx of new users with the
inclusion of just one new useful featureā
āMobile application can achieve greater hits and session count
per day by including features that generate new content dailyā
11. Page10
3. Research Methodology
The first step of the research involved collecting cumulative statistical data of diverse mobile
application dimensions and metrics over a period of eight months from January,1,2014 to
August,31,2014.
The collected data set was then divided into two separate time slices, each four months long as
TimePeriod1 from January,1,2014 to April,31,2014 and TimePeriod2 from May,1,2014 to
August,31,2014.
Data Set Name Time Period of Data Set
TimePeriod1 January,1,2014 - April,31,2014(4 months)
TimePeriod2 May,1,2014 - August,31,2014 (4 months)
The selection of the time periods followed the criteria that one of the time period has no
considerably new features implemented whereas the other time period has at-least one new
feature implemented which is permanently included in the application.
Time Period New Features Implemented
TimePeriod1 0
TimePeriod2 1-Timetable
From the collected metrics, key metrics were identified to evaluate new user pattern against the
stated hypothesis. The identified key metrics were classified as dimensions, which formed the
basis for the data set, consisting of statistics related to the dimensions in key metric areas,
collected using Google Analytics.
The collected data was then visualized using Google Analytics and the visualized data was used
to evaluate the hypothesis, verify the hypothesis and identify the reasons as to why the
hypothesis is true. In case the hypothesis failed, the reasons for failure were identified, inspected
and recommendations were suggested.
12. Page11
4. Data Set
Data Collection: The Data Set is un-sampled mobile application data collected over two periods
TimePeriod1 and TimePeriod2 using Google Analytics tool.
Data Accuracy: 70-90%
Key Dimensions and Metrics:
Dimensions Metrics
User
ga:userType
ga:sessionCount
ga:daysSinceLastSession
ga:userDefinedValue
ga:users
ga:newUsers
ga:percentNewSessions
Session ga:sessionDurationBucket
ga:visitLength
ga:sessions
ga:visits
ga:sessions
ga:visits
ga:bounces
ga:entranceBounceRate
ga:bounceRate
ga:visitBounceRate
ga:sessionDuration
ga:timeOnSite
ga:avgSessionDuration
ga:avgTimeOnSite
ga:hits
Audience
ga:userAgeBracket
ga:visitorAgeBracket
ga:userGender
ga:visitorGender
ga:interestOtherCategory
ga:interestAffinityCategory
ga:interestInMarketCategory
Event Tracking ga:eventCategory
ga:eventAction
ga:eventLabel
ga:totalEvents
ga:uniqueEvents
ga:eventValue
ga:avgEventValue
13. Page12 ga:sessionsWithEvent
ga:visitsWithEvent
ga:eventsPerSessionWithEvent
ga:eventsPerVisitWithEvent
Platform or Device
ga:browser
ga:browserVersion
ga:operatingSystem
ga:operatingSystemVersion
ga:isMobile
ga:isTablet
ga:mobileDeviceBranding
ga:mobileDeviceModel
ga:mobileInputSelector
ga:mobileDeviceInfo
ga:mobileDeviceMarketingName
ga:deviceCategory
Non Visualized Data Set:
New users vs App Versions:
App
Version Date Range Users
New
Users Sessions
Screen
Views
Screens
/
Session
Avg.
Session
Duration
2.1.0
May 1, 2014 - Aug 31,
2014 4,670 3,926 51,283 385,936 7.53 0:01:22
2.1.0
Jan 1, 2014 - Apr 30,
2014 0 0 0 0 0 0:00:00
2.0.9
May 1, 2014 - Aug 31,
2014 2,620 2,086 22,279 154,898 6.95 0:01:11
2.0.9
Jan 1, 2014 - Apr 30,
2014 0 0 0 0 0 0:00:00
2.0.8
May 1, 2014 - Aug 31,
2014 604 324 2,114 12,844 6.08 0:00:46
2.0.8
Jan 1, 2014 - Apr 30,
2014 262 186 718 4,658 6.49 0:01:00
2.0.7
May 1, 2014 - Aug 31,
2014 506 49 1,578 11,770 7.46 0:00:31
14. Page13
2.0.7
Jan 1, 2014 - Apr 30,
2014 2,549 1,931 13,643 107,822 7.9 0:00:55
2.0.6
May 1, 2014 - Aug 31,
2014 121 26 469 2,692 5.74 0:00:38
2.0.6
Jan 1, 2014 - Apr 30,
2014 2,994 2,488 22,584 230,625 10.21 0:01:09
2.0.4
May 1, 2014 - Aug 31,
2014 23 16 80 131 1.64 0:00:09
2.0.4
Jan 1, 2014 - Apr 30,
2014 1,380 1,199 6,215 60,429 9.72 0:01:11
2.0.3
May 1, 2014 - Aug 31,
2014 17 8 44 52 1.18 0:00:05
2.0.3
Jan 1, 2014 - Apr 30,
2014 1,163 953 4,954 56,445 11.39 0:01:17
2.0.5
May 1, 2014 - Aug 31,
2014 5 3 9 10 1.11 0:00:01
2.0.5
Jan 1, 2014 - Apr 30,
2014 588 388 3,430 33,179 9.67 0:01:16
2.0.2
May 1, 2014 - Aug 31,
2014 3 0 3 3 1 0:00:00
2.0.2
Jan 1, 2014 - Apr 30,
2014 30 8 92 1,188 12.91 0:00:57
2.0.0
May 1, 2014 - Aug 31,
2014 0 0 0 0 0 0:00:00
2.0.0
Jan 1, 2014 - Apr 30,
2014 4 0 11 124 11.27 0:01:10
May 1, 2014 - Aug 31,
2014 8,569 6,438 77,859 568,336 7.3 0:01:16
Jan 1, 2014 - Apr 30,
2014 8,989 7,167 51,792 495,806 9.57 0:01:07
New Users vs Features:
Total 2 3 1 0
App
Version Date Range
New
Users
New
Users
New
Users
New
Users
New
Users
2.1.0
May 1, 2014 - Aug 31,
2014 3,926 1,935 1,833 158 0
2.1.0
Jan 1, 2014 - Apr 30,
2014 0 0 0 0 0
2.0.9
May 1, 2014 - Aug 31,
2014 2,086 43 18 1,108 917
2.0.9 Jan 1, 2014 - Apr 30, 0 0 0 0 0
15. Page14 2014
2.0.8
May 1, 2014 - Aug 31,
2014 324 0 0 2 322
2.0.8
Jan 1, 2014 - Apr 30,
2014 186 0 186 0 0
2.0.7
May 1, 2014 - Aug 31,
2014 49 13 4 10 22
2.0.7
Jan 1, 2014 - Apr 30,
2014 1,931 758 1,173 0 0
2.0.6
May 1, 2014 - Aug 31,
2014 26 7 2 10 7
2.0.6
Jan 1, 2014 - Apr 30,
2014 2,488 238 30 2,148 72
May 1, 2014 - Aug 31,
2014 6,438 2,002 1,863 1,291 1,282
Jan 1, 2014 - Apr 30,
2014 7,167 1,007 1,412 2,194 2,554
New Users vs Sessions:
Session
Duration Date Range Sessions
Avg.
Session
Duration
Screens
/
Session
Goal
Conversion
Rate
0-10 seconds
May 1, 2014 - Aug 31,
2014 37,673 0:00:03 2.11 0.00%
0-10 seconds Jan 1, 2014 - Apr 30, 27,060 0:00:03 3.23 0.00%
16. Page15 2014
11-30 seconds
May 1, 2014 - Aug 31,
2014 10,801 0:00:19 4.71 0.00%
11-30 seconds
Jan 1, 2014 - Apr 30,
2014 7,162 0:00:18 8.82 0.00%
31-60 seconds
May 1, 2014 - Aug 31,
2014 7,889 0:00:44 6.46 0.00%
31-60 seconds
Jan 1, 2014 - Apr 30,
2014 4,882 0:00:44 10.2 0.00%
61-180 seconds
May 1, 2014 - Aug 31,
2014 12,669 0:01:47 10.93 0.00%
61-180 seconds
Jan 1, 2014 - Apr 30,
2014 7,562 0:01:46 15.53 0.00%
181-600
seconds
May 1, 2014 - Aug 31,
2014 7,237 0:05:08 23.6 99.96%
181-600
seconds
Jan 1, 2014 - Apr 30,
2014 4,279 0:05:12 29.67 100.00%
601-1800
seconds
May 1, 2014 - Aug 31,
2014 1,453 0:15:25 46.35 99.45%
601-1800
seconds
Jan 1, 2014 - Apr 30,
2014 778 0:15:33 57.21 100.00%
1801+ seconds
May 1, 2014 - Aug 31,
2014 137 0:46:32 74.76 91.24%
1801+ seconds
Jan 1, 2014 - Apr 30,
2014 69 0:39:16 94.84 100.00%
May 1, 2014 - Aug 31,
2014 77,859 0:01:16 7.3 11.31%
Jan 1, 2014 - Apr 30,
2014 51,792 0:01:07 9.57 9.90%
20. Page19
6. Findings
ā¢ Itās been found that, Mobile Application gathers more
new users as its version matures. As the functionality of
the app got better and better users got attracted to it
thus resulting in more traffic to the app.
ā¢ Itās been found out that, Mobile Application has a
sudden influx of new users with the inclusion of just one
new useful feature, āTimetableā. This feature resulted in
a lot of traffic and has is a favourite among students.
The spike in the number of hits during august duration,
when this new feature was implemented serves as a
proof.
ā¢ Itās been found that, Sessions lasting 10-30 seconds
have seen considerable rise due to addition of new
feature called timetable, which lets them see their daily
class schedules, other than that, itās been observed that
there is a decline in sessions, but increase in session
duration, which proves that app has been successful in
delivering user-specific content efficiently.
ā¢ Altogether, this research has been successful in
evaluating all of the stated hypothesis, the hypothesis
three has proven to be partially correct, with several
more factors effecting the end result whereas, the
hypothesis one and two have been proven to be true.
21. Page20
7. Recommendations
ā¢ It is recommended to add more features which engage users
over a longer session duration. Existing features which
require longer session duration such as learn have not been
used much, therefore it is recommended to push more learn
pages when the user initiates a session.
ā¢ New features which ensure continuous update of content
are encouraged. Existing functionalities like campus dabbles
should be pushed daily and it is recommended to provide
some incentives to students who consistently develop hit
garnering campus dabbles.
ā¢ New users are found to increase with app stability, so
continuous updates to the application to make it more
robust and useable should be continued.
22. Page21
6. References
6.1 Online References:
ā¢ Wikipedia
ā¢ Analytics Vidya
ā¢ IBM Mid Market Gartner research paper
ā¢ Gartnerās Business Analytics Framework paper
ā¢ Google Analytics Documentation
6.2 Books
ā¢ Web Analytics 2.0: The Art of Online Accountability and Science
of Customer Centricity by Avinash Kaushik
ā¢ Predictive Analysis by Eric Siegel
ā¢ Keeping Up with the Quants: Your Guide to Understanding and
Using Analytics by Thomas H. Davenport