Data Analytics Applied to a Marketing Problem need to use jmp

Final Assignment (Report) – Data Analytics Applied to a Marketing ProblemThis assignment requires your analysis of a data set taken from the case studies.
Your role is that of a data analyst at the company you have chosen. If there is
no company identified with the data set, then used a well-known company in
that industry space as your moot company. You work for the CMO at the
company and have been charged with bringing some answers and insights from
the data to company executives.
The heart of the assignment is demonstrating your data analytics capabilities
but as you well know this is not enough, you must tie your efforts to business
results. So, it is just as important that you frame the issues in business terms as
well. First you will develop a business context and do the business analysis,
which should lead to the use of the proper technical analytics tools and
techniques. Finally yielding a business answer you can make a credible
argument for to be embodied in some business policy or action.
Assignment
Use the Thinking with Data Framework
Four Part Model
1. Context
2. Need
3. Vision
4. Outcome
The CoNVO model is an analysis tool for the business case for analytics. It helps
plan and execute the technical analysis better and provides a path to a solution,
as well as a justification for the technical analysis.
Develop the Context
Your project has a context, the defining frame that is apart from other particular
problems we are interested in solving.
•
•
•
Select a moot organization in an industry interested in the problem you
are solving.
Who are the people in the organization with an interest in the results of
this project?
What are they generally trying to achieve?
•
What work, generally, is the project going to be furthering?
Discover the Information Need
Everyone faces challenges – things that, were they to be fixed or understood,
would advance the goals they want to reach.
•
•
•
•
What are the specific needs that could be fixed by intelligently using your
data?
Present these needs should in terms that are meaningful to the
organization.
Remember that even if we build a model, the need is not to build a model.
The need is to solve the problem that having the model will solve.
When we correctly explain a need, we are clearly laying out what it is that
could be improved by better knowledge.
Develop a Vision for a Solution
Before starting to acquire data, perform transformations, test ideas, there is
need for some vision of where we are going and what it might look like to
achieve the goal.
•
•
The vision is a glimpse of what it will look like to meet the need you’re
your data.
Make a mockup describing the intended results, or a sketch of the
argument that we’re going to make, or some particular questions that
narrowly focus our aims.
Imagine the Outcome of your Analysis
We need to understand how the work will actually make it back to the rest of
the organization and what will happen once it is there.
•
•
•
•
•
How will it be used?
How will it be integrated into the organization?
Who will own its integration?
Who will use it?
How will its success be measured?
If we don’t understand the intended use of what we produce, it is easy to get
lost in the weeds and end up making something that nobody will want or use.
•
What’s the purpose of all this work if it does nobody any good?
•
The outcome is distinct from the vision; the vision is focused on what
form the work will take at the end, while the outcome is focused on what
will happen when we are “done.”
Perform the Technical Analysis
Use the Analytical Framework:
•
•
•
•
•
Frame
Find
Prep
Analyze
Apply
This is where the technical results of your analysis are to be found.
Frame useful questions for analysis
Understand and define the customer question you are trying to answer based
on the business analysis above. These should come from your Vision statement
above. You must frame at least three questions that your analysis will
specifically answer and that satisfy the client needs posed above. At least one
of the framing questions should be addressable through hypothesis testing and
at least one other framing question should be addressable through advanced
analytical techniques such as linear or logistic regression, factor analysis, or
cluster analysis.
Find the datasets needed for the analysis
Describe what data sets and variables you used. Clearly, you used the data set
from the case study, but you may have to discover, profile, and acquire
additional data to help answer the question. Here there is only one data set, the
one provided with your case study.
Prep the data to ensure accuracy and useability
Clean, shape, and organize the data so that they are ready for analysis. Do not
assume your data files are error free or are already in the structure or format
needed.
Analyze
Use data analytics to get the insights and intelligence needed. This is the heavy
lifting of the technical analysis, all the statistical, predictive and data mining
techniques.
Here is where you will do the bulk of your work. Use at least three analytic
techniques such as hypothesis testing, regression, clustering, partitioning,
Pareto, etc. that meet the needs of your framing questions and are appropriate
to the data and analytical goals. Simple descriptive statistics do NOT
qualify! Be sure to use the analytical techniques correctly and in a way that
adequately satisfies any assumptions or requirements that may be applicable.
Apply
Apply intelligence to the question you are trying to answer. If you have some
special data visualization or regular report or a dashboard you are feeding with
your data, this is where you describe it.
Create at least three business rules based on the answers you
found. Remember that business rules are policies, instructions, or procedures
that specific parts of your company must follow under certain circumstances.
For example, Starbucks created a business rule for its loyalty card program that
card members will receive a free drink on their birthday. Another example might
be a company deciding to generate a weekly report on all prospective
customers who fit a certain demographic profile and distributing that report to
its sales staff. You are expected to propose at least three new business rules
that result from the insights you gained from your analysis.
Make an Argument
Conclude your paper with the final argument. This final argument is NOT about
the analysis part of your report. It is about justifying to your audience the
business rules/recommendations that you are proposing as a result of your
analysis and conclusions. For example, if your analysis leads you to propose a
specific new policy or business rule for retaining or attracting a certain segment
of customers, how would you justify that proposal to other people who may be
skeptical of it and prefer to do something else or nothing at all? Even if people
agree with your analysis, they may legitimately disagree with your proposed
solution(s). You must anticipate their most likely criticism in your report and
give an effective counterargument to them.
Elements of an Argument
1. Claims
2. Evidence
3. Justifications
4. Potential Rebuttals
Claims
Arguments are built around claims. Before hearing an argument, there are some
statements the audience would not endorse. After all the analyzing, mapping,
modeling, graphing, and final presentation of the results, we think they should
agree to these statements. A claim is a statement that could be reasonably
doubted but that we believe we can make a case for.
•
Make claims for your proposed business rules/recommendations.
Evidence
A key part of any argument is evidence. Claims do not demonstrate themselves.
Evidence is the introduction of facts into an argument. Evidence is not raw data.
Raw data needs to be transformed into something more compact, before it can
be part of an argument: a graph, a model, a sentence, a map.
Transformations make data intelligible, allowing raw data to be incorporated
into an argument. A transformation puts an interpretation on data by
highlighting things that we take to be essential.
•
Present your evidence to support your claims.
Justification
Justification of why this evidence should compel the audience to believe our
claim. We need a reason, some logical connection, to tie the evidence to the
claim.
•
Make your justifications, what are the reason that connects the evidence
to the claims.
Potential Rebuttals
There are always reasons why a justification won’t hold in a particular case,
even if it is sound in general. A rebuttal is the yes-but-what-if question that
naturally arises in any but the most self-evident arguments.
•
What are the possible rebuttals you should be prepared for?
Deliverable
Create a memo presenting of your findings using the template provided. Your
memo should be at least 6 pages single-spaced, and no more than 12 pages (not
including bibliography/references, end notes and any appendices). Quality
always counts more than quantity. You should write this document as if it were
a professional business memo that will be read by colleagues and senior
executives at a major company or other leading organization (such as a
nonprofit or government agency).
The main body of your memo should only include tables and data visualizations
that are clearly essential to the narrative of your report. Background tables and
data visualizations of less importance should be moved to an appendix.
I suggest you write a longer draft and edit it to required length. Be sure to work
on the grammar and spelling with someone from the writing center or some
other competent proofreader who is a native English speaker. At the very
minimum, you should use spellcheck and Grammarly.
Finally, be sure to upload a copy of your JMP file(s) with your work saved as
scripts in the data table. If you have trouble uploading a JMP file because of its
size, you can upload it to an online file storage account like Dropbox or Google
Drive which will let you generate a URL link that you can paste into the
appendix or a footnote in your report. If you don’t have an online file storage
account, you can sign up for a free one at this link: https://db.tt/1hJ9elJ3
Business Analytics and Data Visualization
Grading Rubric
Final Assignment ‐ Predictive Analytics Applied To A Marketing Problem ‐ Report
Name
Student Name(s):
What I’ll be looking for in your solution
Overall Presentation
A
Used the template provided
1
1
1
1
1
Well written. Shows good command of the language
Correct length
Shows correct grammar and spelling throughout
Written in way that is appropriate for the intended audience
Assigned Grade
B
C
D
F What I’ll be looking for in your solution
Outcome
How will the results be used?
How will the results be integrated into the organization?
Who will be responsible for its integration?
Who will use it?
How will its success be measured?
Executive Summary
A
1
1
1
1
1
Frame
1
1
1
1
1
1
I was impressed
Proposal is appropriate to the case
Correct length
Free of grammatical and spelling errors
Answered all major questions posed in the case study
Gives an understandable summary of the solution proposed
Posed at least three framing questions
…that satisfy the client’s needs that were identified previously
Find
Adequately described what data sets and variables you used
Identified the people with an interest in the results of this project
What are they generally trying to achieve?
Identified part of organiztion needing results?
What work, generally, is the project going to be furthering?
The appropriate analytical techniques were used correctly
Applied at least three techniques such as clustering, partitioning
1
1
1
1
1
Provided evidence of actually using these techniques
Included appropriate tables or graphs of business professional quality
Attached JMP file with correct variable classifications and saved work
At least three business rules were created
…based on the answers found
Are they meaningful to the organization?
Is it addressing a believable business problem for the organization?
Answers show a deep understanding of the case and real insights
1
1
1
1
Do these rules address the organization’s stated needs/problems?
Answers show a deep understanding of the case and real insights
Acceptable final set of arguments
Vision
Mockup describing the intended results, or a sketch of the argument
1
1
100
100
1
1
Level of Work
Is this graduate‐level (A or B) or undergraduate‐level (C) work?
Student demonstrates mature analytical skills and reasoning
Points
Assignment Grade
1
1
1
1
Recommend
Acceptable proposal/recommendations
What does it look like to meet the organization’s needs with your data?
5
5
2
2
1
Apply
Need
What are the specific needs?
1
Extract
Context
Selected a relevant organization in the industry covered by the case study
1
1
85
0
Total points
70
0
200
Late? Score X .9
180
60
0
0
0
1
1
Assigned Grade
B C D F
RE: Proposal for Analyzing a Predictive Analytics Business Case
Context
● We selected the education industry as our focus area and chose NYU SPS for
this case. We think the admissions office under NYU SPS needs this result. Our
result may help NYU SPS to better understand their students, and to consider
what elements they should focus to promote.
● Also, based on the result, NYU SPS can better analyze which factors will be the
reason why students decide whether to admit to or not, and which factors are
student heavily concerned. Moreover, they can know what types of students will
be interested in the NYU SPS program, and what points they target students
care more about, so they know what are some elements that should be on the
page.
● Page editors and admissions officer may be interested in this project because
they can know what elements they should put on the school information page,
and when their customers ask for advice, they can plan ahead, what kind of
questions they may ask, and what elements customers will care about, so they
can give a better advice to them.
● Our project analyzes what factors will influence students’ admission decisions,
what factors have the most influence, and what factors have the least influence.
So when in the application season, universities will know what points they should
focus on to promote. Also, we can help the NYU SPS admission office to adjust
their admission offer decision, and to increase their enrollment rate.
Need
● We can use regression analysis, correlation analysis and hypothesis testing to
find out which of the 20 factors such as student level, GPA, application date and
ethnicity have a significant effect on the number of students admitted.
● The organization needs to understand how to help the NYU SPS improve its
competitiveness, compare and contrast the students who were accepted and
enrolled with those who were accepted but did not enroll, and analyze what
factors influenced their final decision.
● We can analyze the data set to find out which high schools, regions, religions
and ethnicities have high acceptance rates and whether they will enroll. It is very
useful information for the NYU Admissions office to target students who have
high acceptance rates but will not end up enrolling and address their concerns.
Vision
Before starting to acquire data, perform transformations, test ideas, there is a need
for some vision of where we are going and what it might look like to achieve the
goal.
The vision is a glimpse of what it will look like to meet the need for your data
● Our vision is aiming to utilize the dataset abstracting actionable information from
students that are already enrolled at the university and the group that doesn’t
enroll as first-edge reference to form unique advantages, enabling admission
officers to achieve the goal more accurately and effectively
Make a mockup describing the intended results, or a sketch of the argument that
we’re going to make, or some particular questions that narrowly focus our aims
● For students who applied, admitted and enrolled successfully into the university,
to figure out their high school attended, and academic achievements such as
overall GPA, SAT and ACT scores. That will be helpful for admissions officers
supposed to head the fair event for attracting more qualified and potential lead
generations.
● For students who applied, admitted but didn’t enroll eventually into the university,
to evaluate their ACRK Index, residency description and financial aid request.
More hypotheses to test if there is any relationship between financial problems
and willingness to enroll.
Framed Questions
Use the SMART technique of parsing the information needs and creating good
analytics questions that meet the SMART criteria.
1- Make SPECIFIC goal and write it down
●
Increasing numbers of qualified and potential lead generations from secondary
school
● Help some groups of students who didn’t go to university providing multiple
financial related solutions
2 – Define your MEASURES
● The number of applicants with higher GPA (range between 3.0~3.8)
● More research project or any useful solutions for students
● Conversion rate
5 – Define your TIMING
A goal must have a deadline.This will provide the necessary focus and sense of
urgency to make it happen. For example, if some of them are early birds, applied
applicant (i.e. Early Action applies before Nov. 15; Regular applies before February 1)
rolling process.
Outcome
● After we define all our metrics, we can have a brief idea of what GPA students
are better suited to apply to our program, whether the application date affects a
student’s probability of admission, what happens to students after they receive an
offer, etc.
● Our main purpose is to help Admission officers know their target students, so the
final analysis of GPA, test scores, potential acceptance rate, and many other
related data will be considered metrics when the admission officer chooses
candidates.
● NYU SPS will own its integration, and the Admission Officer in NYU SPS will use
it. We can track the acceptance rate, enrollment rate of students that we send
offers to see if we help the admission officer.to improve admission efficiency.