Getting Started With Data Analytics A Primer on Transforming Data into Action MAY 25, 2017
Getting Started With Data Analytics
A Primer on Transforming Data into Action
MAY 25, 2017
What you will/will not get from today
The Cadence Group : Data Analytics
Won’t Get:
• PreCrime from “Minority Report”
• Magic ability to “sprinkle some of that big data on it” and get
“the right answers”
Will Get:
• A roadmap to developing an analytics program
• Three, brief case studies to learn from & build on
• CPE, Donuts, Bagels!
Discussion Topics
The Cadence Group : Data Analytics
• Big Data, Data Science, Analytics, Business Intelligence, Data Visualization –
what are we talking about?
• Challenges in building an analytics program
• The Cadence Analytical Framework – How we approach analytic projects
• Three case studies: gross margin analysis, mileage reimbursement, &
inventory holding costs
• Closing comments & questions
Here’s What We’re Talking About
The Cadence Group : Data Analytics
• Big Data – This is EVERYTHING (e.g. Facebook likes, Amazon
clicks, phone calls, credit card purchases, etc.)
• Data Science – Disaggregated data used in an exploratory way
• Analytics/Business Intelligence – interchangeable terms; take
raw, structured data and draw conclusions from it
• Data Visualization – taking data and telling an insightful story
Challenges Faced In Building an Analytics Program
The Cadence Group : Data Analytics
Bloomberg Businessweek Research Report – “The Current State of Business Analytics: Where Do We Go From Here?”
The Cadence Analytical Framework
Our Approach to Analytics
• Identify the Problem – What is the
question we’re trying to answer? Issue at
hand?
• Data Exploration – What can we get our
hands on? What shape is it in?
• Build the Model – What do we want to
measure? What are the
calculations/assumptions behind it?
• Evaluate & Present Results – How is it
to be presented? What is the
visualization? What happened and why?
The Cadence Group : Data Analytics
Identify the Problem
Data Exploration
Build the Model
Evaluate & Present Results
Case Study #1Gross Margin Analysis
• Identify the Problem – Company XYZ
approached with the question “We have an
accessories business, have no clue what’s
selling, what our margin is, can you help?”
• Data Exploration – Able to get our hands on
three data sets; sales, cost of goods sold, and
a child/parent hierarchy. Built out an Extract,
Transform, Load (ETL) process to merge into
a single data set.
• Build the Model – Developed an automated
process to calculate gross margins within the
source data.
• Evaluate and Present Results – Visualization
of how each accessory line’s margin
fluctuated quarter of quarter.
The Cadence Group : Data Analytics
Case Study #2Mileage Reimbursement
• Identify the Problem – Company ABC
approached with the question “It feels that
our mileage reimbursement requests have
been high, what do you think?”
• Data Exploration – Data point around
employees who had been assigned a car and
data point around employees who had
submitted a mileage reimbursement request.
• Build the Model – Who are the employees
who have been assigned a company car AND
submitted a mileage reimbursement request?
• Evaluate and Present Results – More text
than a visualization. Question behind the
data, why did these individuals submit a
request? Was their company car in the shop?
The Cadence Group : Data Analytics
Case Study #3Inventory Holding Costs
• Identify the Problem – Company XYZ
approached with the question “We feel like
there’s slow moving inventory which drives
up our holding costs, what are the outliers?”
• Data Exploration – Could not get inventory
levels/costs but were able to receive sales
numbers by SKU then filtered by color.
• Build the Model – Initially, the structure was
less defined but after data were presented,
we were able to draw conclusions based on
outlay.
• Evaluate and Present Results – Three
distinct tiers of product movement.
Suggestion was to have >40 on hand at all
times, 20 – 40 on two week notice, and <20
on an “an order” basis. Reduce the slow
moving colors in the warehouse, thereby
reduce holding costs.
The Cadence Group : Data Analytics
Closing Comments & Questions
The Cadence Group : Data Analytics
• Start with a high priority, high return project
• Find an ally on the business side, build the culture
• Have a good handle on the data
• Understand the story behind the numbers
• You’re opening the kimono, exercise restraint and due diligence
www.theCadenceGroup.com
http://www.linkedin.com/company/the-cadence-group
503.936.3276