Open Data Analytics(ODA)- Part 3
High Level Implementation Steps
- Analytics : “R” and Python
- Data Source : Sales, Marketing, Traffic and Customer
- Data Technology: HADOOP, SAP HANA, Oracle BDA & MS Big Data
In this article, we are analyzing the effectiveness of digital vs. traditional marketing as a use case to implement enterprise data science. To analyze the effectiveness of digital vs. traditional marketing we are using many data sets but the key data sets that’s been used in this solutions are [a] Customer, [b] Marketing, [c], Sales Volume, [d] Cost, [e] Revenue, [f] Capital, [g] Channel, [h] Traffic, [i] Sentiment, [j] Social Network , [k] Branding, and Others.
This is a proven use case, and we are NOT sharing every details of our solution. In case, If anyone interested to know more about our solution please call or mail me directly.
Competitive and Market Intelligence:
- What should be the strategy to drive digital marketing?
- How does digital marketing differs from our traditional marketing?
- What type of data we need to develop digital marketing intelligence?
- How to empower my marketing team with digital marketing tools and techniques?
- Are we left behind in digital marketing as compare to the rest of the world? – Industry benchmarking
- What are the leading practices to start the digital marketing journey
Our enterprise data science addresses all of the above business challenges, and we enable business for “DATA DRIVEN DIGITAL MARKETING”
Figure 1: Analyze The Effectiveness of Digital Marketing vs. Traditional Marketing Using Multiple Maxima.
Figure 2: Determine Combinations That Yield The Highest Conversion Rates Using Multivariate Distribution.
Figure 3: LY (Last Year) & TY (This Year) Sales Data (statistical) Analysis and Comparison Using Clustering (K-Means).
Figure 4: Probability That a Shop Receives a Total of “X” Customers in “Y” Hours.
Figure 5:Effectiveness of Digital Marketing vs. Traditional
Figure 6: Marketing Spend Benchmarking & Customer Data Profiling
Figure 7: Digital Humanities Connected by Influenced Brand.
Figure 8: Our Digital Innovation Lab – Big Data Lake. Our big data lake integrates data across technology, and interoperates data regardless of data format , structure, and source.
Our goal is to bring enterprise perspective to data science even though data science involves statistical and mathematical data analysis techniques. Further, make everyone to realize data sciences are driven from industry, enterprise, and business than just academic research, statistics and applied mathematics.
To show the steps to develop data sciences using statistical and mathematical techniques for industry, enterprise, and business.
To prove that enterprise data science should only be developed from business perspective without industry expertise it’s impossible to develop enterprise data science.
Important Note: This is our part three (3) of open data analytics (ODA) series. For continuity and better understanding, please make sure you read our part one(1) and part two(2) of ODA series.
Statistical and Mathematical Data Analysis for Industry, Enterprise, and Business Problems
Example: Multivariate with Clustering
Multivariate Data Analysis refers to statistical technique used to analyze data that arises from more than one variable. A multivariate data analysis equipped with powerful methods including Clustering (K-Means), Multivariate Curve Resolution (MCR), PLS regression, 3-Way PLS regression, etc. Multivariate data analysis Identifies the dominant patterns in data, such as groups, outliers, trends, and etc.
Business Usage of Multivariate Data Analysis:
What type of enterprise data science we can develop using Multivariate statistical techniques ?
Here is some proven examples of data science that we have developed using various statistical and mathematical techniques and data sources (including open data).
In our data science solution, we are addressing the following business challenges
- Hypothesis of changes that could impact your business such as revenue ,cost, volume, traffic, customer
- Customer segmentation
- Sales classification
- Determine combinations that yield the highest conversion rates
- Identify revenue per order
- Customers buying patterns and similarities (ex: Customers who is buying running shoes at the sports store often buy running track shoot at the same time)
- Predict the market size of a target market based on location, number of customer, sales size, and other factors
- Customers who are likely to increase spending if given an extra 5% discount
- Score the new customer profile data – new customer is somewhat of an anomaly
Statistical and Mathematical Techniques Used:
- Clustering (K-Means) (Statistical Technique for Data Analysis)
- Multivariate Distribution(Statistical Technique for Data Analysis)
- Multiple Maxima(Mathematical Technique for Data Analysis)
- Association (Ex:Customers who buy running shoes at the sports store often buy running track shoot at the same time)
- Regression (Ex:Predict the value of a target marketing based on location, number of customer, sales size, and other factors)
- Classification (Ex:Customers are likely to increase spending if given an extra 5% discount)
- Anomaly Detection(Ex: Score the new customer data – new customer is somewhat of an anomaly)
Disclaimer. This “PARTIAL” code block strictly shared for learning and researching. We have no liability on this particular code under any circumstances. Users should use this code on their own risk. All software, hardware and other products that are referenced in these materials belong to the respective vendor who developed or who owns this product. – This is not a production ready code.
We Are Global Leader in Solving Data Science Skill Gaps.
Our Community Service:
Our Goal: Simplify enterprise data science learning through proven use cases and integrated data science lab infrastructure.
- Our effort is to address global skill gap issues on data science. This is a free community service to help individuals to learn enterprise data science using various analytics techniques on big data.
- This is not a sales material and/or marketing tool kits to market on what we do for the data science learning community.
We are PRACTICAL:
- We bring only reality to public for on job learning experience
- We share our practical experience and lessons learned in our presentation
- We have on cloud integrated data science lab and our lab empowered by HADOOP, SAP HANA, Oracle BDA & Microsoft’s Big Data
- We have digital learning content
- We have proven industry use cases
- We have hands-on expertise to solve data science & big data problems from strategy to execution
- WE ARE PRACTICALand we bring only reality to our discussion. We share our practical experience and lessons learned in our presentation.
Our Value To Data Science Community:
As you may have realized, we are starting our posting with measurable business values by presenting real analytics. I am sure you may have seen 100’s of presentation on big data , analytics, data science and predictive analytics but have you seen one single presentation which explains practical process and steps that’s required to implement predictive analytics at an enterprise level?, and the answer is “NO”.
With us, now you have an opportunity to understand the steps to implement big data and also a true learning platform to learn to implement Predictive analytics.
Learn beyond concepts and presentation
- Jothi Periasamy
- Durga Prasad
- Karthikeyan Rajamanickam
- Uday Bhoomagoud
- Gourav Reddy
- Sankar Janakiraman
Learn the roots of data science through proven use cases and integrated data science lab.
Author, Speaker, Thought Leader & Community Contributor
We Approach Data Science From Enterprise Perspective. Just not form academic perspective.
Author Bio :
Jothi Periasamy is an Author, Speaker, Thought Leader & Community Contributor with Seventeen years of experience in management consulting, entrepreneurial and process excellence with Deloitte, E&Y, KPMG. Deeply “hands-on” on SAP HANA, HADOOP, & BI Tools .