BUSINESS

ANALYTICS
Supervised By:
Prof. Esraa A. Raslan
 TEAM MEMBERS

• Omar SirajEldin Sayed

• Nazer Radwan Radwan

• Yosef Gerges Abdullah

• Tarek Mohamed Yosef

• Adham Nasser Ali

• Ahmed Ismail Sayed

 01 02
Business Analytics Key Roles for Successful

Overview Problem Approach Busines Analytics Project

Business 03 04
Analytics
Business Analytics
Business Analytics Plan
Lifecycle Lifecycle Phases

05 06
Key Outputs from Successful Core Deliverables to Meet
BA Project, by Role Most Stakeholders Needs
 INTRODUCTION
• M any probl em s seem huge at f i rst, but a w el l -def ined
process enabl es you to break dow n com pl ex probl em s
i nto smal l er steps that you can more easi l y address.
• Busi ness anal yti cs probl em s w i l l ensure that you
establ i sh a com prehensi ve, repeatabl e m ethod f or
conducti ng anal ysi s.
• dem onstrate your f i ndi ngs, w hi ch w i l l provi de
addi ti onal credi bi l i ty to the proj ect w hen you share
your f i ndi ngs.
• I t al so enabl es you to teach others to adopt the
m ethods and anal ysi s so i t can be repeated next
quarter, next year, or by new hi res.
 BUSINESS ANALYTICS
01 PROBLEM APPROACH

How to approach your business analytics problems?

• How do you currently approach your analytics problems?

• Do you follow a methodology or some kind of framework?
• How do you plan for an analytic project?
 BUSINESS ANALYTICS
01 PROBLEM APPROACH

Value of using business analytics lifecycle

• Focus your time.

• Ensure rigor and completeness.
• Enable better transition to members of the cross- functional
analytic teams.
• Repeatable.
• Scale to additional analysts.
• Support validity of findings.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

01 Business User
Someone who benefits from the end results
and can consult and advise project team.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

02 Project Sponsor
Person responsible for the beginning of the
project, providing core business problem, and
generally provides the funding and will
measure the degree of value from the final
outputs of the working team.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

03 Project Manager
Ensure key milestones and objectives are met
on time and at expected quality.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

04 Business Intelligence Analyst

Business domain expertise with deep

understanding of the data, KPIs, key metrics
and business intelligence from a reporting
perspective.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

05 Data Engineer
Deep technical skills to assist with tuning SQL
queries for data management, extraction and
support data ingest to analytic sandbox.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

06 Database Administrator (DBA)

Database Administrator who provisions and

configures database environment to support
the analytical needs of the working team.
 KEY ROLES FOR SUCCESSFUL
02 BUSINESS ANALYTIC PROJECT

07 Data Scientist
Provide subject matter expertise for analytical
techniques, data modeling, applying valid
analytical techniques to given business
problems and ensuring overall analytical
objectives are met.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

• There are six phases in the business analytics

lifecycle.
• Work on a project can be done in several
phases at once.
• Movement from any phase to another and
back again to previous phases occurs
throughout the Lifecycle.
• For this reason, the graphic is shown as a
cycle and the circular arrows are intended to
convey that you can move iteratively
between phases until you have sufficient
information to continue moving forward.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

01 Discovery

Determination of domain knowledge needed and

• Learn the Business Domain
business analytics problem type.

Solve the problem with learn from past mistakes,

• Learn from the Past
identify the reasons for failure.

Support the project in term of people, technology,

• Assess the resources
time, data, and teamwork.

State the analytics problem, identify stakeholders,

• Frame the business problem
objectives, and criteria of success and failure.

Gather and assess hypotheses from stakeholders,

• Formulate initial hypotheses (IH)
and preliminary data exploration with them.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

01 Discovery
• Analytics Sponsor Interview Tips:
• Prepare for the interview.
• Use open-ended questions.
• Don't fill every silence - give them time to think.
• Use active listening.
• Be mindful of your body language and theirs.
• Document what you heard and review it back with the sponsor.

• Frame an analytics problem:

• Identify the initial hypotheses (IH).
• Identify data will need to test (IH).
• Identify data dependencies will have.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

02 Data Preparation
• Prepare work environment in which you can
Determine the work space for the analytic team.
work for the duration of the project

• Perform ELT and begin transforming the data

Determine needed transformations.
so you can work with it and analyze it

• Assess data quality and structuring Determine and establish data connections for raw data.

• Derive statistically useful measures Execute Big ELT and/or Big ETL.

Data transformation & clustering: SQL, Hadoop,

MapReduce, Alpine Miner.
• Useful tools for this phase
Next phase: assistant from IT and DBAs.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

02 Data Preparation

• Familiarize yourself with the data thoroughly • List your data sources.
and take steps to condition the data • What's needed v.s what's available.

• Clean and normalize data.

• Data Conditioning
• Discern what you keep v.s what you discard.

• Overview, zoom & filter, details-on-demand.

• Survey & Visualize • Descriptive Statistics.
• Data Quality.

Descriptive Statistics on candidate variables for

• Useful tools for this phase
diagnostics & quality, and Visualization.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

03 Model Planning

• Select methods based on hypotheses, data structure and

• Determine the methods, techniques
volume.
and workflow you intend to follow for
• Ensure techniques and approach will meet business
the model scoring
objectives.

• Candidate tests and sequence.

• Techniques & Workflow
• Identify and document modeling assumptions.

R/PostgresSQL, SQL Analytics, Alpine Miner,

• Useful Tools for this phase
SAS/ACCESS, SPSS/OBDC.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

03 Model Planning
• Variable Selection:
• Inputs from stakeholders and domain experts.
• Capture essence of the predictors, leverage a technique for dimensionality
reduction.
• Iterative testing to confirm the most significant variables.

• Model Selection:
• Conversion to SQL or database language for best performance.
• Choose technique based on the end goal.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

• sample analytical methods used

across multiple market segments
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES
CONCEPT 1
Develop data sets for testing, training, and production
purposes.

04
• Need sufficiently model data.
• Need smaller test set & training set.

CONCEPT 2

Model Building
Get the best environment you can for executing models
and workflows, including fast hardware and parallel
processing.

USEFUL TOOLS

R, PL/R, SQL, Alpine Miner, SAS Enterprise Miner

 BUSINESS ANALYTICS
03 LIFECYCLE PHASES
CONCEPT 1
Determine if you succeeded or failed, based on the
criteria you developed in the Discovery phase, in

05 collaboration with your stakeholders.

CONCEPT 2

Communicate Results Identify your key findings, quantify the business value.

CONCEPT 3
Develop a narrative to summarize your findings
and convey to stakeholders.
 BUSINESS ANALYTICS
03 LIFECYCLE PHASES

Deliver final reports, briefings, code, and technical documents.

06 Run a pilot project.

Implement your models in a production environment.

Operationalize
Assess the benefits.

Provide final deliverables.

Define process to update, vetrain, and retire the model, as needed.

 Discovery Business How do we identify churn/no churn for
Problem Framed a customer?

Transaction volume and type are key

Initial Hypotheses
04 predictors of churn rates.

Components of Data 5 months of customer account history.

Business
Analytics Plan Model Planning
Analytic Teqnique
Logistic regression to identify most influential
factors predicting churn.

Result & Key Once customers stop using their accounts

for gas and groceries, they will soon erode
Findings
their accounts and churn.

If we can target customers who are high-

Business Impact risk for churn, we can reduce customer
attrition by 25%.
 Key Outputs from
05 Successful BA
Project, by Role
• In a successful Business Analysis (BA) project, key
outputs are produced by different roles within the
project, contributing to its overall success.

• Each role is responsible for specific deliverables that

together ensure the project is aligned with business
goals, stays on track, and is completed successfully.
 Core Deliverables to Meet
06 Most Stakeholder Needs
1. Presentation for Project Sponsors.
• Present big picture.
• Determine key messages.
• Focus on clean, easy visuals.
2 . Presentation for Analysts.
• Business process changes.
• Reporting changes.
• details of technical graphs.
3. Code for technical people.

4 . Technical specs of implementing the code.

THANK YOU!