Business Analytics: Definition and Types

Definition of Business Analytics

• Business Analytics involves collecting, organizing, analyzing and interpreting data.
• It generates insights to make well informed business decisions.

Types of Business Analytics

• Business analytics includes descriptive, diagnostic, predictive, and prescriptive analytics.

Descriptive Analytics

• It pulls trends from data, describing what happened or is happening.
• It answers the question: "What happened?"
• Descriptive analytics is the simplest type and the base for the other types.

Diagnostic Analytics

• It provides information about why trends occurred.
• In-depth understanding of issues helps find solutions.
• Diagnostic Analytics is useful for data-driven decision-making.
• The question it addresses is: "Why did this happen?"

Predictive Analytics

• Predictive analytics makes future trend or event predictions.
• The question it answers is: "What might happen in the future?"

Prescriptive Analytics

• It takes all possible factors into account and suggests actionable recommendations.
• It provides the answer to the question: "What should we do next?"

Data Ecosystem

• The Data Ecosystem refers to programming languages, packages, algorithms, cloud-computing services, and the general infrastructure used by corporations.
• These tools and systems help organizations collect, store, analyze, and leverage data.

Data Life Cycle

• The Data Life Cycle describes the path from initial data generation to the moment data is interpreted into actionable insights.
• The eight steps of a data life cycle are: generation, collection, processing, storage, management, analysis, visualization, and interpretation.

Data Analytics Model: I-M-P-A-C-T

• I - Identify the Questions: Help business partners identify the questions they need help answering, in a non-intrusive way. Be sure to set a clear expectation of the time and effort involved to get an answer.
• M - Master the Data: Analyst can use assemble, analyze and synthesize all information in answering the critical business question. They can create easy to comprehend visual presentations such as charts, graphs, and tables.
• P - Provide the Meaning: Articulate clear and concise interpretations of data and visuals in the context of the questions identified.
• A - Actionable Recommendation: Provide thoughtful business recommendations based on an interpretation of the data. You can tie a rough dollar figure to any revenue improvements or cost savings in your recommendations.
• C - Communicate Insights: Focus on a multi-pronged communication strategy that will get your insights out to as wide of an audience as possible. It can be an interactive tool, a reported webex, a thoughtful memo, etc.
• T - Track Outcomes: Set up a way to track the impact of your insights, making sure that there is future follow up with your business partners on the outcomes.

Population vs. Sample

• Population: The entire group of individuals or objects to study. For example, all the people living in a certain city, all students in a school, or all cars on a certain road.
• Sample: A subset of a population; a smaller group of individuals or objects selected from a larger population and used as a representation of the population, from which statistics are used to make inferences or predictions about the population.

The Science of Data - Variable Analysis

• Univariate Analysis: Examines a single variable at a time
• Bivariate Analysis: Looks at the relationship between two variables.
• Multivariate Analysis: Explores the relationships between three or more variables simultaneously.

Variable

• A variable is any property, characteristic, number, or quantity that can take on different values or vary from one instance to another.
• Variables can also refer to attributes like age, gender, income, expenditure, family size, country of birth, etc.

Qualitative Variable

• Qualitative variables talk about an attribute that's not numeric, such as gender, race, or color of eyes.
• Variables describing attributes that are not measurable is a qualitative variable (gender, hair, eye-color, religion, favorite movie).

Nominal Variable (Categorical)

• A nominal variable labels categories and has no numerical value or order.
• Nominal variables have two or more categories and can include sex, type of dwelling, or hair color.

Ordinal Variable (Categorical)

• An ordinal variable has a clear order to its possible values.
• Used to gain insights into the customer behavior (ex: educational level, economic status, job satisfaction, sleep quality).

Quantitative Variable

• Variables that can be measured and expressed as numbers (weight, age, height, shoe size, profit, the number of attendees at a webinar).

Discreet (Numerical)

• Discreet variables can only take a limited number of distinct values ​​within a given range. Examples include # of pets in the household, # of cousins, # of business sites, # of children in the family.

Continuous (Numerical)

• Continuous variables can assume any value within a specified range, such as the amount of time, height, time, age, temperature, and weight.

Sources of Data

• Sources of data are the locations from which the data comes.
• Essential to understand because they are important for conducting studies thoroughly and with impact.

Primary Data Sources

• Refer to the original data collected directly by researchers for their research purposes.
• Surveys and questionnaires are widely used data collection methods that gather information directly from respondents.

Direct Observations

• Consist of systematically watching and recording events or behaviors.

Interviews and Focus groups

• Qualitative data collected through interviews and focus groups gives an in-depth exploration of the respondent's opinions, beliefs, and experiences.

Secondary Data

• This involves data collected by another party for another specific purpose than the research at hand.

Published Literature

• Academic papers, books, and reports offer a foundation for research, and the ability to build upon existing knowledge.

Government Sources

• Government agencies that collect and maintain data on a wide range of topics make them available for public use.

Online Databases

• The internet has increased access to a wealth of data through online databases, repositories, and open initiatives.

Market Research Reports

• Market research companies conduct surveys and gather data to analyze consumer behavior, and industry insight.

Comparison Chart: Primary vs Secondary Data

• Primary data refers to first-hand data gathered by the researcher himself; secondary data is data that's be collected by someone else earlier.
• Primary data is real time; secondary data is past data.
• Gathering primary data is involved; getting secondary data is quick and easy.
• Sources of primary data: surveys, observations, personal interviews, questionnaires. Sources of secondary data are government publications, websites, books, journals, internal records, etc.
• Primary data is expensive to acquire; secondary data economical.
• Gathering primary data takes a long time; secondary data, a short time.
• Primary data is always specific to the researcher's needs; secondary data is may or may not be specific the researcher's needs.
• Availability of primary data in crude form; secondary data, in refined form.
• Primary data has more accuracy and reliability; secondary data comparatively less accurate and reliable.