Chapter 4: Analyze the Data

Vocabulary

Exploratory business analytics: Initial descriptive and diagnostic business analytics that analysts use to summarize and explain performance

Confirmatory business analytics: Predictive and prescriptive analytics that use statistics to judge the likelihood of a future event or outcome occurring

Human Resource Management System (HRMS): Information system for managing all interactions with employees

Customer Relationship Management System (CRM): Information system for managing all interactions with past, current, and potential customers

Supply Chain: Sequence of processes to move a product from raw materials to final customer delivery

Supply Chain Management System (SCM): System used to track supply chain processes

Financial Reporting Systems (FRS): Accounting systems which capture and measure financial transactions and communicate financial performance to interested parties

Financial statements: Collection of reports that communicate a company’s financial results, condition, health, and cash flows

Balance sheet: Summarizes a company’s assets and how much are financed at a certain point in time

Income statement: Summarizes how profitable the company has become een over a period of time

Descriptive Statistics: Summary statistics that briefly summarize features of a data set or variable

Horizontal Analysis: Identifies changes in various line items of the income statement, balance sheet, or statement of cash flows across time

Vertical Analysis: Expresses financial information in relation to some relevant base

Diagnostic Analytics: Investigate the underlying reasons for past results that cannot be answered by simply looking at the descriptive data

Outlier: Observation that lies outside its expected distribution

Management by Exception: Management technique in which managers attend to specific aspects of the business that depart from expectations

Concepts

Data for Descriptive Statistics comes from:

• Human Resource Management Systems

  • Data includes:

    • Recruiting data and leads

    • Training

    • Payroll and compensation

    • Benefits

    • Annual reviews

    • Absenteeism

    • Career Progression

    • Satisfaction and Sentiment

  • Privacy is highly valued

• Customer Relationship Management Systems

  • Data includes:

    • Contact history

    • Order history

    • Level of trade discounts or payment terms

    • Credit score

    • Credit limit

    • Payment history

  • Privacy is highly valued

• Supply Chain Management Systems

  • Data includes:

    • Active vendors

    • Orders placed

    • Demand schedules for which components of the final product are needed

    • Transportation status and schedules

    • Current location of the product in the supply chain

• Financial Reporting Systems

Descriptive Analytics Techniques

• Counts: Show how frequently an event or process occurs

• Total, sums, averages, subtotals, cross-tabulations: Summarize measures of performance

• Minimums, maximums, medians, standard deviations: Summarize measures showing extreme values to help explain what happened

• Graphs, histograms, and other visualizations: Summarize visually how often or how much there is of an object or event in comparison to other values

• Vertical analysis and horizontal analysis of financial statements

• Ratio analyses such as return on assets, return on sales, asset turnover ratios, or debt-to-equity ratios

Diagnostic Analytics

• Identifying anomalies and outliers

  • Important to start by establishing the norm, then investigate to understand why they occurred

  • Can be mistakes, truth, or fraud

  • Techniques

    • Cash/Bank Reconciliation: Reconcile bank acct with financial records

    • Benford’s Law: Principle that in any large, randomly produced set of natural numbers, there is an expected distribution of the first or leading digit.

      • Smaller values occur more frequently than larger ones

      • $P\left(d\right)=\log_{10}\left(1+\frac{1}{d}\right)$

    • Duplicates

    • Fuzzy Matching: Potential equivalents without exact fit

    • Sequence Check: Technique to determine if key number field in record is in the correct order

• Finding previously unknown linkages, patterns, or relationships between variables

  • Performing drill-down detailed analytics: Evaluates a greater level of detail in the data to gain insight

  • Performing statistical analyses

    • Techniques

      • Correlation: Extent to which variables are related to each other

      • Regression: Assessment of how a specific outcome is related to specific inputs

      • Hypothesis Testing: Statistical test of an assumption or theory