1.7: Detailed Figure Captions - Chapter 1

Foundations of Data Analytics — Figure Specifications

Overview

• Chapter: 1 — Foundations of Data Analytics
• Total Figures Proposed: 8
• Figure Types Used: Conceptual Illustration, Flowchart, Comparison Table, Architecture Diagram, Infographic

---

Figure 1.1.1

Title: From Raw Data to Actionable Insight 

Detailed Caption: A light-gray "Raw Data" panel on the left holds six muted source tiles — Spreadsheet, Database, Text Docs, Sensors, Logs, and APIs — feeding into a long navy "DATA ANALYTICS" pill across the middle, decorated with three teal step circles labeled "Examine → Process → Visualize." On the right, a teal-bordered "Actionable Outputs" panel stacks three colored outcomes: a teal "Dashboards (Charts & KPIs)" tile, a warm-gold "Insights (Patterns & Trends)" tile, and a navy "Decisions (Actions & Strategy)" tile. The figure compresses the chapter's opening claim into one image — analytics is the transformation that converts disparate raw sources into a deliberately structured set of outputs, each tier serving a different audience from analysts (dashboards) up to executives (decisions).

Figure 1.2.1

Title: The Four Types of Analytics — Progression of Value and Complexity 

Detailed Caption: Four ascending stair-steps climb from lower-left to upper-right against a y-axis labeled "VALUE" and an x-axis labeled "COMPLEXITY." The bottom step, "Descriptive" (cool-gray), poses "What happened?" with the goal label Hindsight; step two, "Diagnostic" (light teal), poses "Why did it happen?" with Insight; step three, "Predictive" (deep teal), poses "What will happen?" with Foresight; and the top step, "Prescriptive" (warm gold), poses "What should we do?" with Action. The deliberate parallel — each box pairs a question with a one-word outcome — makes the staircase double as a quick mnemonic. The figure delivers the chapter's organizing framework: the four analytics types are not parallel choices but a maturity path, where each step's payoff requires the capabilities of the steps below it. 

Figure 1.2.1.2

Title: Visual Progression of Analytics Types (Staircase Diagram)

Detailed Caption: The diagram presents four levels of data analytics arranged as an ascending staircase, plotted against two axes: - Y-axis (vertical): VALUE — increases upward, indicating each successive analytics type delivers greater business value. - X-axis (horizontal): COMPLEXITY — increases to the right, indicating each successive type requires more sophisticated techniques and effort. The staircase metaphor illustrates that analytics maturity progresses from simply looking backward (descriptive/hindsight) through understanding causes (diagnostic/insight) and forecasting (predictive/foresight), culminating in actionable recommendations (prescriptive/action) —  with each step adding both more complexity and more business value.

Figure 1.2.2

Title: E-Commerce Analytics Scenario — Four Types Applied 

Detailed Caption: Four horizontal panels stack vertically, color-graded from navy at the top through teal in the middle to warm gold at the bottom, with downward arrows linking them. Panel 1, navy "Descriptive — What happened?", reports "Sales dropped 20% | Conversion rate fell from 3.2% to 2.1%" with the gloss "Dashboard shows declining revenue trend over past quarter." Panel 2, teal "Diagnostic — Why did it happen?", reports "Page load times 40% slower after redesign | Cart abandonment rose 68% → 78%" with "Root cause: website performance degradation after recent update." Panel 3, deeper teal "Predictive — What will happen?", projects "15–20% further sales decline | Customer satisfaction to drop 25%" with "If no action taken, continued revenue loss and customer churn expected." Panel 4, warm gold "Prescriptive — What should we do?", lists three actions (Rollback to previous website version, Deploy optimized version, Send targeted discounts) with the expected outcome "recover 85% of lost sales within 30 days." By threading a single e-commerce scenario through all four levels, the figure shows how each analytics type adds capability rather than replacing the previous one — and why a mature team needs all four working together, not just the most advanced. 

Figure 1.3.1

Title: The Data-to-Insight Lifecycle 

Detailed Caption: Seven rounded-rectangle stages are arranged in a horseshoe loop and color-graded from navy on the top-left through teal across the middle to warm gold at the upper-right return point. Reading clockwise: 1. Define Goals & Requirements (navy) → 2. Data Collection (navy) → 3. Storage & Management (deep teal) → 4. Cleaning & Prep (teal) → 5. Analysis & Modeling (teal) → 6. Visualization & Comm. (teal) → 7. Insight to Action (warm gold), with a long dashed gold "Iterate & Refine" arrow looping from Stage 7 back to Stage 1, and the central label "Iterative Process" in light gray. The deliberate color shift mirrors the value progression — engineering-heavy stages start in navy, analytical work cools through teal, and the final action stage glows in gold. Drawing the lifecycle as a closed loop rather than a linear pipeline communicates the chapter's argument: each round of analysis yields refined questions for the next, so a team that runs the cycle once and stops captures only a fraction of the value the framework offers. 

Figure 1.3.2

Title: MobileFlow Case Study — Lifecycle in Practice 

Detailed Caption: Seven numbered milestone tiles run left to right above a horizontal timeline, each carrying a short title and a one-line case detail: 1. Define Goals (navy, "Why is Q4 churn up 15%?"), 2. Collect Data (navy, "Billing, app logs, support tickets"), 3. Store & Manage (deep teal, "Cloud warehouse, nightly ETL"), 4. Clean & Prep (teal, "Fix 8% missing logs, standardize UTC"), 5. Analyze & Model (teal, "Social features predict churn"), 6. Visualize (teal, "2x/week social = 40% less churn"), and 7. Act (warm gold, "Engagement campaign, 30-day monitor"). A long dashed gold "Re-evaluate" arc curves from Stage 7 back under the timeline to Stage 1, completing the loop. The deliberately concrete language — exact churn percentages, specific data sources, named campaigns — turns the abstract lifecycle into a story students can follow, and demonstrates that analytics maturity is less about exotic techniques than about disciplined progression through every stage. 

Figure 1.4.1

Title: Key Roles in Data Analytics — Responsibilities and Focus Areas 

Detailed Caption: Three column headers — navy "Data Analyst," teal "Data Scientist," and warm-gold "Data Engineer" — sit above five row labels: Primary Focus, Key Skills, Typical Outputs, Analytics Types, and Data Types. Each cell carries a one- or two-line phrase; for example, Primary Focus maps Analyst to "Interpret & report," Scientist to "Model & predict," and Engineer to "Build & maintain infra," while Key Skills contrasts "SQL, Excel, visualization tools" with "Python/R, ML, statistics" and "ETL, cloud platforms, distributed systems." Analytics Types pins each role to its dominant quadrants — Descriptive & Diagnostic for Analysts, Predictive & Prescriptive for Scientists, "Enables all types" for Engineers — and Data Types ranges from Structured (Analyst) through Structured & Unstructured (Scientist) to All formats at scale (Engineer). The matrix collapses the chapter's role discussion into one scannable reference and makes the practical distinction immediate: analysts answer questions on existing data, scientists build models that predict, and engineers make both possible by building the data plumbing underneath. 

Figure 1.4.2

Title: How the Three Roles Collaborate in the Data Lifecycle 

Detailed Caption: A horizontal pipeline of seven lifecycle stages — Define Goals, Collection, Storage & Mgmt, Cleaning, Analysis & Model, Visualization, Insight to Action — runs across the top, color-graded from navy through teal to warm gold. Below the pipeline, three swim-lane bars represent the roles: a warm-gold Data Engineer bar spanning Collection through Cleaning ("Collection → Storage → Cleaning (pipelines, ETL, infrastructure)"), a navy Data Analyst bar spanning Cleaning through Visualization ("Cleaning → Analysis → Visualization (dashboards, reports, KPIs)"), and a teal Data Scientist bar spanning Cleaning through Analysis ("Cleaning → Analysis & Modeling (ML, experiments)"). The three bars overlap at the Cleaning stage, marked as the shared "Collaboration zone" in the center. The deliberate overlap is the figure's pedagogical anchor: rather than treating these roles as siloed specialties, the diagram shows that data cleaning is where engineering, analysis, and science converge — and that the cleanest division of labor across the rest of the lifecycle still requires everyone to agree on the shape and quality of the prepared data. 

Figure 1.1.2

Title: Data Analytics and Related Fields 

Detailed Caption: Three translucent circles overlap in a triangular Venn layout — navy "Business Intelligence" (Reporting, Dashboards, KPIs) on the upper left, teal "Data Science" (Machine Learning, Experimentation, Model Building) on the upper right, and warm-gold "Data Analytics" (Statistical Analysis, Pattern Discovery) at the bottom. Each pairwise overlap labels the shared territory: BI ∩ Analytics holds "Visualization & Trends," Analytics ∩ Data Science holds "Exploratory Analysis," and BI ∩ Data Science is annotated "Predictive Reporting." At the three-way intersection, the bold phrase "Data-Driven Decisions" sits as the unifying purpose. By placing each discipline's distinctive techniques in its own zone and naming the overlaps explicitly, the figure resolves the common confusion students face — Analytics, BI, and Data Science aren't synonyms, but they share enough territory that practitioners routinely move between them, and all three converge on the same goal of evidence-based decision-making.