Get comfortable navigating a workbook before touching real analysis work. Learn to enter, organize, and verify healthcare records inside Excel.

• Workbook orientation and navigation
• Entering and organizing healthcare records
• Formulas, tables, and filters
• Verification habits appropriate to the task

Build the calculation and lookup skills the rest of the course assumes you already have. Practice pivot tables and simple decision checks on healthcare data.

• Calculated fields and structured tables
• Lookups, joins, and reusable formulas
• Pivot tables and decision checks

Connect a prepared healthcare dataset and build your first Tableau view from scratch. Save and verify your work inside Tableau Public.

• Connect a prepared healthcare dataset
• Dimensions, measures, and shelves
• Build a clear first view
• Save and verify in Tableau Public

Move from a first view to a dashboard built to support an actual decision. Practice choosing the right chart for the question in front of you.

• Connect a prepared healthcare dataset
• Dimensions, measures, and chart choice
• Build and verify a clear view
• Use dashboards to support a decision

Discover how operational questions turn into decision contexts across the analytics maturity curve, from descriptive to prescriptive. Learn the Orient, Explore, Translate framework that runs through every class.

• Operational questions and decision contexts
• Analytics maturity: descriptive through prescriptive
• Orient Your Thinking, Explore and Understand, Translate to Action
• Evidence, limits, and verification before action

Map a real healthcare workflow before you ever open a dataset. Trace how a question connects to a unit, a table, and a row.

• Map a healthcare workflow before analyzing data
• Connect question to unit to table to row
• Encounters, episodes, orders, and procedures
• Grain, keys, and workflow-driven data quality
• AI tests assumptions; records verify them

Define a metric in plain language before translating it into Excel-ready logic. Practice catching edge cases before a measure ever gets published.

• Define population, event, time, and exclusions
• Counts, rates, denominators, and targets
• Build a measure specification in plain language
• Translate the measure into Excel-ready logic
• Verify edge cases before publishing a metric

Separate genuine data-quality problems from real operational performance shifts. Profile missingness, duplicates, and impossible values, then document what you find.

• Completeness, validity, consistency, and timeliness
• Profile missingness, duplicates, and impossible values
• Separate data quality from operational performance
• Document provenance and transformation steps
• Use Excel checks and reconciliation
• Red-team AI suggestions against the source

Assignment #1: Is the Lactate Measure Ready?
Submit one completed, checked Excel analyst workbook.

Choose the summary statistic that actually matches the operational question in front of you. Build and verify a decision-ready Tableau view around it.

• Choose summaries that match the operational question
• Counts, rates, medians, percentiles, and distributions
• Segment without hiding important variation
• Build and verify a decision-ready Tableau view

Frame a group comparison before running it, then judge whether the difference is big enough to matter. Practice stating exactly what the evidence does and doesn't support.

• Frame the comparison and expected difference
• Compare groups with absolute and relative differences
• Judge practical importance alongside uncertainty
• State what the evidence does and does not support

Tell common-cause variation apart from special-cause variation before jumping to a driver. Build an explanation that survives a verification pass.

• Distinguish common-cause from special-cause variation
• Explore drivers without implying causation
• Use stratification to locate operational variation
• Build an explanation that survives verification
• Identify the next evidence needed

Set up a control chart with a real centerline and decision rules, not just a trend line. Practice spotting shifts and outliers before acting on them.

• Choose a stable measure and time grain
• Plot performance over time
• Establish centerline and decision rules
• Detect shifts, trends, and outliers
• Annotate operational changes
• Verify the signal before action

Assignment #2: Build the Winter Readiness View
Submit Tableau dashboard evidence plus a completed Excel verification planner.

Define a prediction target and decision window before trusting any model output. Evaluate predictions against a simple baseline instead of taking accuracy at face value.

• Define the prediction target and decision window
• Separate useful signal from leakage
• Evaluate predictions against a simple baseline

Turn a supported pattern into a small, safe test rather than a permanent policy overnight. Set your own stop, adjust, and scale rules in advance.

• Turn a supported pattern into a testable change
• Define aim, intervention, prediction, and measures
• Design a small, safe test of change
• Set stop, adjust, and scale rules

Lead with the decision instead of the chart, and match your evidence depth to the audience in the room. Practice stating your recommendation, uncertainty, and next action clearly.

• Lead with the decision, not the chart
• Match evidence depth to the audience
• Use visual hierarchy and annotations
• State recommendation, uncertainty, and next action

Treat AI as a thought partner rather than an authority on your data. Practice detecting hallucination, bias, and overreach before documenting your own human review.

• Use AI as a thought partner, not an authority
• Protect privacy and avoid sensitive data
• Verify calculations, citations, and assumptions
• Detect hallucination, bias, and overreach
• Document responsible human review

Pull question, workflow, measure, analysis, and action into one evidence chain for the first time. Combine a Tableau view with an Excel appendix and stress-test its limits.

• Integrate question, workflow, measure, analysis, and action
• Build the surgical growth decision evidence chain
• Combine Tableau communication with an Excel appendix
• Test recommendation sensitivity and limits
• Prepare a monitored decision proposal

Assignment #3: Defend the Surgical Growth Decision
Build a Tableau decision product with an Excel evidence appendix and a monitored recommendation. (Final cornerstone, 60 points.)

Defend your recommendation live under real questioning about evidence, limits, and trade-offs. Close the loop by defining ownership, cadence, and escalation for what happens next.

• Defend the recommendation under questioning
• Explain evidence, limits, and trade-offs
• Define ownership, cadence, and escalation
• Close the loop from analysis to action