Excelligence · Learning Paths

Two paths.
One governed graph.

Curated entry sequences through the knowledge graph. Each step connects to the next via governed LEADS_TO edges. Click a path to begin. Click any step to expand it.

Path 01 · Beginner
Your First Hour
Five foundational entries every Excel user needs before anything else. No prerequisites. Start here.
5 entries Beginner → Intermediate KEY · CON · FRM
Path 02 · Advanced
The Star Schema Path
The DDL methodology applied to Excel. From grain definition to institutional-grade matrix architecture. This is how the Assurance Map was built.
5 entries Beginner → Expert CON · PTN · ARC
Path 01 · Your First Hour
Five entries. One solid foundation.
These are the five things every Excel user should know before learning anything else. Each one is a stone. Together they build the floor everything else stands on.
1
KEY-0001 KEY Beginner

Ctrl+T (Table Conversion)

Converts a data range into a structured Excel Table, enabling auto-expanding ranges, structured references, built-in filtering, consistent formatting, and Power Query readiness.

▾
Intent
When you need structured references, auto-expansion, and Power Query readiness, use Ctrl+T to convert a range into a proper Excel Table.
Example
Select any cell in your data range → Ctrl+T → Confirm range → Rename table in Table Design tab. Result: =SUM(Sales[Revenue]) instead of =SUM(B2:B100)
Every data range in every DDL workbook must be a Table. This is the single most impactful Excel action. The foundation everything else builds on.
Explore in graph →
2
CON-0004 CON Beginner

No Blank Rows or Columns

Ensures that data ranges contain no empty rows or columns within the data boundary, preventing silent failures in tables, filters, sorts, pivot tables, and range detection.

▾
Intent
When you structure any data range, use continuous rows and columns with no internal blanks to ensure Excel correctly detects the data boundary for all operations.
Example
WRONG: Data in A1:D10 with row 5 completely blank → Ctrl+Shift+End selects A1:D4 (stops at blank) → Pivot table only reads rows 1-4 → COUNTA returns wrong count RIGHT: Data in A1:D10 with no blank rows → All operations detect the full range correctly
No blank rows or columns within any DDL data range. Period. This is the most common cause of 'my formula works but the answer is wrong' in Excel. If you need visual separation, use borders or conditional formatting — never blank rows.
Explore in graph →
3
CON-0001 CON Beginner

Center Across Selection

Centers text visually across multiple columns without merging cells, preserving sorting, filtering, references, and copy-paste operations.

▾
Intent
When you need centered text spanning multiple columns, use Center Across Selection to achieve visual alignment without merging cells.
Example
Select cells A1:D1 → Format Cells (Ctrl+1) → Alignment tab → Horizontal → Center Across Selection → OK. Text in A1 appears centered across A1:D1. Cells B1:D1 remain independently addressable.
The approved alternative to merged cells (ANT-0001). Mandatory in any DDL workbook where visual centering is needed. Resolves the most common formatting anti-pattern.
Explore in graph →
4
FRM-0001 FRM Intermediate

XLOOKUP

Retrieves a value from a range or array based on a lookup key, returning exact or approximate matches without requiring a column index number.

▾
Intent
When you need to retrieve a value from another table based on a lookup key, use XLOOKUP to return exact or approximate matches without column index dependency.
Example
=XLOOKUP(A2, Products[SKU], Products[Price], “Not Found“, 0)
Replaces VLOOKUP in all new builds. VLOOKUP should never appear in DDL workbooks.
Explore in graph →
5
FRM-0010 FRM Beginner

IF/IFS

Evaluates a condition and returns one value if true and another if false. IFS extends this to multiple conditions without nesting, improving readability for multi-branch logic.

▾
Intent
When you need conditional logic with multiple branches, use IF or IFS to evaluate conditions and return governed outcomes without deep nesting.
Example
=IF(A2>100, “High“, “Low“) =IFS(A2>100, “High“, A2>50, “Medium“, TRUE, “Low“)
IF is acceptable for binary conditions. For 3+ branches, use IFS or SWITCH instead of nested IF. For 5+ branches with complex logic, wrap in LET for named intermediate values.
Explore in graph →

You’ve walked the foundation. The graph has 50 more entries waiting — and 130 edges connecting them.

Explore the Full Graph →
Path 02 · The Star Schema Path
From raw data to governed architecture.
This is the DDL methodology applied to Excel. Start with documentation and grain discipline. End with a self-constructing matrix that updates as your data changes. This is how the Assurance Map was built.
1
CON-0006 CON Beginner

README Tab Convention

Establishes a dedicated README sheet as the first tab in every governed workbook, documenting purpose, owner, schema, change log, and usage instructions — making the workbook self-describing.

▾
Intent
When building any workbook intended for reuse or sharing, add a README tab first to document what the workbook does, who owns it, and how it is structured.
Example
Tab name: README (or _README to sort first) Required sections: Purpose — one sentence Owner — name + date Sheet Index — list all tabs + their role Change Log — date, author, change Governance Notes — any rules or constraints
Mandatory on all DDL workbooks per STD-EXCEL-001 Rule 0. A workbook without a README is ungoverned by definition. README must be updated on every structural change.
Explore in graph →
2
CON-0002 CON Advanced

Consistent Grain

Enforces that every row in a data table represents the same unit of measurement — one transaction, one period, one entity — preventing aggregation errors and ambiguous analysis.

▾
Intent
When you need reliable aggregation and filtering, use consistent grain to ensure every row in a table represents exactly one instance of the same thing.
Example
CORRECT: Every row = one sale (Date, Product, Qty, Revenue) WRONG: Row 1 = daily total, Row 2 = individual sale, Row 3 = monthly summary Test: “Can I SUM this column and get a meaningful number?“ If yes → grain is consistent. If no → grain is mixed.
Every DDL fact table must have documented grain. STD-FACT-001 enforces this — one row = one fact at one grain. Mixed grain is the most common cause of 'the numbers don't add up' in enterprise spreadsheets.
Explore in graph →
3
PTN-0001 PTN Advanced

Star Schema Layout

Organizes workbook data into dimension tables (descriptive attributes) and fact tables (transactional measures), enabling scalable lookups, consistent grain, and pivot-ready architecture.

▾
Intent
When you need to organize workbook data for scalable analysis, use star schema layout to separate dimension tables from fact tables, enabling governed lookups and pivot-ready architecture.
Example
Dimension tables: dim_Product (ProductID, Name, Category) Fact table: fct_Sales (Date, ProductID, Qty, Revenue) Lookup: =XLOOKUP(fct_Sales[@ProductID], dim_Product[ProductID], dim_Product[Name])
The DDL standard for any workbook with more than one entity type. All Assurance Map, BlindSpot, and dashboard workbooks use this pattern. Non-negotiable at Advanced tier and above.
Explore in graph →
4
PTN-0008 PTN Intermediate

Surrogate Key Generation

Generates a stable, system-assigned integer key for each row using ROW() offset or a counter column, decoupled from any business attribute — providing a reliable join key that survives source data changes.

▾
Intent
When a fact table row needs a stable, non-business join key, use surrogate key generation to produce an integer ID that never changes even when source attributes do.
Example
=ROW()-1 // in a table starting at row 2 — simple offset =TEXT(ROW()-1,“00000“) // zero-padded for display // Or: named column SK_ID with sequence starting at 1
Required on all DDL fact tables per STD-EXCEL-001 Rule 2b. Never use source system IDs as surrogate keys — they couple your workbook to upstream naming conventions.
Explore in graph →
5
ARC-0001 ARC Expert

MAKEARRAY Engine

Generates a dynamic grid of calculated values across two dimensions using MAKEARRAY with LAMBDA, building self-constructing matrices that eliminate manual cell-by-cell formulas.

▾
Intent
When you need to generate a dynamic grid of calculated values across two dimensions, use MAKEARRAY with LAMBDA to build self-constructing matrices that eliminate manual cell-by-cell formulas.
Example
=LET( rows, COUNTA(RowHeaders), cols, COUNTA(ColHeaders), MAKEARRAY(rows, cols, LAMBDA(r,c, XLOOKUP(INDEX(RowHeaders,r), Data[Key], INDEX(Data,,c+1)) )) )
The DDL signature architecture. Used in Assurance Map, Keith's audit dashboard, BlindSpot analytics. Every DDL dashboard matrix should use this pattern.
Explore in graph →

You’ve walked the star schema path. This is the architecture behind every DDL analytical workbook.

Continue in the Graph →