1 Spreadsheet Fundamentals and Data Organization
Learn how spreadsheet grids and cell references work, how stored values differ from their display formats, and how to organize consistent datasets for reliable analysis.
Spreadsheet grids and addresses
A spreadsheet is a grid of columns labeled with letters and rows labeled with numbers. Their intersections are cells, each identified by an address: B3, for example, is in column B and row 3. A group of cells is a ; A1:C4 means every cell from A1 through C4.
A is one grid within a spreadsheet file, or workbook. Cell references and ranges identify locations for describing data and referring to it in calculations.
How cell references behave
Cell references can behave differently when a formula is copied:
A , such as
A2, changes according to the formula’s new location.An , such as
$A$2, stays fixed.A fixes just one part:
$A2fixes the column, andA$2fixes the row.
For example, if a tax rate is stored in F1, a formula copied down a column can use $F$1 so every row uses the same rate. Spreadsheet applications may offer other reference styles, but the letter-column, number-row is common.
Values, data types, and formats
A cell’s is the information stored in it; its controls how that information appears. Common kinds of data include:
Text: Names, categories, and codes. Identifiers such as ZIP codes or product codes are often best treated as text, especially when leading zeros matter.
Numbers: Quantities and measurements that may be used in arithmetic.
Dates and times: Calendar or time values. A consistent, unambiguous convention such as
YYYY-MM-DDhelps keep dates consistent.Logical values:
TRUEorFALSE.Formulas: Expressions that calculate a result from values or references.
Number formatting can display a stored number as currency, a percentage, or a date without necessarily changing its underlying . Formatting a cell to look right does not guarantee that its contents have the intended . For example, a number stored as text may not sort or calculate as expected. Check imported data and use consistent types within each column.
Build a consistent dataset
Before analysis, decide what one row represents and apply that rule throughout. A row might represent one order, one person, or one daily measurement. Each column should describe one characteristic of the records. If parts of a field need separate analysis, put them in separate columns—for example, store city and state separately rather than combining them in one location field.
Use a header row with short, meaningful, unique column names, and avoid empty headers. Keep entries consistent within each column: use the same units, spelling, date convention, and . For example, do not mix 12, twelve, and N/A in a numeric field.
Represent missing information by leaving a blank or using a documented missing- convention. Do not use a number such as 0 to mean missing unless zero is the true . Check for duplicate records, unexpected blanks, inconsistent labels, and values outside plausible ranges.
Keep data analysis-ready
Keep the dataset in one continuous rectangular block. Do not put blank rows or columns inside it, use merged cells in it, or mix totals into the records. Keep explanations, subtotals, and charts outside the data block or on another so they are not mistaken for records.
Organized data is easier to select, sort, filter, and analyze. For related data in Excel, Microsoft recommends placing similar items in the same column and using a table.
Example: order-line records
In the example dataset, the columns are OrderID, Date, Item, Quantity, and UnitPrice. The row with 1001, 2026-09-03, Notebook, 2, and 4.50 represents one order line; the row with 1002, 2026-09-04, Pen, 5, and 1.25 represents another. Each row is one , and each column describes an attribute of that .