Count Table Rows: Methods, SQL, and Spreadsheet Tips

Updated 5 November 2026 · Standard formula, shown below

Counting table rows means finding how many records a table contains, but the right method depends on what “table” means and whether you need every row or only rows matching a condition. For a database, use a row-count query; in a spreadsheet or document, use its row indicators or selection tools. The result is a count, not a count of columns or cells.

What Counts as a Table Row?

A row is one horizontal record in a structured table. In a database, a row—also called a record or tuple—contains values for the table’s columns. In a spreadsheet, a worksheet row is a numbered horizontal band of cells, while a formatted table may occupy only part of the worksheet. In a document, a row is one horizontal set of cells inside a table.

These meanings matter because the number shown by a tool may refer to different things. A worksheet can have many numbered rows that are entirely blank, yet its data table may contain only a few records. A database table can also have rows that are hidden from a particular view or excluded by a filter. Decide whether you mean all stored records, visible records, or nonempty records before counting.

Counting Rows in a Database

To count every row in a relational database table, use the SQL aggregate function COUNT with an asterisk:

SELECT COUNT(*) FROM orders;

This returns one number: the number of rows in orders visible to that query. The table name is an example; replace it with the name used in your database. If you need rows that meet a condition, add a WHERE clause:

SELECT COUNT(*)
FROM orders
WHERE status = 'shipped';

Here, the count includes only rows whose status matches the condition. Multiple conditions can be combined with AND or OR. For example, a date range can limit the result to orders placed within a specified period.

COUNT(*) counts rows regardless of whether individual columns contain NULL. By contrast, COUNT(column_name) counts only rows where that particular column is not NULL. This is a frequent source of confusing results. If a row has no value for email, it still contributes to COUNT(*), but not to COUNT(email).

Counting Rows in a Spreadsheet

In a spreadsheet, first identify the range that represents the data. If each record has a required field that is filled in on every record, a count function can count nonblank cells in that field. For example, if order IDs are in cells A2 through A100, a spreadsheet formula such as =COUNTA(A2:A100) counts the nonempty cells in that range.

This works only if the chosen field is reliably populated once per record. If some records lack an order ID, the result will be too low. If the range includes headings, notes, or stray values, the result may be too high. In spreadsheets that support structured references, counting a reliably populated table column can make a formula easier to maintain as the table grows.

For a quick manual check, select the cells in a column that should contain one value per record. Many spreadsheet applications display a selection count in a status area, though what that count includes can depend on the selection and application. Check the displayed label and compare it with a formula when accuracy matters. Filtering also changes what is visible, and some tools provide a count of visible cells separately from a count of all cells in the range.

Visible Rows, Blank Rows, and Filters

“How many rows are in the table?” can mean at least three different counts:

Total rows: every record stored in the table, whether or not it is currently visible.
Matching rows: records that meet a filter or query condition.
Nonempty records: rows with data in a chosen field, often counted through a spreadsheet column.

Blank rows deserve special attention. In a database, a row can exist even if some or all of its fields are empty; it is still a row. In a spreadsheet, a blank worksheet row between two groups may not belong to the data table at all. Conversely, a row with formulas that display an empty string may look blank but still contain formulas. A nonblank-cell count can therefore differ from a visual count.

Filters do not necessarily delete records. A database query with a WHERE clause returns a count of matching rows, while a spreadsheet filter generally hides nonmatching records from view. Use a tool or formula specifically intended to count visible rows if that is what you need, and verify whether manually hidden rows are included.

Choosing the Right Counting Method

Use the method that matches the question, rather than choosing a formula first. For all rows in a database table, COUNT(*) is the direct choice. For rows meeting a condition, count the query result with a WHERE clause. In a spreadsheet, count a dependable identifier column or use a feature that reports table records. In a document, the application’s table-selection or properties controls may show row information, but available features differ.

Before trusting a result, check these points:

Does the count include a header row?
Are hidden or filtered-out rows meant to count?
Are blank records included, or only records with a value in a particular field?
Does the selected range stop at the actual end of the table?
Could duplicate records be mistaken for one record? A row count counts rows, including duplicates.

Common Counting Mistakes

One common mistake is counting cells instead of records. A table with 12 records and 5 columns has 12 rows and 60 data cells, not 60 rows. Another is counting a column’s non-null values and assuming that equals the row count; it does only when every row has a value in that column.

People also confuse a table’s data rows with the row numbers shown along a spreadsheet’s edge. Those numbers identify worksheet positions, not necessarily records in the table. A heading row may be part of the selected range but not part of the record count. State explicitly whether the header is included, especially when reporting results to someone else.

Finally, be aware that database counts can take time on very large tables, depending on the database system, storage, indexes, and query conditions. Do not assume a displayed estimate is the same as an exact count. If exactness matters, run a count query against the relevant table and conditions, and confirm that the query is looking at the intended data and transaction state.

Frequently Asked Questions

What SQL command counts all rows in a table?
Use SELECT COUNT(*) FROM table_name;, substituting the actual table name. Add a WHERE clause when you want to count only rows meeting a condition.

Does COUNT(*) include rows with NULL values?
Yes. It counts each row returned by the query, even if some or all column values are NULL. COUNT(column_name), in contrast, excludes rows where that column is NULL.

How do I count rows with data in Excel or another spreadsheet?
Count a column that should contain one nonblank value per record, such as an ID column. A function such as COUNTA can count nonempty cells in the chosen range, but verify that the range excludes headings and unrelated entries.

Does a filter change the total number of table rows?
No. A filter usually changes which rows are displayed, not which records exist. A count of visible rows may be lower than the total count, so check which one your tool or formula reports.

Are blank rows counted?
A database row is counted even when its fields are blank or NULL. A completely empty spreadsheet row may not be part of the data table, so its inclusion depends on the selected range and counting method.

Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.