Sample CSV Datasets for Download

Each file is UTF-8, comma-separated, and starts with a header row. Download a file, then use your spreadsheet or database import command. Choose comma as the delimiter, double quotes as the text qualifier, and the first row as column names. Check that the result has the column and row counts shown above, excluding the header.

CSV stores text fields, not a saved column-type definition. The kinds below describe how to interpret the values. Set identifier columns to text, choose date and number types where needed, and check true/false conversion in your importer. Decimal amounts have no currency symbol or currency code; do not assume they are dollars.

These files are for checking imports and learning CSV behavior. They do not represent real sales, stock, or support activity and are too small for a large-file performance benchmark.

Use this eight-row example to check dates, decimal amounts, true/false values, and product names that contain a comma.

month,region,product,units,revenue,returned
2026-01-01,North,Field Notes,142,3408.00,false
2026-01-01,South,Canvas Tote,96,2688.00,false
2026-02-01,East,"Desk Tray, Small",74,2812.00,true
2026-02-01,West,Field Notes,165,3960.00,false
2026-03-01,North,Canvas Tote,108,3024.00,false
2026-03-01,South,"Desk Tray, Small",89,3382.00,true
2026-04-01,East,Field Notes,151,3624.00,false
2026-04-01,West,Canvas Tote,117,3276.00,false

month

Date text. First day of the reporting month, written as YYYY-MM-DD.

region

Text. Fictional sales region: North, South, East, or West.

product

Text. Product name. Desk Tray, Small contains a comma inside a quoted field.

units

Whole number text. Number of units in this fictional product and region entry.

revenue

Decimal text. Fictional revenue amount with two decimal places. No currency is specified.

returned

Boolean text. A fictional return flag, written as lowercase true or false.

The field "Desk Tray, Small" must remain one cell. Splitting each line at every comma would incorrectly create seven columns in those rows.

Use this six-row example to test stock identifiers, quantities, reorder thresholds, and decimal costs. Compare on_hand with reorder_point to practice finding low stock.

sku,item,warehouse,on_hand,reorder_point,unit_cost
INV-001,Recycled Notebook,Auckland,84,30,4.50
INV-002,Canvas Pouch,Wellington,21,25,8.75
INV-003,"Desk Tray, Small",Christchurch,43,18,12.40
INV-004,Refill Pack,Auckland,160,60,2.10
INV-005,Brass Clip Set,Wellington,12,20,6.80
INV-006,Weekly Planner,Christchurch,57,24,7.25

sku

Identifier text. Fictional stock keeping unit, such as INV-001. The zero-padded suffix is part of the identifier.

item

Text. Item name, including the quoted name Desk Tray, Small.

warehouse

Text. Warehouse location label: Auckland, Wellington, or Christchurch.

on_hand

Whole number text. Fictional number of units currently in stock.

reorder_point

Whole number text. Fictional stock threshold used for comparison with on_hand.

unit_cost

Decimal text. Fictional cost per unit with two decimal places. No currency is specified.

Import sku as text to preserve identifiers exactly. If you extract the 001 suffix, keep it as text too: converting it to a number would remove the leading zeros.

Use this seven-row example to check UTC timestamps and group fictional tickets by channel, category, priority, or status.

ticket_id,opened_at,channel,category,priority,status
TKT-1001,2026-06-02T09:15:00Z,web,Billing,medium,resolved
TKT-1002,2026-06-02T10:42:00Z,chat,Delivery,high,open
TKT-1003,2026-06-03T01:08:00Z,web,Product question,low,resolved
TKT-1004,2026-06-03T06:30:00Z,phone,Returns,medium,pending
TKT-1005,2026-06-04T11:22:00Z,chat,Account,high,resolved
TKT-1006,2026-06-05T03:54:00Z,web,Delivery,low,open
TKT-1007,2026-06-05T08:16:00Z,phone,Billing,medium,resolved

ticket_id

Identifier text. Fictional unique ticket reference, such as TKT-1001.

opened_at

Timestamp text. Ticket opening time in ISO 8601 format. The Z suffix means UTC.

channel

Text. Contact channel: web, chat, or phone.

category

Text. Fictional topic such as Billing, Delivery, or Product question.

priority

Text. Priority label: low, medium, or high.

status

Text. Ticket state: open, pending, or resolved.

Keep the Z timezone when reading opened_at. A spreadsheet may display UTC timestamps in your local timezone; check that behavior before comparing times or grouping by day.

A three-record parser exercise. A quoted newline is part of one field, so physical line count is not record count.

id,label,note,active
00123,"Tray, small","He said ""hello""",true
00007,Café,"Line one
Line two",false
00900,東京,,true

id

Identifier text. Keep 00123, 00007 and 00900 as strings; leading zeros are intentional.

label

Unicode text. Includes a comma, Café and 東京. Read as UTF-8.

note

Quoted text. Escaped double quotes, a real embedded newline and an empty value.

active

Boolean text. Lowercase true/false strings.

This is valid CSV. Use a CSV parser, not line splitting. Store identifiers as text in a spreadsheet.

A deliberately dirty but syntactically valid CSV. The first two records match after trimming; amount and status need review.

id,name,amount,status
001,  Widget  ,12.50,active
001,Widget,12.50,active
002,Gadget,,pending
003,Thing,not-a-number,active
004,Thing,9.00,UNKNOWN

id

Identifier text. 001 repeats; use it to investigate duplicates.

name

Text. Widget includes leading and trailing spaces in the first record.

amount

Mixed text. Includes a blank and not-a-number. No currency is specified.

status

Category text. UNKNOWN intentionally differs from active/pending.

Trimming and duplicate removal do not repair amounts or category choices. These intentional data-quality issues are not malformed CSV syntax.

Practice distinguishing a readable CSV from values that satisfy a schema. Record 3 intentionally breaks typed-value expectations; record 4 repeats an ID.

id,quantity,date,active
001,12,2026-01-01,true
002,-3,2026-02-02,false
003,twelve,not-a-date,yes
001,4,2026-03-03,true

id

Identifier text. Intended to be required and unique; 001 deliberately repeats.

quantity

Integer text. Includes -3 and twelve to exercise numeric and range checks.

date

Date text. Expected YYYY-MM-DD, but one value is not-a-date.

active

Boolean text. Expected true/false, but one value is yes.

The syntax is valid. Configure the expected field types and any supported constraints in your validator; CSV itself does not carry a schema.

Exactly 1,000 fictional records generated by a deterministic pattern. Preview below shows only the first five records; download and viewer contain all 1,000.

Preview: first five of 1,000 records

order_id,region,quantity,amount
000001,North,2,4.50
000002,South,3,6.50
000003,East,4,8.50
000004,West,5,10.50
000005,North,6,12.50

order_id

Identifier text. Sequential six-character identifiers from 000001 to 001000.

region

Category text. Repeating North, South, East and West labels.

quantity

Integer text. Deterministic values from 1 to 9.

amount

Decimal text. Synthetic amounts, with no currency implied.

Created here without external source data. Small enough for an import exercise; use Generate CSV for configurable records rather than interpreting this as a scale benchmark.

Use Generate CSV to create a custom sample with named columns and a chosen record count. Use Validate CSV for syntax and supported schema checks, or Clean CSV to trim the dirty fixture and remove records that then match.

These datasets are fictional and contain no personal identifiers or public email addresses. They cover quoted fields, dates, timestamps, identifiers, booleans, whole numbers, and decimals without pretending to be real operational data.

They are dedicated to the public domain under CC0 1.0. You may copy, adapt, and redistribute them without attribution.