Intermediate14 columnsUp to 10,000 rows Fictional data

Employee & Attrition Records dataset

An HR file of employees with salary, satisfaction, overtime and whether they left: built for attrition analysis.

About this dataset

A fictional company's HR export: one row per employee, with department, role, city, joining date, monthly salary, whether they regularly work overtime, how satisfied they said they were, their last performance rating, and the column everyone cares about: attrition, whether they left.

Attrition is a classic first machine-learning problem because the answer is a simple Yes/No and the reasons are human and easy to reason about. The file is also handy for plain spreadsheet practice: salary bands, headcount by department, average tenure, and lookups.

Good for practising: Classification, Feature analysis, Excel lookups, Data visualisation.

What's built into the data

The rows are random, but they follow rules, so there is something real to find. These are the patterns to look for:

Employees who work overtime, report low satisfaction (1–2), are under 30, or are junior and paid under ₹30,000 a month are noticeably more likely to leave.
People with five or more years at the company are less likely to leave.
Engineering pays the most and Customer Support the least; each step up in seniority adds roughly 65% to pay.
Roughly one employee in five leaves, so the classes are imbalanced, as they usually are in real HR data.
years_at_company is counted up to 31 March 2026 and always agrees with date_of_joining.

Columns

Column	Type	Description
`emp_id`	ID (text)	Employee number.
`full_name`	Text	Fictional name.
`gender`	Category	Male or Female.
`age`	Whole number	Age in years.
`department`	Category	One of seven departments.
`job_role`	Category	Role title; three levels of seniority per department.
`city`	Category	Office city.
`date_of_joining`	Date	Date the employee joined.
`years_at_company`	Whole number	Complete years served as of 31 Mar 2026.
`monthly_salary`	Amount (₹)	Gross monthly salary in rupees.
`overtime`	TRUE / FALSE	True if the employee regularly works overtime.
`job_satisfaction`	Whole number	Self-reported, 1 (very low) to 5 (very high).
`performance_rating`	Whole number	Last appraisal, 1 to 5.
`attrition`	Category	Yes if the employee left, No otherwise.

Preview: the first 8 rows

A sample with seed 2026. Your download will have different values but the same columns and patterns.

emp_id	full_name	gender	age	department	job_role	city	date_of_joining	years_at_company	monthly_salary	overtime	job_satisfaction	performance_rating	attrition
EMP00001	Varun Fernandes	Male	48	Operations	Operations Manager	Kolkata	2016-09-24	9	69000	FALSE	4	4	No
EMP00002	Aarav Bansal	Male	21	Sales	Sales Executive	Pune	2026-03-01	0	28400	FALSE	4	3	No
EMP00003	Mary Banerjee	Female	31	Marketing	Marketing Executive	Pune	2024-07-11	1	27300	TRUE	4	4	Yes
EMP00004	Simran Ansari	Female	29	Human Resources	HR Executive	Hyderabad	2023-05-16	2	31200	FALSE	4	4	No
EMP00005	Revathi Chauhan	Female	25	Marketing	Marketing Executive	Chennai	2026-01-27	0	31200	FALSE	4	4	No
EMP00006	Abhishek Deshmukh	Male	26	Customer Support	Support Associate	Hyderabad	2023-07-14	2	24400	FALSE	5	3	No
EMP00007	Varun Sheikh	Male	29	Customer Support	Support Associate	Hyderabad	2026-03-01	0	21600	FALSE	3	3	No
EMP00008	Ankit Ansari	Male	43	Finance	Finance Analyst	Kolkata	2013-08-27	12	68000	TRUE	3	3	Yes

Practice exercises

Find the attrition rate for each department, and for overtime vs no overtime. Which single factor matters most?
Plot monthly salary by job role as box plots. Are there outliers inside a role?
Build a logistic regression or decision tree to predict attrition. Report precision and recall, not just accuracy.
Because only about 1 in 5 employees leave, compare a model that always says 'No' with yours. Did you beat it?
Make an Excel salary band table (under 30k, 30–50k, 50–80k, 80k+) with COUNTIFS and the attrition rate in each band.

Stuck on one? Each exercise uses ideas from the lessons below, and trying it with the messy option on turns it into a cleaning exercise as well.

Load it straight into Python

You don't have to download first. Pandas can read the link directly, in Jupyter, Google Colab or a script. Keep the seed so you get the same data every time you run the notebook:

import pandas as pd

url = "https://beeprotonai.info/resources/datasets/employees/download/?rows=1000&seed=42"
df = pd.read_csv(url, parse_dates=["date_of_joining"])
df.head()

For SQL practice, download the SQL format and run it against an empty database, for example sqlite3 practice.db < employees-1000-rows-seed42.sql or mysql -u root -p practice < employees-1000-rows-seed42.sql (use your file's name). It creates a table called employees and fills it.

About the data

Everything in this file is invented. Names are random pairings of common first names and surnames, email addresses use example.com (reserved for examples, so it can never reach anyone), and phone numbers start with +91 00, which no real Indian mobile number does. Any resemblance to a real person or business is coincidence. Dates run from 1 April 2024 to 31 March 2026 unless a column says otherwise. Use it freely for learning, teaching and projects, including commercial ones, with no attribution required (CC0).

Lessons that go with it

More datasets

Sweet Shop Sales

Two years of orders from a fictional chain of Indian sweet shops, with Diwali …

Student Exam Results

Marks in five subjects with attendance and study hours, plus totals, percentage, result and …

Customer List (CRM)

Customer records with contact details, segment, orders and spend: ideal for cleaning, deduplication and …

Stock Register (Inventory)

Monthly stock register for five warehouses: opening, received, issued, closing, reorder level and value.

Website Traffic (Analytics)

Daily sessions, users, bounce rate and conversions by page, traffic source and device.

Daily City Weather

Simulated daily temperature, rain, humidity and AQI for Indian cities, with monsoon and winter …

See all 14 datasets →