Practice Now
pandas
python
data manipulation
string operations
data analysis

Add a string prefix to each value in a pandas string column

ML System Design practice on Codemia

Design recommenders, ranking systems and training pipelines the way ML interviews actually ask for them, with worked solutions.

Practice ML system design
Data SciencePython

Introduction

Adding a prefix to every value in a Pandas string column is a common data transformation. The simplest approach is string concatenation with the + operator: df['col'] = 'prefix_' + df['col']. The .str accessor provides cat() for more complex operations, and apply() with a lambda works for custom formatting. All these methods are vectorized (except apply) and handle the operation without Python loops.

Method 1: String Concatenation with +

python
1import pandas as pd
2
3df = pd.DataFrame({'id': ['001', '002', '003'], 'name': ['Alice', 'Bob', 'Charlie']})
4
5# Add prefix to id column
6df['id'] = 'USR-' + df['id']
7print(df)
8#       id     name
9# 0  USR-001    Alice
10# 1  USR-002      Bob
11# 2  USR-003  Charlie

This is the fastest and most readable approach for simple prefixes.

Method 2: .str.cat() (Concatenation Accessor)

python
1# Using str accessor
2df['name'] = df['name'].str.cat(['Dr. '] * len(df), sep='')
3# Alternatively:
4df['name'] = 'Dr. ' + df['name']

str.cat() is more useful for joining two columns or adding a separator:

python
# Combine two columns with a separator
df['full'] = df['first_name'].str.cat(df['last_name'], sep=' ')

Method 3: apply() with Lambda

python
1df['id'] = df['id'].apply(lambda x: f'USR-{x}')
2
3# Or with a custom function
4def add_prefix(value):
5    return f'ID_{value.upper()}'
6
7df['id'] = df['id'].apply(add_prefix)

apply is slower than vectorized operations but useful for complex transformations.

Method 4: map() with Format String

python
1df['id'] = df['id'].map('USR-{}'.format)
2print(df)
3#       id     name
4# 0  USR-001    Alice
5# 1  USR-002      Bob
6# 2  USR-003  Charlie
7
8# f-string equivalent (Python 3.6+)
9df['id'] = df['id'].map(lambda x: f'USR-{x}')

Adding Both Prefix and Suffix

python
1df['name'] = '[' + df['name'] + ']'
2# [Alice], [Bob], [Charlie]
3
4# Or with str methods
5df['name'] = df['name'].apply(lambda x: f'({x})')

Conditional Prefix

Add prefix only to rows that meet a condition:

python
1df = pd.DataFrame({
2    'name': ['Alice', 'Bob', 'Charlie'],
3    'role': ['admin', 'user', 'admin']
4})
5
6# Add prefix only for admins
7df.loc[df['role'] == 'admin', 'name'] = 'Admin: ' + df.loc[df['role'] == 'admin', 'name']
8print(df)
9#            name   role
10# 0  Admin: Alice  admin
11# 1           Bob   user
12# 2  Admin: Charlie  admin
13
14# Using np.where
15import numpy as np
16df['name'] = np.where(df['role'] == 'admin', 'Admin: ' + df['name'], df['name'])

Handling Non-String Columns

If the column contains numbers, convert to string first:

python
1df = pd.DataFrame({'id': [1, 2, 3]})
2
3# This fails: TypeError
4# df['id'] = 'USR-' + df['id']
5
6# Convert to string first
7df['id'] = 'USR-' + df['id'].astype(str)
8# USR-1, USR-2, USR-3
9
10# With zero-padding
11df['id'] = 'USR-' + df['id'].astype(str).str.zfill(3)
12# USR-001, USR-002, USR-003

Handling NaN Values

python
1df = pd.DataFrame({'name': ['Alice', None, 'Charlie']})
2
3# Concatenation with NaN produces NaN
4result = 'Hello ' + df['name']
5# 0    Hello Alice
6# 1           NaN
7# 2  Hello Charlie
8
9# Fill NaN first if needed
10df['name'] = 'Hello ' + df['name'].fillna('Unknown')
11# Hello Alice, Hello Unknown, Hello Charlie
12
13# Or skip NaN rows
14mask = df['name'].notna()
15df.loc[mask, 'name'] = 'Hello ' + df.loc[mask, 'name']

Performance Comparison

python
1import timeit
2
3df = pd.DataFrame({'col': [f'val_{i}' for i in range(100_000)]})
4
5# Fastest: string concatenation
6%timeit 'prefix_' + df['col']          # ~1.5 ms
7
8# Fast: map with format
9%timeit df['col'].map('prefix_{}'.format)  # ~15 ms
10
11# Slower: apply with lambda
12%timeit df['col'].apply(lambda x: f'prefix_{x}')  # ~25 ms
13
14# Slowest: list comprehension (not vectorized)
15%timeit [f'prefix_{x}' for x in df['col']]  # ~30 ms

String concatenation with + is 10-20x faster than apply because it uses Pandas' internal C-optimized string operations.

Common Pitfalls

  • TypeError with non-string columns: 'prefix' + df['int_column'] raises TypeError. Convert the column to string first with .astype(str) before concatenation.
  • NaN propagation: String concatenation with NaN produces NaN, not 'prefix_nan'. Use .fillna('') before concatenation if you want to preserve the prefix for missing values.
  • Using apply for simple prefix: apply(lambda x: 'prefix_' + x) works but is 10-20x slower than 'prefix_' + df['col']. Use vectorized string operations for simple transformations.
  • Modifying a copy instead of the original: df['col'].str.upper() returns a new Series. Assigning back with df['col'] = ... is required. Without assignment, the original DataFrame is unchanged.
  • Mixed types in column: If a column has both strings and numbers (object dtype with mixed types), string concatenation may fail on numeric rows. Use df['col'].astype(str) to normalize the column first.

Summary

  • Use 'prefix_' + df['col'] for the fastest and simplest prefix operation
  • Use .astype(str) first if the column contains non-string values
  • Use .fillna() to handle NaN values before concatenation
  • Use np.where or df.loc for conditional prefix application
  • Avoid apply() for simple string operations — vectorized concatenation is 10-20x faster
Related reading
Free course
Beginner
7 lessons
2 hours
Tackling System Design Interview Problems

A short course that equips you with the skills to approach system design interviews methodically.

Start the free course
Track what you have practised

A free account saves your progress, solutions and study plan across every problem on Codemia.

ML System Design practice on Codemia

Design recommenders, ranking systems and training pipelines the way ML interviews actually ask for them, with worked solutions.

Practice ML system design
Codemia
© 2026 Codemia
Resources
Legal
Social
All Rights Reserved.