Text operations

The .str accessor

df['name'].str.lower()
df['name'].str.upper()
df['name'].str.strip()          # whitespace at both ends
df['name'].str.len()
df['name'].str.replace('Ltd', 'Limited')
df['name'].str.contains('Corp', na=False)
df['name'].str.startswith('A')

Everything text-related goes through .str. Without it you are calling the method on the Series object rather than on its values.

The standard clean

df['email'] = df['email'].str.strip().str.lower()

Run this on every text key before joining or deduplicating. Trailing whitespace and mixed case are the two reasons merges silently fail to match on values that look identical.

Splitting

df[['first', 'last']] = df['full_name'].str.split(' ', n=1, expand=True)
df['domain'] = df['email'].str.split('@').str[1]

expand=True returns a DataFrame you can assign to several columns. n=1 splits only on the first occurrence, so a middle name does not break the assignment.

Extracting

df['digits'] = df['phone'].str.replace(r'\D', '', regex=True)
df['code']   = df['sku'].str.extract(r'^([A-Z]{3})')
df['num']    = df['label'].str.extract(r'(\d+)').astype(float)

extract takes a regular expression with a capture group and returns what it captured, or NaN where there was no match.

na=False