Exporting data

Export scrape_jobs() results to CSV, Excel, JSON, Parquet, and more.

scrape_jobs() returns a standard Pandas DataFrame. You can use any pandas export method to save results to disk.

DataFrame columns

Results are always returned in the following column order:

Column	Description
id	Internal job ID
site	Source job board
job_url	URL to the job listing
job_url_direct	Direct application URL (if available)
title	Job title
company	Company name
location	City, state, and/or country
date_posted	Date the job was posted
job_type	Employment type (e.g. fulltime, contract)
salary_source	direct_data or description
interval	Pay interval (yearly, monthly, weekly, daily, hourly)
min_amount	Minimum salary/pay
max_amount	Maximum salary/pay
currency	Currency code (e.g. USD)
is_remote	Whether the job is remote
job_level	Seniority level (LinkedIn only)
job_function	Job function category (LinkedIn only)
listing_type	Listing type
emails	Contact emails extracted from description
description	Full job description
company_industry	Industry (LinkedIn and Indeed)
company_url	Company profile URL
company_logo	Company logo URL (Indeed)
company_url_direct	Direct company website URL
company_addresses	Company office addresses (Indeed)
company_num_employees	Employee count label (Indeed)
company_revenue	Revenue label (Indeed)
company_description	Company description (Indeed)
skills	Required skills (Naukri)
experience_range	Experience required (Naukri)
company_rating	Company rating (Naukri)
company_reviews_count	Number of company reviews (Naukri)
vacancy_count	Number of open positions (Naukri)
work_from_home_type	WFH type, e.g. Hybrid, Remote (Naukri)

Export formats

CSV

import csv
from jobspy import scrape_jobs

jobs = scrape_jobs(
    site_name=["indeed", "linkedin"],
    search_term="software engineer",
    location="Austin, TX",
)

jobs.to_csv(
    "jobs.csv",
    quoting=csv.QUOTE_NONNUMERIC,
    escapechar="\\",
    index=False,
)

Selecting specific columns

You can select a subset of columns before exporting to keep files smaller.

import csv
from jobspy import scrape_jobs

jobs = scrape_jobs(
    site_name=["indeed", "linkedin"],
    search_term="data analyst",
    location="New York, NY",
)

columns = ["site", "title", "company", "location", "job_url", "min_amount", "max_amount", "interval"]
jobs[columns].to_csv("jobs_slim.csv", quoting=csv.QUOTE_NONNUMERIC, escapechar="\\", index=False)

Sorting results

The DataFrame is pre-sorted by site and date_posted (newest first). You can re-sort by any column.

# Sort by minimum salary, highest first
jobs_sorted = jobs.sort_values("min_amount", ascending=False)

# Sort by date posted, most recent first
jobs_sorted = jobs.sort_values("date_posted", ascending=False)

Normalizing salary to annual figures

Job boards report salary in different intervals — some hourly, some monthly, some yearly. Use enforce_annual_salary=True to automatically convert all non-yearly wages to annual equivalents before they are returned.

from jobspy import scrape_jobs

jobs = scrape_jobs(
    site_name=["indeed", "zip_recruiter"],
    search_term="nurse",
    location="Houston, TX",
    enforce_annual_salary=True,  # all salary amounts converted to yearly
)

# Now min_amount and max_amount are comparable across all rows
print(jobs[["title", "company", "interval", "min_amount", "max_amount"]].head())

Conversion rates used internally:

Interval	Multiplier
Hourly	× 2,080
Weekly	× 52
Monthly	× 12
Daily	× 260

Controlling description format

Job descriptions can be returned as Markdown or HTML. Use the description_format parameter.

jobs = scrape_jobs(
    site_name="indeed",
    search_term="technical writer",
    location="Remote",
    description_format="markdown",  # default — human-readable
)

jobs_html = scrape_jobs(
    site_name="indeed",
    search_term="technical writer",
    location="Remote",
    description_format="html",  # raw HTML for rendering in a UI
)

Accepted values: "markdown" (default), "html", or "plain" (plain text with HTML tags stripped).