Glassdoor Company Scraper - Most Comprehensive

Glassdoor Company Scraper

Collect Glassdoor company profiles by company page URL, employer ID, company name, or website. The default Dataset holds one company record for each employer, with details, ratings, locations, benefits, photos, and updates. Interviews, salaries, benefit reviews, and current jobs go to four more Datasets, one for each kind. See Output datasets.

Company profile coverage

Group	Data in a full company record
Company profile	Identity, website, description, size, revenue, headquarters, industry, leadership, relationships, and recognition
Ratings and locations	Overall and detailed ratings, distributions, demographics, review locations, and offices
Company content	Managed profile content, company updates, and photos
Optional research	Salary estimates, benefits and benefit reviews, interview data, and current jobs. Salaries, benefit reviews, interviews, and jobs are items of their own Datasets.

Compact mode returns the core company fields as one flat row. Full mode keeps the nested sections and company relationships.

Choose full or compact records

Need	Set	Result
A nested company profile for research	Keep compactMode: false	Core company data, ratings, locations, and each enabled section
One flat row for a table or CSV file	Set compactMode: true	Core fields only; section switches are ignored

Collect a company from its page

{
  "urls": [
    "https://www.glassdoor.com/Overview/Working-at-Nucor-EI_IE489.htm"
  ]
}

Any supported Glassdoor company page that contains the employer ID can identify the company.

Collect a company by employer ID

{
  "employerIds": [489]
}

Find a company by name or website

{
  "searchQueries": ["Nucor", "https://www.nucor.com"],
  "maxSearchResults": 1
}

maxSearchResults applies to each search value. With a website value, strictDomainMatch can keep results that use the same registered domain.

Create flat rows for a table

{
  "employerIds": [489],
  "compactMode": true
}

Add salary, benefit, interview, and job data

Enable only the research sections that you need:

{
  "employerIds": [489],
  "includeSalaries": true,
  "maxSalaries": 20,
  "includeBenefits": true,
  "maxBenefitReviews": 20,
  "includeInterviews": true,
  "maxInterviews": 20,
  "includeJobs": true,
  "maxJobs": 20
}

The company record keeps the benefits overview, the salary and interview statistics, and the row counts. The salaries, benefit reviews, interviews, and jobs themselves go to their own Datasets.

Output datasets

Each run writes up to five Datasets. A Dataset for a section stays empty when you do not switch that section on.

Dataset	One item is	Switch	Count in the company record	Link in the run output
Default Dataset	One company record for each employer	None	None	Companies
interviews	One interview	includeInterviews	rowsWritten.interviews	Interviews
salaries	One salary estimate for a job title	includeSalaries	rowsWritten.salaries	Salaries
benefitReviews	One benefit review	includeBenefits	rowsWritten.benefitReviews	Benefit reviews
jobs	One current job listing	includeJobs	rowsWritten.jobs	Jobs

Each Dataset has its own table view in the Output tab of the run. You can also open each Dataset in the Storage tab of the run. To read one with the API, read the run and take the Dataset ID from storageIds.datasets, for example storageIds.datasets.interviews. Then request that Dataset's items. The Export button on the run page exports only the default Dataset. Use the Storage tab to export the other four. They are kept for as long as the other storage of the run.

Read all Datasets of a run

This example uses the Apify client for JavaScript. It reads the company records and the interviews of a run, and it finds the company record of the first interview by employerId. Replace interviews with salaries, benefitReviews, or jobs to read another Dataset.

import { ApifyClient } from "apify-client";

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.run("<RUN_ID>").get();
const { datasets } = run.storageIds;

const readAll = async (datasetId) =>
  JSON.parse(await client.dataset(datasetId).downloadItems("json"));

const companies = await readAll(datasets.default);
const interviews = await readAll(datasets.interviews);

const company = companies.find((item) => item.employerId === interviews[0].employerId);

To start the Actor and read the result in one script, call client.actor("<ACTOR_ID>").call(input). The run it returns has the same storageIds.

Breaking change. Interviews, salaries, benefit reviews, and jobs are no longer arrays inside the company record. The fields interviews, salaries, benefitReviews, and jobs are gone from that record. Read them from their own Datasets. Join each one to the default Dataset on employerId. Photos, company updates, and all other sections stay in the company record.

Each item in the four Datasets starts with employerId and employerName, followed by the fields of the interview, salary, benefit review, or job. The items of one employer are written in the order of collection. They are not sorted.

rowsWritten in the company record counts the items written for each enabled section. Compare it with the reported totals:

• salaryMeta.resultCount is the number of salary estimates that Glassdoor reports.
• interviewStats.filteredInterviewCount is the number of interviews that Glassdoor reports.
• benefitsTotalReviews is the number of benefit reviews that Glassdoor reports.

The Actor writes each interview once, even when Glassdoor shows it in more than one list. If a run stops while it collects an employer, the four Datasets can hold items of an employer that has no company record. A run that you resume can write those items again. Keep only the items whose employerId is in the default Dataset. Remove repeats by id for interviews and benefit reviews, and by job.listingId for jobs. Salary items have no ID.

Input reference

Provide at least one non-empty urls, employerIds, or searchQueries list. The Actor resolves duplicate inputs to one company record.

Complete input example

Enable only the sections that you need. The limits keep the optional arrays short.

{
  "urls": [
    "https://www.glassdoor.com/Overview/Working-at-Nucor-EI_IE489.htm"
  ],
  "employerIds": [
    489
  ],
  "searchQueries": [
    "https://www.nucor.com"
  ],
  "maxSearchResults": 1,
  "strictDomainMatch": true,
  "compactMode": false,
  "minReviewCount": 1,
  "minOverallRating": 1,
  "includeRatings": true,
  "includeDemographics": true,
  "includeReviewLocations": true,
  "includeOfficeLocations": true,
  "includeManagedContent": true,
  "includeCompanyUpdates": true,
  "includePhotos": true,
  "maxPhotos": 100,
  "includeSalaries": true,
  "maxSalaries": 21,
  "includeBenefits": true,
  "maxBenefitReviews": 21,
  "benefitsCountryId": 1,
  "includeInterviews": true,
  "maxInterviews": 21,
  "includeJobs": true,
  "maxJobs": 21,
  "proxyConfiguration": {
    "useApifyProxy": true
  }
}

For a normal run, use only one input method. Compact mode ignores all section switches.

Field reference

Field	Type	Default	Description
urls	string[]	[]	Company page URLs. Overview, reviews, salaries, and other company-page URLs containing an employer ID are accepted.
employerIds	integer[]	[]	Employer IDs from 1 through 2,147,483,647.
searchQueries	string[]	[]	Company names or company website addresses.
maxSearchResults	integer	20	Requested maximum matching companies per search query. A query can return fewer.
strictDomainMatch	boolean	true	For website searches, keep only results whose company website uses the same registered domain.
compactMode	boolean	false	Return a flat company row instead of the full nested record.
minReviewCount	integer	0	Skip companies with fewer reviews. 0 disables this filter.
minOverallRating	number	0	Skip companies below this rating. Accepted range is 0 through 5; 0 disables this filter.
includeRatings	boolean	true	Include the detailed ratings section and distributions. The overall rating and review count remain in the core profile.
includeDemographics	boolean	true	Include ratings grouped by demographic category.
includeReviewLocations	boolean	true	Include cities and metro areas represented in reviews.
includeOfficeLocations	boolean	true	Include company office locations.
includeManagedContent	boolean	true	Include company-managed profile content.
includeCompanyUpdates	boolean	true	Include up to 400 recent company updates.
includePhotos	boolean	true	Include company photos.
maxPhotos	integer	100	Maximum photo entries per company. 0 collects all available photos.
includeSalaries	boolean	false	Include salary estimates by job title as items of the salaries Dataset.
maxSalaries	integer	20	Maximum salary items per company. 0 collects all that Glassdoor serves, up to 10,000.
includeBenefits	boolean	false	Include the benefits overview in the company record and benefit reviews as items of the benefitReviews Dataset.
maxBenefitReviews	integer	20	Maximum benefit review items per company. 0 collects all available reviews.
benefitsCountryId	integer	1	Country used for benefits data. When benefits are enabled and available, benefitCountries lists the supported countries and IDs.
includeInterviews	boolean	false	Include interview experiences as items of the interviews Dataset and aggregate statistics in the company record.
maxInterviews	integer	20	Maximum interview items per company. 0 collects all available interviews.
includeJobs	boolean	false	Include current jobs as items of the jobs Dataset.
maxJobs	integer	20	Maximum job items per company. 0 collects all available jobs.
proxyConfiguration	object	{"useApifyProxy":true}	Optional connection settings for the run.

Company record reference

The default Dataset contains one record for each included employer. The interviews, salaries, benefit reviews, and jobs of an employer are items of their own Datasets, described in Output datasets.

Mode	Shape	Section behavior
Full (default)	Nested company profile	Includes the enabled default and optional sections
Compact	Flat core-company row	Ignores section switches such as includeSalaries and includeJobs

A filtered company, a missing company, or a search with no match produces no Dataset record.

The examples shorten long arrays to one item. Empty arrays and null values keep their original meaning.

Compact record

This compact-field projection contains every documented compact key. Its scrapedAt value dates the point-in-time record to 16 August 2026.

{
  "recordType": "company",
  "mode": "compact",
  "status": "complete",
  "failedSections": [],
  "employerId": 489,
  "inputSource": "https://www.nucor.com",
  "glassdoorUrl": "https://www.glassdoor.com/Overview/Working-at-Nucor-EI_IE489.11,16.htm",
  "name": "Nucor Corporation",
  "overallRating": 3.9,
  "reviewCount": 876,
  "revenue": "$10+ billion (USD)",
  "size": "10000+ Employees",
  "type": "Company - Public",
  "yearFounded": 1971,
  "website": "https://www.nucor.com",
  "headquarters": "Charlotte, NC",
  "ceoName": "Leon Topalian",
  "industry": "Metal & Mineral Manufacturing",
  "sector": "Manufacturing",
  "description": "Nucor continues to electrify the steel industry with a simple concept: The minimill is mighty. At its various minimills, Nucor produces about 15 million tons of steel annually, including hot- and cold-rolled steel, steel joists, and metal buildings. North America's largest recycler of scrap metal, Nucor produces steel by melting scrap in electric arc furnaces. Most of its products are sold to steel service centers, manufacturers, and fabricators. Subsidiaries such as Vulcraft -- one of the US's largest producers of steel joists, girders, and decking -- use the balance of the steel. The company has grown mainly through acquisition since 2006, including spending $1 billion to buy The David J. Joseph Company.",
  "scrapedAt": "2026-08-16T15:12:05.665Z"
}

Full record with optional sections

This field-shape projection contains point-in-time values dated 16 August 2026. It includes the salary, benefit, interview, and job summaries and the row counts. The rows themselves are in their own Datasets. Long arrays keep one item, so the JSON is not a complete full record.

Interview, salary, benefit review, and job items

Each item in the interviews, salaries, benefitReviews, and jobs Datasets starts with employerId and employerName. The other fields follow. These examples are field-shape projections with values dated 16 August 2026. Long text is kept as Glassdoor returns it.

Common and compact fields

Output field	JSON type and nullability
recordType	string; always company
mode	string; full or compact
status	string; complete or partial
failedSections	string[]; empty when status is complete
employerId	number
inputSource	string or null
overallRating, reviewCount	number or null; present in full and compact records
scrapedAt	ISO 8601 string
Compact: glassdoorUrl, name, revenue, size, type, website, headquarters, ceoName, industry, sector, description	string or null
Compact: yearFounded	number or null

Full profile fields

Full mode includes the current Glassdoor profile fields and adds the selected sections. Glassdoor can add child keys. In the child tables below, the type is the JSON type of a non-null value. A child field can be absent or null. Arrays can be empty. The Actor keeps new keys that Glassdoor adds.

The top-level section branches have these contracts:

Branch	JSON type	Presence
ratings, ratingCountDistribution	object	Optional; controlled by includeRatings; added together.
filteredReviewsCountByLang	object[]	Optional; added with ratings.
allReviewsCount, filteredReviewsCount, ratedReviewsCount	number	Optional; added with ratings.
demographicRatings	object[] or null	Optional; controlled by includeDemographics.
reviewLocations	object[]	Optional; controlled by includeReviewLocations.
officeLocations	object[]	Optional; controlled by includeOfficeLocations.
employerManagedContent	object[]	Optional; controlled by includeManagedContent.
companyUpdates	object[]	Optional; controlled by includeCompanyUpdates.
companyUpdatesTotalCount	number	Optional; added with companyUpdates.
photos	object[]	Optional; controlled by includePhotos.
salaryMeta	object	Optional; controlled by includeSalaries, when salary data is available.
benefitCountries	object[]	Optional; controlled by includeBenefits.
benefits	object[] or null	Optional; controlled by includeBenefits.
benefitsOverallRating, benefitsTotalReviews	number or null	Optional; added when a benefit summary is available.
benefitEmploymentStatuses	string[]	Optional; added with a benefit summary.
interviewStats	object	Optional; controlled by includeInterviews, when interview data is available.
jobsCompanyFilterOptions	object[]	Optional; controlled by includeJobs.
rowsWritten	object	Optional; present when includeSalaries, includeBenefits, includeInterviews, or includeJobs is on. It has one number for each enabled section: salaries, benefitReviews, interviews, and jobs. Each number counts the items written to the Dataset of that name.

If a section fails at the start, its branch can be absent. If it fails later, companyUpdates and photos can keep the items collected before the failure, and the four row Datasets keep the items written before the failure. The record then has status: "partial", and failedSections names the incomplete section.

Profile and relationships

The profile can contain these containers:

Field	JSON type
primaryIndustry, bestProfile, overview, ceo, counts, links, coverPhoto	object
parent	object
officeAddresses, awards, bestPlacesToWork, bestLedCompanies, subsidiaries, siblings	object[]

name is a string and is present in every saved record. Other profile scalars use these JSON types when present:

Scope	Number fields	String fields	Boolean fields
Profile root	id, yearFounded, primaryIndustryId	shortName, headquarters, revenue, size, sizeCategory, website, type, activeStatus, approvalStatus, squareLogoUrl, regularLogoUrl, largeLogoUrl	isValid
primaryIndustry	industryId, sectorId	industryName, sectorName	—
bestProfile	id	—	—
overview	—	description, mission	—
ceo	id	name, title, photoUrl	—
counts	benefitCount, reviewCount, salaryCount, interviewCount, photoCount, globalJobCount.jobCount	—	—
links	—	benefitsUrl, interviewUrl, jobsUrl, overviewUrl, reviewsUrl, salariesUrl	—
coverPhoto	—	hiResUrl	—
officeAddresses[]	id, officeLocationId, country.id, country.continent.id	addressLine1, addressLine2, cityName, countryName, postalCode, administrativeAreaName1, country.name, country.continent.name	—
awards[]	year	name, source	featured
bestPlacesToWork[]	id, rank	listType, name, timePeriod	isCurrent
bestLedCompanies[]	id, rank	name, timePeriod	isCurrent

parent.employer, subsidiaries[].employer, and siblings[].employer use the same relationship shape:

Number fields	String fields	Boolean fields
id, ratings.overallRating, counts.reviewCount, counts.salaryCount, counts.globalJobCount.jobCount	name, shortName, headquarters, squareLogoUrl, approvalStatus	isValid

Ratings, demographics, and locations

Scope	Number fields	String fields	Other fields
ratings	overallRating, reviewCount, ceoRating, ceoRatingsCount, recommendToFriendRating, businessOutlookRating, cultureAndValuesRating, careerOpportunitiesRating, workLifeBalanceRating, seniorManagementRating, compensationAndBenefitsRating, diversityAndInclusionRating	—	ratedCeo: object or null
ratings.ratedCeo	id	name, title, regularPhoto, largePhoto	—
ratingCountDistribution	Values under 1, 2, 3, 4, and 5; recommendToFriend.RECOMMEND; recommendToFriend.WONT_RECOMMEND	—	overall, cultureAndValues, careerOpportunities, workLifeBalance, seniorManagement, compensationAndBenefits, diversityAndInclusion, and recommendToFriend are objects.
filteredReviewsCountByLang[]	count	isoLanguage	—
Rating counts at the record root	allReviewsCount, filteredReviewsCount, ratedReviewsCount	—	—
demographicRatings[]	—	category	categoryRatings: object[]
demographicRatings[].categoryRatings[]	Rating fields listed for ratings, including reviewCount	categoryValue	ratings: object
reviewLocations[]	id	name, type; state and country are string or null	—
officeLocations[]	id, officeLocationId, country.id, country.continent.id, employerReviews.ratings.overallRating	addressLine1, addressLine2, cityName, countryName, postalCode, administrativeAreaName1, country.name, country.continent.name	country and employerReviews: object

Managed content, updates, and photos

Scope	Number fields	String fields	Boolean fields and containers
employerManagedContent[]	—	—	managedContent: object[]
managedContent[]	id	body, type, title	media: object[]
managedContent[].media[]	—	caption, source, mediaType	—
companyUpdates[]	employerId, likes; id can be a number or string	text, date	isLiked: boolean; media and category: object
companyUpdates[].media	—	title, description, linkUrl, linkDomain, imageUrl, videoUrl	—
companyUpdates[].category	id	name	—
photos[]	photoId	caption, photoUrl2x	—

Salaries and benefits

The salary item and benefitReview item scopes are items of the salaries and benefitReviews Datasets. The other scopes are fields of the company record.

Scope	Number fields	String fields	Boolean fields and containers
salary item	employerId, basePayStatistics.mean, totalAdditionalPayStatistics.mean	employerName, payPeriod	currency, employer, jobTitle, basePayStatistics, totalAdditionalPayStatistics, and totalPayStatistics: object
salary item: currency	id	code	—
salary item: employer	id, ratings.overallRating, counts.globalJobCount.jobCount	name, shortName, squareLogoUrl	ratings and counts: object
salary item: jobTitle	id, gocId, mgocId	text	—
salary item: totalPayStatistics.percentiles[]	value	ident	—
salaryMeta	resultCount, numPages	—	queryLocation: object or null; truncatedByWindow: boolean
salaryMeta.queryLocation	id	name, type	—
benefitCountries[]	id	name	—
benefits[]	rating can be a number or null; reviewCount	category, name	verified: boolean
Benefit summary at the record root	benefitsOverallRating and benefitsTotalReviews can be a number or null	—	benefitEmploymentStatuses: string[]
benefitReview item	employerId, id, rating	employerName, createDate, userEnteredJobTitle	currentJob: boolean; city: object; benefitComments: object[]
benefitReview item: city	—	name, state.name, metro.name, country.name	state, metro, and country: object
benefitReview item: benefitComments[]	helpfulVotes, id	comment	—

Interviews

The interview item scopes are items of the interviews Dataset. interviewStats is a field of the company record.

Scope	Number fields	String fields	Boolean fields and containers
interview item	employerId, id, countHelpful	employerName, advice, difficulty, experience, negotiationDescription, outcome, processDescription, reviewDateTime, source	featured: boolean; employer, jobTitle: object; employerResponses, userQuestions: object[]
interview item: employer	—	name, squareLogoUrl	—
interview item: employerResponses[]	—	response, responseDateTime	—
interview item: jobTitle	—	text	—
interview item: userQuestions[]	answerCount	question	—
interviewStats	difficultySubmissionCount, difficultySum, filteredInterviewCount, totalInterviewCount	—	interviewExperienceCounts, interviewObtainedChannelCounts: object[]; truncatedByWindow: boolean
interviewStats.interviewExperienceCounts[], interviewStats.interviewObtainedChannelCounts[]	count, percentage	type	—

Jobs

The job item scopes are items of the jobs Dataset. A job item can contain job, header, and overview objects. A listing that Glassdoor returns without details is an item with only employerId and employerName. jobsCompanyFilterOptions is a field of the company record.

Scope	Number fields	String fields	Boolean fields and containers
job item	employerId	employerName	job, header, overview: object
job item: job	listingId	jobTitleText, description	—
job item: header	ageInDays, categoryMgocId, sgocId, locId, rating, jobCountryId	goc, normalizedJobTitle, locationName, locationType, employerNameFromSearch, payPeriod, salarySource, payCurrency, jobViewUrl, jobLink, applyUrl, serpUrlForJobListing	easyApply, expired, isSponsoredJob, isSponsoredEmployer: boolean; employer, coverPhoto, payPeriodAdjustedPay, urgencySignal, indeedJobAttribute: object or null; occupations: object[]; remoteWorkTypes: string[] or null
job item: header.employer	id	name, shortName, squareLogoUrl	—
job item: header.coverPhoto	employerId, sgocId	url	—
job item: header.payPeriodAdjustedPay	p10, p50, p90	—	—
job item: header.occupations[]	—	key	—
job item: header.urgencySignal	normalizedCount	labelKey, messageKey	—
job item: header.indeedJobAttribute.extractedJobAttributes[]	—	key, value	—
job item: overview.primaryIndustry	industryId, sectorId	industryName, sectorName	—
jobsCompanyFilterOptions[]	id	name, squareLogoUrl, shortName	—

Company resolution and result behaviour

• A company or section can be unavailable for a country or employer. Missing optional objects can be null, empty, or absent.
• A full record has status: partial when a requested section does not complete. failedSections can contain ratings, demographicRatings, reviewLocations, officeLocations, employerManagedContent, companyUpdates, photos, benefits, benefitReviews, salaries, interviews, or jobs. Check this list before you use an optional section.
• Glassdoor can add or remove nullable child fields. Use the field reference as a guide and allow new keys in your data pipeline.
• A company that does not match the rating or review filters produces no Dataset item. A missing company or a search with no match also produces no item.
• A search term can match more than one employer. Use maxSearchResults and strictDomainMatch to limit website searches.
• A maximum value of 0 collects all available entries and can increase the result size and run time for large employers.
• Glassdoor serves only the first 10,000 salary estimates of a company. For a larger company, the Actor collects 10,000 salary items and stops. No filter or sort order reaches the rest. This stop is not a failure, so the record keeps status: complete. salaryMeta.truncatedByWindow is true when Glassdoor reports more salaries than the Actor can read. The reported total is salaryMeta.resultCount. Some salary items can be exact repeats of each other, because the Actor cannot tell them apart.
• For interviews, the Actor combines several views of the list of a company and keeps each interview once. It reads all interviews of most companies, including companies with more than 10,000 interviews. A very large list can hold a few interviews that no view reaches. Then interviewStats.truncatedByWindow is true. Compare rowsWritten.interviews with interviewStats.filteredInterviewCount. A run that reads more interviews takes more time.
• Each job appears once in the jobs Dataset, even when Glassdoor lists it on more than one page, so rowsWritten.jobs can be lower than the number of jobs that Glassdoor reports.