kununu Employer Data

Paste kununu profiles, get one row per employer: score vs. industry, category ratings, salary ranges per job title, benefits and reviews. kununu has no public API; this reads the public profile pages at the pace kununu accepts and stops cleanly when refused.

Limits: employers on kununu in Germany, Austria and Switzerland only. Without a login kununu shows 10 reviews per view, so large companies cannot be read completely – every row says how many reviews kununu has and how many were delivered. kununu's bot protection can tighten without warning; the run then stops cleanly and lists what is left.

How to use – step by step

1. Click Try for free on this page (a free Apify account is enough) – Siemens is prefilled.
2. Paste kununu profile URLs or slugs into Companies – sap, or at/red-bull for an Austrian profile.
3. Optional: keep Include salary data on, switch on Include individual reviews and pick the Review type and Maximum reviews per company; choose the Output format (one row per company, or one per review).
4. Click Start. Three companies without salary pages took 13.9 seconds in a local test run on 12 October 2026; 100 reviews of one company take about a minute.
5. Open the Output tab: the Employers view shows score, industry average, recommendation rate, reviews, employer response rate and check time. Export as CSV, Excel or JSON, or read the rows through the API. Companies kununu does not know are listed under NOT_FOUND in the key-value store – not charged.

Built to finish: how it handles kununu's bot protection

kununu sits behind AWS WAF. Plain HTTP clients get an interactive CAPTCHA on every page; a real browser gets an automatic JavaScript challenge, which this Actor clears in about 1.5 seconds — from Apify's own machines, no proxy.

kununu also limits how fast one network address may read its pages. Read too quickly and it answers with a CAPTCHA; press on and it blocks the address entirely. So the Actor:

• reads one page at a time at a steady pace (about 14 pages a minute),
• loads only the page itself — none of kununu's trackers, consent or ad scripts, which also halves the run time,
• treats a CAPTCHA or 403 as "slow down": it pauses (60 s, then 120 s, then 240 s — as far as the run's remaining time allows), doubles the gap between pages and only speeds up again after a run of good pages, in the same session,
• and if kununu keeps refusing, stops cleanly: everything finished is delivered, nothing else is charged, and the companies still to do are listed under REMAINING in the run's key-value store, ready to paste into the next run.

Measured on 6 October 2026: 21 test runs, 21 succeeded — including two parallel runs with more than 50 page loads each, which never met a CAPTCHA. On 12 October 2026 a run that met three refusals in a row paused 60 + 120 + 240 s and then finished normally instead of failing.

Too little CPU looks like a block. With a hard limit of a quarter CPU core the browser drew HTTP 403 in 4 of 4 local runs; at 2 GB memory (Apify assigns CPU in proportion) it does not. Keep the default 2 GB.

What makes this actor different

It returns the data an HR-tech product or an employer-branding analysis needs:

• ⭐ Category scores with sub-factors — not just "3.8", but career (salary, image, career development), culture (atmosphere, communication, teamwork, management…), work environment, diversity — each with its own review count
• 💶 Salary ranges per job role — min, max, median and average per position, with the number of entries behind each figure, plus the salary-satisfaction split and how compensation compares to the industry
• 🎁 Benefits with confirmation rates — "68 % of reviewers confirm a canteen"
• 📈 Industry benchmark built in — industryAverageScore and the ready-made scoreVsIndustry delta, so a company is instantly comparable to its sector
• 💬 Employer responsiveness — response rate and average response time in days: a strong signal for employer-branding sales
• 🗣 Reviews in full depth — positive / negative / suggestion text, plus each factor rating with its individual comment, and the employer's replies — for employees, applicants (interview experiences), apprentices, interns and working students
• 🧾 Proof on every row — the input as you gave it, status, checkedAt, the pages read (sourceUrls) and the coverage: how many reviews and salary ranges kununu has and how many were delivered
• 🇩🇪🇦🇹🇨🇭 All three DACH countries in one actor

Use cases

• HR tech & employer branding — enrich company records with employer scores, benefits and salary bands
• Competitive talent research — how does your employer rating compare to the companies you compete with for candidates?
• Recruiting & market research — salary benchmarks per job role and region
• Sentiment analysis — thousands of structured review texts, ready for NLP
• Lead scoring for HR vendors — low score + low response rate + unclaimed profile = a company with an employer-branding problem
• AI agents — clean schema, callable via MCP

Input example

{
    "companies": [
        "https://www.kununu.com/de/siemens",
        "sap",
        "at/red-bull"
    ],
    "country": "de",
    "include_salary": true,
    "include_reviews": true,
    "review_type": "employee",
    "max_reviews_per_company": 100,
    "output_type": "company"
}

Companies can be full kununu URLs or plain slugs. Prefix a slug with a country (at/red-bull) to mix markets in one run.

Output example (shortened)

{
    "companyName": "Siemens AG",
    "profileUrl": "https://www.kununu.com/de/siemens",
    "score": 3.8,
    "industryAverageScore": 3.4,
    "scoreVsIndustry": 0.4,
    "recommendationRate": 85,
    "totalReviews": 6151,
    "totalReviewsEmployees": 5813,
    "employerResponseRate": 11,
    "employerResponseTimeDays": 55,
    "firstReviewYear": 2007,
    "isClaimed": false,
    "locationMain": { "city": "München", "state": "Bayern", "countryCode": "de" },
    "locationsTotal": 136,
    "scoreBreakdown": [
        {
            "category": "career",
            "score": 3.9,
            "totalReviews": 5511,
            "factors": [
                { "factor": "salary", "score": 4.1 },
                { "factor": "image", "score": 3.9 },
                { "factor": "career", "score": 3.7 }
            ]
        }
    ],
    "benefits": [
        { "benefit": "cantine", "confirmedBy": 3930, "percentage": 68 },
        { "benefit": "homeOffice", "confirmedBy": 3173, "percentage": 55 }
    ],
    "salaryRanges": [
        {
            "jobTitle": "Projektmanager:in",
            "salaryMin": 41300,
            "salaryMax": 117600,
            "salaryMedian": 75000,
            "salaryAverage": 78900,
            "basedOnEntries": 223
        }
    ],
    "salaryRangesDelivered": 20,
    "salaryRangesAvailable": 345,
    "salarySatisfaction": { "positivePercentage": 76, "neutralPercentage": 15, "negativePercentage": 9 },
    "compensationVsIndustryPercent": 21,
    "reviewsAvailable": 5904,
    "reviewsScraped": 100,
    "reviewsStatus": "ok",
    "reviews": [
        {
            "title": "Meine Erfahrung als langjähriger Mitarbeiter",
            "score": 4.5,
            "recommended": true,
            "createdAt": "2026-07-30T00:00:00+00:00",
            "positionGroup": "student",
            "department": "it",
            "positiveText": "Alle Benefits und die Kollegen …",
            "negativeText": "…",
            "suggestionText": "…",
            "factorRatings": [
                { "factor": "atmosphere", "score": 5, "comment": "Sehr lockere Atmosphäre" }
            ],
            "employerResponses": []
        }
    ],
    "input": "https://www.kununu.com/de/siemens",
    "status": "found",
    "checkedAt": "2026-10-12T13:27:49Z",
    "sourceUrls": ["https://www.kununu.com/de/siemens", "https://www.kununu.com/de/siemens/gehalt"]
}

Prefer a flat table? Set output_type: "reviews" to get one row per review with the company data attached — ideal for Excel, Google Sheets or a database import.

Performance & cost

The actor opens one browser session per run and reuses it for every page. To stay within kununu's limits it reads about one page every 4 seconds, measured:

Job	Pages	Run time
1 company, profile + salary (default input)	3	~10–25 s
25 companies, profile + salary	51	~3.3 min
30 reviews of one company	4 views	~22 s (measured 12 Oct 2026)
100 reviews of one company	12 views	~53 s (measured 12 Oct 2026)

How reviews are reached. kununu shows visitors who are not logged in only the first 10 reviews of any list — page 2 onwards and the "more reviews" button need a login (measured 12 Oct 2026: ?page=2 and /kommentare/2 return page 1). The actor never logs in. It reads kununu's own views instead — newest, oldest, best, worst, then by year, job status, department and so on — 10 reviews each, and keeps every review exactly once (by its id). It stops when max_reviews_per_company is reached or when the views bring nothing new. Each company row reports reviewsAvailable, reviewsScraped, reviewsViewsLoaded and reviewsStatus (ok, views_exhausted, partial, stopped, not_available, robots_first_view_only), so you always see the coverage. Turn include_salary and include_reviews off for the fastest company-level scan.

Rating factor reference

Rating keys are locale-independent — the same keys come back for Germany, Austria and Switzerland. They are kununu's own keys and kununu does change them now and then (checked 12 Oct 2026; earlier management, colleagues, interestingTasks are now leadership, workLife, tasks).

Category	Factors
career	salary, image, career
culture	atmosphere, communication, teamwork, workLife, leadership, tasks
work_environment	workConditions, environment
diversity	equality, oldColleagues

Limitations – honestly stated

• The bot protection can tighten without warning. kununu changed its answer to browsers twice in September 2026. The Actor paces itself, pauses when asked to and never tries to trick the protection — if kununu refuses for good, the run stops with what it has.

• kununu publishes no official API; this actor reads the public website, so a site redesign can require an update.

• Reviews: 10 per view, no login. Large companies cannot be read completely — the actor reaches as many distinct reviews as kununu's views show and says how many that was (reviewsAvailable vs reviewsScraped).

• Salary ranges: the top job titles. kununu's salary page lists about 20 job titles; salaryRangesAvailable says how many it has (Siemens: 20 of 345).

• Locations: kununu publishes the main location and the number of locations (locationMain, locationsTotal), not the full list.

• kununu forwards some addresses to another profile (e.g. basf → Master Builders Solutions). redirected: true and input show when that happened.

• Review texts are in German, as written by reviewers. Field names, rating keys and all metadata are English.

• If kununu refuses part way through, the run stops cleanly: finished companies are delivered and charged, the rest are listed under REMAINING. A run that could not load a single page fails visibly instead of writing empty rows — a silent empty dataset is worse than an error.

• Not charged: companies that do not exist on kununu (NOT_FOUND), that could not be loaded (FAILED), duplicates in the input, and — in the one-row-per-review layout — companies without reviews of the chosen type (NO_REVIEWS). All are listed in the key-value store and the status message.

Data protection

Only public, aggregate and anonymous data is collected:

• kununu reviews are anonymous by design — no reviewer identity exists on the platform and none is extracted. No user profiles are visited.
• No individual salary entries tied to a person — only the aggregated ranges kununu publishes, each with its sample size (basedOnEntries).
• No login-protected content, no employer dashboards.
• robots.txt: kununu allows the first review page per company and disallows sorted and paginated review lists. By default the actor reads kununu's sorted and filtered review views; set respect_robots_txt: true to stay strictly within robots.txt (first 10 reviews per company only).

Not affiliated with or endorsed by kununu GmbH or New Work SE.

Changelog

• 12 October 2026 – Duplicate inputs are read once and charged once; companies kununu does not know are listed under NOT_FOUND and not charged; every row carries the input, status, checkedAt, the pages read and the review and salary coverage. Reviews come from kununu's own views without a login, each review once; respect_robots_txt keeps to the first 10 reviews.
• 6 October 2026 – Paced to kununu's limits: one page at a time, pauses on a CAPTCHA, stops cleanly and lists the companies still to do under REMAINING (21 of 21 test runs succeeded).

Auf Deutsch: kununu-Daten exportieren

kununu-Profile rein, je Arbeitgeber eine Zeile raus: Score mit Branchenvergleich, Bewertungen je Kategorie, Gehaltsspannen je Jobtitel, Benefits und auf Wunsch einzelne Bewertungen. kununu hat keine öffentliche API; gelesen werden die öffentlichen Profilseiten in dem Tempo, das kununu zulässt.

Enthalten sind der kununu-Score inklusive Branchenvergleich, die Bewertungen je Kategorie mit allen Unterfaktoren, bestätigte Benefits mit Prozentwerten, Gehaltsspannen je Jobtitel (Minimum, Maximum, Median, Durchschnitt und Anzahl der Angaben), die Gehaltszufriedenheit sowie auf Wunsch einzelne Bewertungen mit Volltext, Faktor-Bewertungen samt Kommentar und den Antworten des Arbeitgebers.

Typische Anwendungsfälle: Employer-Branding-Monitoring, Wettbewerbsanalyse im Recruiting, Gehaltsbenchmarks, Anreicherung von CRM- oder Bewerberdaten sowie Sentiment-Analysen. Export als JSON, CSV oder Excel — oder direkt per API.

FAQ

Is there an official kununu API? No. kununu offers no public API or data feed; this actor reads the public profile pages.

Do I need proxies? No. The actor clears kununu's automatic challenge itself and keeps to the pace kununu accepts. The proxy input is there for unusual network setups only.

How do I get the data into Excel or Google Sheets? Run the actor, then export the dataset as CSV or XLSX from the Apify console — or use output_type: "reviews" for a flat one-row-per-review table.

Can I monitor employers continuously? Yes. Schedule the actor (daily, weekly) with the companies you care about and send the dataset to a webhook, Google Sheets, Slack or your database.

What does a run cost?

What you export	Cost
1 company, company level (default input)	$0.041
100 companies, company level	$0.14
1,000 companies, company level	$1.04
1 company, 1,000 reviews as rows (output_type: "reviews")	$1.04
100 companies, 100 reviews each as rows	$10.04

Free plan prices: the start fee is $0.02 per GB of memory — at the default 2 GB that is $0.04 per run — plus $0.001 per record (lower on paid plans, see the Pricing tab). In output_type: "company" one company is one record, reviews nested inside it; in "reviews" each review is a record. Not found, failed or duplicate companies cost nothing.

More from Perceptron Data

Actor	What it does
Advertiser Check (Google & Bing)	Company list in, one row per company out: Google ads, Bing ads, ad counts, accounts
Imprint & VAT Check	Domains in, company rows out: legal name, address, register number and EU VAT ID checked against VIES (Swiss UIDs excluded)
Free Imprint & Contact Check	The same company row, free, for up to 10 websites per run
PDF to Markdown	PDF, Word, Excel and PowerPoint to Markdown for AI – tables as JSON, OCR for scans
Research Agent	A question in, cited passages out – every source graded by evidence level
App Store & Google Play Data	App IDs or App Store keywords in, one row per app out, plus App Store reviews
Arbeitsagentur Jobs	German job listings from the Jobbörse of the Bundesagentur für Arbeit
Messe München Exhibitors	Exhibitor lists of bauma, IFAT, BAU and more
Koelnmesse Exhibitors	Exhibitor lists of Anuga, ISM, imm cologne and more
TED Tenders	EU public tenders and contract awards from TED
MaStR Energy Assets	Solar, wind and storage units in Germany with operator and capacity
EUDAMED Medical Devices	EU medical devices, manufacturers and certificates
Data Source Finder	Free: tells you which ready-made scraper covers your data source

Support

Bug reports are welcome in the Issues tab.