Catho Jobs Scraper

Catho Jobs Scraper: Brazilian job listings with salary, contract, benefits and the full description

Type what you would type into Catho's own search box, analista de dados or motorista, add a city or a state if you want one, and every matching job comes back as a row: title, company, city and state, the salary Catho shows, the contract, the benefits, the full description and when it went up. No Catho account and no cookies. Two things catch people out: most employers publish a salary band rather than a number, and Catho only reveals a job's work model through its own filter, so workModel and homeOffice stay empty unless you filter on one.

🔍 What Catho Jobs Scraper does

Each line of search words is its own Catho search, and a job found by two of them is kept once. With Full job details on, the default, it opens each job's own record for the description, contract, benefits, salary band and dates. Off, you get only what the search card shows, which reads about six times less.

A city needs its two-letter state: São Paulo, SP. Catho has more than one city of most names and picks a small one if the state is missing, so a bare city is refused before anything is searched. Catho also answers a place it does not recognise with jobs from the whole of Brazil, so the run checks which place Catho actually applied, and if it was not yours, that search stops with a free row instead of a nationwide crawl you did not ask for.

The home office point, plainly: Catho does not put a work model on the job card, in the job's own record or in the page markup. The only place it exists is Catho's own Modalidade filter. Set workModel to remote and every row is a job Catho files as home office, with homeOffice: true on it. Leave it on any and both fields come back null, because guessing from the description would be making it up.

Salary is mostly a band. Across 120 jobs from a broad nursing search, salaryBand was filled on every row and salaryMin, an actual figure, on 26%. Catho shows a signed-out visitor the same salary it shows everyone else; often there is just no number to show.

📥 What you give it

{
  "searches": ["analista de dados", "cientista de dados"],
  "city": "São Paulo, SP",
  "workModel": "remote",
  "sortBy": "recent",
  "maxResults": 200,
  "includeDescription": true
}

With no search words at all, the run returns one sample job from Catho's own list, labelled and free.

A city without its state stops the run before anything is searched, with a message saying what to fix.

📤 What you get back

A real row from a run on 30 September 2026, with the description cut short:

{
  "jobId": "38566637",
  "title": "Analista de Dados",
  "company": "RANDSTAD SP",
  "confidentialCompany": false,
  "city": "São Paulo",
  "state": "SP",
  "stateName": "São Paulo",
  "country": "Brasil",
  "otherLocations": [],
  "seniority": null,
  "workModel": null,
  "homeOffice": null,
  "salaryText": "R$ 4.000",
  "salaryMin": 4000,
  "salaryMax": null,
  "salaryBand": "A partir de R$ 3.000,00",
  "salaryCurrency": "BRL",
  "contractType": "CLT (Efetivo)",
  "contractTypes": ["CLT (Efetivo)"],
  "workSchedule": "Período Integral",
  "benefits": [],
  "category": "Informática",
  "subCategory": "Informática /TI / Engenharia da Computação",
  "roles": ["Analista de Banco de Dados"],
  "openings": 1,
  "sponsored": false,
  "hasScreeningQuestions": false,
  "screeningQuestionCount": 0,
  "description": "A Randstad é líder global em soluções de Recursos Humanos, atuando como um parceiro do talento em toda a sua jornada profissional...",
  "postedAt": "2026-09-25T16:32:19",
  "postedText": "Publicada em 25/09",
  "updatedAt": "2026-09-25T16:32:19",
  "expiresAt": "2026-12-24T16:32:19",
  "url": "https://www.catho.com.br/vagas/analista-de-dados/38566637",
  "searchTerm": "analista de dados",
  "detailsLoaded": true,
  "scrapedAt": "2026-09-30T02:24:49.966Z"
}

🧾 Reading the output

Real jobs carry no charged field. The sample is a real job from Catho's list, so drop rows with _sample or _diagnostic rather than looking at the job itself.

A search with no jobs writes nothing to the dataset and costs nothing; the status message names it. Every run also writes a RUN_REPORT record to its key-value store with, per search, the jobs Catho claimed, pages read, jobs returned, repeats skipped, jobs without details, and why it stopped.

▶️ How to run it

1. Open Catho Jobs Scraper and click Try for free. 2. Type one search per line into Search words. 3. Add a City with its state, a State, or a Work model if you want one. 4. Set Maximum jobs, and pick Order if you want the newest first. 5. Click Start, then download the dataset as JSON, CSV or Excel.

The form arrives with analista de dados filled in and a cap of 100, so a run started without editing it is a real search.

💰 How much does it cost?

$1.40 per 1,000 jobs, flat on every Apify plan, with no volume tiers. That is $0.0014 a job.

Not charged: the sample job, a search that finds nothing, a job an earlier search in the same run already delivered, a job whose full record did not load when you asked for details, and a search stopped because Catho did not recognise the place. If you set a maximum charge on the run, it is read before the first request and the run stops at it.

💡 What people use it for

• A morning feed of new postings: sortBy set to recent, a modest maxResults, and jobId to skip what you had

yesterday.

• A home-office job list, through workModel set to remote, which is the only honest way to get one from Catho.
• Reading salary bands for a role across states, from salaryBand and salaryMin.
• Seeing which employers in a field post as "Empresa Confidencial".

🚧 What it does not do

• It does not invent a work model. Null wherever Catho says nothing.
• Catho's results count is not what you can get. Its page claimed 8,782 results for desenvolvedor; paging end to

end gave 763 distinct jobs. The run stops after two pages with nothing new rather than paying for repeats.

• No other filters. Catho's contract, salary-band and posting-date filters have no working form, fourteen shapes

were tried and every one was ignored, so they are not offered. Filter contractType and postedAt on the rows.

• seniority is only there when the title says so: about a fifth of jobs on a broad search, half on developer

searches.

• No résumés, candidates or employer contact details. Job listings only.

🧭 Which jobs scraper do you need?

❓ Questions people ask

How do I get only home-office jobs? Set workModel to remote. That is Catho's own Modalidade filter.

How many jobs can I pull for one search term? As many as Catho will page: 763 for desenvolvedor, over 880 for enfermeiro. The number Catho prints above its list is much bigger than what it serves.

Does it work for terms in English? Where Brazilian employers post in English, yes. software engineer returns real listings, and Portuguese terms return far more.

Why did my run stop straight away? A city was given without its state, or the maximum charge set for the run did not cover a single job. The run's message says which.

🆘 If something breaks

Open the Issues tab on the actor page. Send the input you used and the run ID, and the RUN_REPORT record if you can.