Buscar empresas que usam um ERP
find_companies_using_erp
Finds companies that run one of the ERPs asked (totvs, sap_business_one, sap, omie, sankhya, senior, linx, netsuite, dynamics, bling, tiny, contaazul, egestor, vhsys, marketup, gestaoclick, nomus), each proved with public evidence: a job post the company itself published, its site or its domain's e-mail setup. The same search, company check and filters as the "Usa ERP" signal (recruiters, accounting offices, ERP vendors and consultancies, and posts published for another company are left out), as a one-off list instead of a daily feed. Candidates come from job posts that name the ERP: Nock's job index (LinkedIn read every 6 hours, Gupy, Indeed, Vagas.com, Glassdoor, Sólides and others; a post whose employer has no site in the index is read too and the company found by name and place), LinkedIn searched live in the same call (one search per ERP asked with the area's words, and a second round while there is time and spend left), then openings in an area that runs it (hiringArea: fiscal_contabil by default, fiscal, contabil, controladoria, financeiro, compras), then companies of the market (location, industry, companySizes). Filters on the job post that brought the company, read BEFORE any check (a candidate that fails them is never checked nor charged): postedWithinDays (1 to 180, only posts published within that many days; a company with no dated post is left out) and jobPostTitles (words or titles, e.g. ["analista", "coordenador", "controller"], accents and case aside; a title of several words needs each). An ask that pairs the ERP with a job post in an area always maps to these filters: "usam TOTVS e abriram vaga no financeiro de analista, coordenador ou controller nos últimos 30 dias" is hiringArea "financeiro", jobPostTitles ["analista", "coordenador", "controller"], postedWithinDays 30. With hiringArea every company delivered has a post of that area within the days (hiringPost: title, date, link; namesTheErp true when that same post names the ERP, the strongest, listed first), apart from the ERP's proof; a company without one is never delivered. Each proof has its kind (evidence.kindLabel: the company's own job post, its site or domain, an employee's LinkedIn profile, another page), its date or age (evidence.age, "sem data" when unknown) and a grade (evidence.grade: forte when the post requires the ERP, média when desirable, fraca when it accepts other ERPs, "ou similares"; fraca is left out unless includeWeakProof true). An employee's LinkedIn profile is proof when any position of the person at this company, current or past (even after they left), names the ERP in its title, description or its own skills: graded média and delivered by default, with evidence.position and evidence.age ("cargo atual, desde 03/2024" or "ex-funcionário, cargo até 08/2023"). The ERP only in a position at another company, or only in the profile's general skills, headline or about, is no proof at all, not even fraca; the funnel says why. A company is delivered only as the post's employer: never a recruiter (by name, by its CNPJ's CNAE 78.10, 78.20 or 78.30, or a post that says "nosso cliente"; its people are never returned), a hidden employer ("Confidencial") or an accounting office, never a site of another company with a similar name (checked against the Registry's CNPJ and state), and, with companySizes, never one whose headcount (the job index's or the Registry's estimate) is outside them. Each company comes with its CNPJ, city and UF from the Receita; when the source, the site and the name in the post's state give none, the "Achar empresa" engine finds it by name, site and place. It also looks up one decision maker at each company (without jobTitle the finance decision makers, tried in order: CFO, diretor financeiro, gerente financeiro, controller, controller financeiro, head de finanças, gerente de controladoria, coordenador financeiro last; area "financeiro" leaves out other areas' and recruiters' titles; a jobTitle names anyone else). decisionMakerFallback role_only (the default) never puts someone else in the role's place: a company where nobody holds it comes with decisionMakerNotFound saying so; partner accepts the managing partner the Receita names (free) or the owner, each marked roleAsked false, with their LinkedIn looked up by name and company (3 credits, only when found). findPeople false skips people; withEmail adds the work e-mail. It saves every company in a sheet before it returns: the ERP, the evidence excerpt and its address on the company, the decision maker as a person row of that company. sheetId adds to an existing sheet and skips the companies it already has; without it a sheet named sheetName (default "Usa <ERPs>") is reused or created. To check a list the user already has: companies [{name, domain, cnpj}] (50 at most) or companiesSheetId; each one comes back in list[] with its status (entregue, erp_nao_provado, prova_fraca, fora_do_pedido, empresa_nao_achada, ja_na_planilha, nao_checada), the reason and what Nock's job index has of it (posts, the latest date, the post naming the ERP and its date: whether it was left out of a search by the date or never seen), with the same rules and prices. limit 10 by default, 200 at most. Two steps: without confirmed it runs nothing, charges nothing and returns maxCredits, the most the call can cost, candidates (how many companies of Nock's job index pass the filters before any check) and expected (expected.line: how many should pass and the likely cost, from the conversion of the searches recorded); say both, the ceiling and the expected. With confirmed true it runs and charges. Prices, only for the companies delivered (the companies checked and left out are never charged): 6 per company delivered with the ERP proved (its check and the job post read for it included); 8 per decision maker found by search (a partner the Receita names is free); 3 per work e-mail found. Finishes within about 85 seconds and returns the companies (name, domain, cnpj, city and uf; ERP, evidence kind, age, grade, source, url and excerpt; jobPost and hiringPost; foundOn, the job site its post came from; the decision maker with LinkedIn, or decisionMakerNotFound), counts (withCnpj, withPerson, proofGrades, proofKinds), the funnel (funnelText: candidates, delivered, then one short line per reason the rest was left out), sources (candidates and companies delivered per job site: LinkedIn, Indeed, Gupy, InfoJobs, Catho, Greenhouse/Lever...) and routes (Nock's index, LinkedIn now, the job posts bought, the area's openings), creditsCharged, sheetId and sheetPath. Each source runs within its own time, so a late one never takes the call. The call never spends with providers more than half of what its companies can be charged: past that it stops buying job posts and paid checks (stoppedAtCostCeiling true). partial true means the time ran out, a source was late or the cost ceiling was reached before the limit ("more" says which, in plain words); continuation.call is the next call, ready (the same input with the returned sheetId and limit = what is left; the companies already in the sheet are neither checked nor charged again, and this call's checks are reused free): with sendNow true it is the rest of the search the user already approved, within approvedCreditsLeft, so send it right away and say it continues; with sendNow false offer it.
Schema valid · SHA-256 0cba40231dd5…
View input schema
{
"type": "object",
"properties": {
"erps": {
"minItems": 1,
"maxItems": 8,
"type": "array",
"items": {
"type": "string",
"enum": [
"totvs",
"sap_business_one",
"sap",
"omie",
"sankhya",
"senior",
"linx",
"netsuite",
"dynamics",
"bling",
"tiny",
"contaazul",
"egestor",
"vhsys",
"marketup",
"gestaoclick",
"nomus"
]
},
"description": "Any of these ERPs (totvs, sap_business_one, sap, omie, sankhya, senior, linx, netsuite, dynamics, bling, tiny, contaazul, egestor, vhsys, marketup, gestaoclick, nomus); sap_business_one needs evidence naming Business One, sap is any SAP."
},
"hiringArea": {
"description": "The area of the job post the user asks for: fiscal, contabil, controladoria, financeiro or compras (fiscal_contabil is the default search when none is asked). Pass it whenever the ask names an area (\"abriram vaga no financeiro\" is financeiro): then every company delivered has a post of that area within postedWithinDays (30 when left out), shown as hiringPost apart from the ERP's proof, and the ones whose own area post names the ERP come first. Without it, no area post is required.",
"type": "string",
"minLength": 2,
"maxLength": 80
},
"location": {
"description": "A state or city, e.g. \"SP\" or \"Curitiba, PR\". Brazil when left out.",
"type": "string",
"minLength": 2,
"maxLength": 200
},
"industry": {
"description": "The sector in plain words, e.g. \"indústria de alimentos\".",
"type": "string",
"minLength": 2,
"maxLength": 200
},
"companySizes": {
"description": "Headcount bands, e.g. [\"51,200\", \"201,500\"].",
"maxItems": 8,
"type": "array",
"items": {
"type": "string",
"enum": [
"1,10",
"11,50",
"51,200",
"201,500",
"501,1000",
"1001,5000",
"5001,10000",
"10001,1000000"
]
}
},
"postedWithinDays": {
"description": "Optional, 1 to 180: only companies whose job post was published within this many days (\"abriram vaga nos últimos 30 dias\" is 30). Applied before any check: an older post, or a company with no dated post, is never checked nor charged. Each company returns its post's date (jobPost.postedAt).",
"type": "integer",
"minimum": 1,
"maximum": 180
},
"jobPostTitles": {
"description": "Optional: only companies whose job post's title has one of these words or titles, accents and case aside, e.g. [\"analista\", \"coordenador\", \"controller\"] (a title of several words needs each of them). Applied before any check: a post of another title is never checked nor charged. With hiringArea they are the titles of the area's post (\"vaga no financeiro de analista, coordenador ou controller nos últimos 30 dias\" is hiringArea \"financeiro\", jobPostTitles [\"analista\", \"coordenador\", \"controller\"], postedWithinDays 30).",
"minItems": 1,
"maxItems": 10,
"type": "array",
"items": {
"type": "string",
"minLength": 2,
"maxLength": 80
}
},
"includeWeakProof": {
"description": "true also delivers companies whose only proof is a post that accepts other ERPs (\"Winthor ou similares\"), graded fraca. Default false: left out. An employee's profile is never fraca: it is proof (média) when a position of the person at the company, current or past, names the ERP.",
"type": "boolean"
},
"limit": {
"description": "Companies to bring, 10 by default, 200 at most.",
"type": "integer",
"minimum": 1,
"maximum": 200
},
"jobTitle": {
"description": "Who to find at each company, titles tried in order, comma-separated. Default: the finance decision makers (\"CFO, Diretor Financeiro, Gerente Financeiro, Controller, Controller Financeiro, Head de Finanças, Gerente de Controladoria\", area \"financeiro\").",
"type": "string",
"minLength": 2,
"maxLength": 400
},
"area": {
"description": "Optional: only decision makers of this area. \"financeiro\" keeps finance people (CFO, finance director or manager, a finance controller) and leaves out other areas' titles such as \"Controller Jurídico\"; without jobTitle it also looks for the finance titles. Left out: financeiro when jobTitle is left out too, else anyone matching jobTitle.",
"type": "string",
"enum": [
"financeiro"
]
},
"decisionMakerFallback": {
"description": "When nobody holds the role asked (the finance titles by default): role_only, the default, leaves the company without a decision maker and says so (decisionMakerNotFound); partner accepts who decides there instead (the managing partner the Receita names, free, or the owner), with their LinkedIn looked up by name and company, charged only when found.",
"type": "string",
"enum": [
"role_only",
"partner"
]
},
"findPeople": {
"description": "false brings only the companies, no decision maker. Default true.",
"type": "boolean"
},
"withEmail": {
"description": "Also look up each decision maker's work e-mail. Default false.",
"type": "boolean"
},
"sheetId": {
"description": "An existing sheet to add the companies to; companies already in it are skipped. Without it, a new sheet is made.",
"type": "string",
"pattern": "^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$"
},
"sheetName": {
"description": "The name of the sheet to use or make when there is no sheetId (a sheet of that name is reused).",
"type": "string",
"minLength": 1,
"maxLength": 200
},
"companies": {
"description": "Check a list the user already has instead of searching (50 at most): each company by name, domain or CNPJ. Each one comes back in list[] with its status (entregue, erp_nao_provado, prova_fraca, fora_do_pedido, empresa_nao_achada, ja_na_planilha, nao_checada), the reason, and what Nock's job index has of it (posts, the latest date, the post naming the ERP and its date): it answers whether a company was left out of the search by the date or never seen. The same rules and prices: only the companies delivered are charged.",
"minItems": 1,
"maxItems": 50,
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {
"type": "string",
"minLength": 2,
"maxLength": 200
},
"domain": {
"type": "string",
"minLength": 3,
"maxLength": 200
},
"cnpj": {
"type": "string",
"minLength": 11,
"maxLength": 20
}
},
"additionalProperties": false
}
},
"companiesSheetId": {
"description": "Check the companies of this sheet instead (its name, site and CNPJ columns), the first 50.",
"type": "string",
"pattern": "^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$"
},
"confirmed": {
"description": "false or absent: returns the most it can cost and runs nothing. true: runs and charges.",
"type": "boolean"
}
},
"required": [
"erps"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}