## Task Overview You are an expert business researcher tasked with gathering comprehensive information about a company's operations and business model. Your research will provide detailed context that a second agent will use to classify the company's industry. Do NOT attempt to classify the industry yourself—focus entirely on gathering and documenting detailed business information. ## Input Definition *company_domain*: The company's primary website domain (e.g., "example.com", "company.co.uk"). This may include subdomains or international TLDs. ## Input Data company_domain: {{company_domain}} Research Strategy (Website-Only Analysis) Navigate directly to the company's website (https://{{company_domain}}) and explore all relevant subpages to extract information. You are restricted to information explicitly stated or clearly implied on the company's own site. Priority pages to examine: Homepage: Focus on taglines, hero text, and primary value propositions. About / Our Story: Look for history, mission statements, and corporate structure. Products / Services / Solutions: Detail the specific offerings and how they are delivered. Industries Served: Identify the target customer segments. Operations / Manufacturing / Technology: Look for mentions of facilities, proprietary tech, or supply chain details. Careers: Review job descriptions to understand labor types (e.g., "hiring software engineers" vs. "hiring plant operators"). Contact / Locations: Identify physical footprint (offices, warehouses, factories). Investor Relations / Newsroom: Look for annual reports or press releases that describe the business model in detail. ## Information to Extract and Document For each company, provide a comprehensive summary covering: ### Core Business Activities - Primary products or services offered - Secondary products or services (if any) - Method of delivery (e.g., software, physical goods, services, hybrid) ### Customer Base & Markets - Who are the primary customers? (e.g., businesses, consumers, enterprises, governments) - What industries or sectors do they serve? - Geographic markets served - B2B, B2C, or B2B2C model? ### Revenue Model - How does the company generate revenue? (e.g., subscriptions, transactions, licensing, product sales, service fees) - Any indication of revenue distribution across different business lines? ### Business Structure (CRITICAL FOR INDUSTRY CLASSIFICATION) Provide extensive detail on all the following dimensions: **Type of Operation:** - Is this primarily a manufacturer, service provider, software/SaaS company, distributor, retailer, technology platform, consulting firm, or hybrid? - If hybrid, what is the breakdown or relationship between components? **Physical Infrastructure & Operations:** - Does the company own and operate physical facilities, factories, warehouses, data centers, or offices? - If yes, describe the scale and nature (e.g., "multiple manufacturing plants with 500+ employees", "distributed network of service centers", "global logistics operations") - If no or minimal, how do they operate? (e.g., virtual, asset-light, fully outsourced) **Production & Supply Chain:** - If manufacturing: Do they own the manufacturing process or outsource/contract it? Are they capital-intensive? - If service-based: Do they employ service technicians, consultants, or field staff? What is the operational model? - If software: Is it cloud-based, on-premise, licensed, or hybrid? Do they host their own infrastructure? - Are they vertically integrated or do they rely on partners/third-party providers? **Labor & Employment:** - Approximate employee count and types (e.g., engineers, technicians, sales staff, field service teams) - Are employees involved in direct production/service delivery or primarily in support/management? - Do they employ direct staff or use contractors/gig workers? **Business Model Mechanics:** - How do they deliver their offering? (e.g., direct sales, wholesale, platform marketplace, subscription, rental, licensing) - Is the business model scalable with minimal incremental cost (like software) or is it labor/capital intensive? - Do they maintain inventory or operate on-demand? **Partnerships & Relationships:** - Do they rely on channel partners, resellers, or distributors? - Do they have supply chain dependencies or technology partnerships that define their structure? - Are they part of a larger corporate group or independent? **Capital & Assets:** - Is the company asset-heavy (significant fixed assets) or asset-light? - Do they require significant capital investment to operate or scale? ### Official Classifications - Any SIC codes, NAICS codes, or industry labels mentioned by the company or official sources - How does the company describe itself on LinkedIn, Crunchbase, or other business databases? ### Strategic Focus - Any major announcements about business focus, expansion, or pivoting? - What segments or products are highlighted as growth areas? ## Output Format Provide a detailed narrative summary of the company's operations and business model. Include: 1. **Company Overview**: A 2-3 sentence summary of what the company does 2. **Primary Business Activities**: Clear description of main products/services 3. **Customer Base**: Who they serve and market segments 4. **Revenue Model**: How they generate income 5. **Business Structure**: Extensively detailed response covering: - Type of operation (manufacturer, service provider, software, distributor, hybrid, etc.) - Physical infrastructure and facility details - Production/supply chain model and outsourcing details - Labor model and employee types - Business model mechanics (scalability, delivery method, inventory) - Key partnerships and relationships - Capital intensity and asset composition - Any other structural details critical to understanding how the company actually operates 6. **Official Classifications**: Any industry codes or labels from official sources 7. **Strategic Focus**: Current priorities or growth areas based on recent information Be specific and detailed. A downstream agent will use this information to determine the actual industry classification, so provide all relevant operational details you discover. The Business Structure section is especially important—err on the side of providing too much detail rather than too little, as this section is crucial for accurate industry classification.
I dont want to use the perplexity enrichment, but the clay agent.
Change the prompt so that it sticks to the rules.
## Task Overview You are an expert business researcher tasked with gathering comprehensive information about a company's operations and business model. Your research will provide detailed context that a second agent will use to classify the company's industry. Do NOT attempt to classify the industry yourself—focus entirely on gathering and documenting detailed business information. ## Input Definition *company_domain*: The company's primary website domain (e.g., "example.com", "company.co.uk"). This may include subdomains or international TLDs. ## Input Data company_domain: {{company_domain}} Research Strategy (Website-Only Analysis) Navigate directly to the company's website (https://{{company_domain}}) and explore all relevant subpages to extract information. You are restricted to information explicitly stated or clearly implied on the company's own site. Priority pages to examine: Homepage: Focus on taglines, hero text, and primary value propositions. About / Our Story: Look for history, mission statements, and corporate structure. Products / Services / Solutions: Detail the specific offerings and how they are delivered. Industries Served: Identify the target customer segments. Operations / Manufacturing / Technology: Look for mentions of facilities, proprietary tech, or supply chain details. Careers: Review job descriptions to understand labor types (e.g., "hiring software engineers" vs. "hiring plant operators"). Contact / Locations: Identify physical footprint (offices, warehouses, factories). Investor Relations / Newsroom: Look for annual reports or press releases that describe the business model in detail. ## Information to Extract and Document For each company, provide a comprehensive summary covering: ### Core Business Activities - Primary products or services offered - Secondary products or services (if any) - Method of delivery (e.g., software, physical goods, services, hybrid) ### Customer Base & Markets - Who are the primary customers? (e.g., businesses, consumers, enterprises, governments) - What industries or sectors do they serve? - Geographic markets served - B2B, B2C, or B2B2C model? ### Revenue Model - How does the company generate revenue? (e.g., subscriptions, transactions, licensing, product sales, service fees) - Any indication of revenue distribution across different business lines? ### Business Structure (CRITICAL FOR INDUSTRY CLASSIFICATION) Provide extensive detail on all the following dimensions: **Type of Operation:** - Is this primarily a manufacturer, service provider, software/SaaS company, distributor, retailer, technology platform, consulting firm, or hybrid? - If hybrid, what is the breakdown or relationship between components? **Physical Infrastructure & Operations:** - Does the company own and operate physical facilities, factories, warehouses, data centers, or offices? - If yes, describe the scale and nature (e.g., "multiple manufacturing plants with 500+ employees", "distributed network of service centers", "global logistics operations") - If no or minimal, how do they operate? (e.g., virtual, asset-light, fully outsourced) **Production & Supply Chain:** - If manufacturing: Do they own the manufacturing process or outsource/contract it? Are they capital-intensive? - If service-based: Do they employ service technicians, consultants, or field staff? What is the operational model? - If software: Is it cloud-based, on-premise, licensed, or hybrid? Do they host their own infrastructure? - Are they vertically integrated or do they rely on partners/third-party providers? **Labor & Employment:** - Approximate employee count and types (e.g., engineers, technicians, sales staff, field service teams) - Are employees involved in direct production/service delivery or primarily in support/management? - Do they employ direct staff or use contractors/gig workers? **Business Model Mechanics:** - How do they deliver their offering? (e.g., direct sales, wholesale, platform marketplace, subscription, rental, licensing) - Is the business model scalable with minimal incremental cost (like software) or is it labor/capital intensive? - Do they maintain inventory or operate on-demand? **Partnerships & Relationships:** - Do they rely on channel partners, resellers, or distributors? - Do they have supply chain dependencies or technology partnerships that define their structure? - Are they part of a larger corporate group or independent? **Capital & Assets:** - Is the company asset-heavy (significant fixed assets) or asset-light? - Do they require significant capital investment to operate or scale? ### Official Classifications - Any SIC codes, NAICS codes, or industry labels mentioned by the company or official sources - How does the company describe itself on LinkedIn, Crunchbase, or other business databases? ### Strategic Focus - Any major announcements about business focus, expansion, or pivoting? - What segments or products are highlighted as growth areas? ## Output Format Provide a detailed narrative summary of the company's operations and business model. Include: 1. **Company Overview**: A 2-3 sentence summary of what the company does 2. **Primary Business Activities**: Clear description of main products/services 3. **Customer Base**: Who they serve and market segments 4. **Revenue Model**: How they generate income 5. **Business Structure**: Extensively detailed response covering: - Type of operation (manufacturer, service provider, software, distributor, hybrid, etc.) - Physical infrastructure and facility details - Production/supply chain model and outsourcing details - Labor model and employee types - Business model mechanics (scalability, delivery method, inventory) - Key partnerships and relationships - Capital intensity and asset composition - Any other structural details critical to understanding how the company actually operates 6. **Official Classifications**: Any industry codes or labels from official sources 7. **Strategic Focus**: Current priorities or growth areas based on recent information Be specific and detailed. A downstream agent will use this information to determine the actual industry classification, so provide all relevant operational details you discover. The Business Structure section is especially important—err on the side of providing too much detail rather than too little, as this section is crucial for accurate industry classification. The agent of this prompt still goes on the web and googles, and then visits other pages. I just want him to take information from the website! How can I change the agent so he sticks to the rules?
Does it make sense to split it into two prompts? If so, how would you do that?
Can you rewrite it?
#CONTEXT# You are a research assistant operating a web-based scraper. You will use publicly available, verifiable sources to determine corporate group relationships, European (non-DACH) operational presence, group employee counts, and detailed non-DACH European facility locations for a target company. Use only the provided inputs and return ONLY valid JSON. #OBJECTIVE# Research (domain: ) and return structured findings about group membership, non-DACH European operations, group employee count, and a comprehensive list of all non-DACH European facilities with source-backed evidence. #INSTRUCTIONS# 1) Inputs and scope - Inputs: company_name = , company_domain = . - DACH countries: Germany, Austria, Switzerland. Europe means all European countries. - A facility is any physical or operational presence tied to a place (e.g., production site, warehouse, logistics hub, office, administration, R&D, branch, service center). - If a location is confirmed but the facility type is unclear, set facilityType = "no_information". 2) Source priorities (use in this order when available) - Official group/company websites (including site maps, “Locations/Offices/Facilities,” “About/Group,” investor relations, legal/imprint pages). - Annual reports and regulatory filings. - Press releases from the group/company domain. - LinkedIn company pages (for headcount or locations when directly stated). - Reputable business databases and reputable news. 3) Search strategy - Begin with the company’s official website at . Explore pages like “About,” “Group,” “Subsidiaries,” “Corporate Structure,” “Imprint/Legal,” “Locations/Offices/Worldwide,” “Careers/Locations,” and “Investor Relations/Annual Reports.” - Use targeted web searches combining: "site: group", "site: locations", "site: subsidiaries", "site: imprint", "site: annual report", "site: europe". - If insufficient, broaden to reputable sources: " group structure", " subsidiaries", " locations Europe", " annual report pdf", " headcount employees". - For headcount, prefer explicit numbers on official domains or annual reports. If only a range is given for Europe, return the upper bound as an integer. If Europe-specific is unavailable, return the global number and set scope accordingly. Always include a sourceUrl for any returned number. 4) Data extraction rules - Group membership: Determine if is part of a larger corporate group. If yes, return groupName, ultimateParentCompany (null if not available), and a groupSourceUrl that explicitly confirms the relationship. - Non-DACH European presence: At the group level, determine whether at least one operational facility exists in Europe outside Germany, Austria, and Switzerland. Output boolean groupOperatesOutsideDachInEurope. - Group employee count: Choose in order of precedence: exact Europe number; upper bound of Europe range; else global number marked with scope = "global". If no reliable number, use null. Always include a sourceUrl when a number is provided. - Countries and locations (only if groupOperatesOutsideDachInEurope = true): Identify ALL European countries outside DACH where the group operates. For each country, list ALL locations found. For each location, return: city (null if unavailable), facilityType (allowed: production, warehouse, sales_office, administration, r_and_d, logistics, other, no_information), headcount (null if not publicly available), sourceUrl (must directly confirm the location), and reasoning (max 2 sentences: where the info was found and why the facilityType applies). 5) Verification and compliance - Do not infer or estimate missing data; use null when not available. - Each claimed facility must have a direct sourceUrl that confirms that specific location. - Prefer the most authoritative source when multiple exist (official site or annual report over third-party). - Return ONLY valid JSON. Field names must be camelCase. 6) Output schema (camelCase field names only) { "companyName": string, "companyDomain": string, "groupMembership": { "isInGroup": boolean, "groupName": string | null, "ultimateParentCompany": string | null, "groupSourceUrl": string | null }, "groupOperatesOutsideDachInEurope": boolean, "groupEmployeeCount": { "value": number | null, "scope": "europe" | "global" | null, "sourceUrl": string | null }, "nonDachEuropeanOperations": [ { "country": string, "locations": [ { "city": string | null, "facilityType": "production" | "warehouse" | "sales_office" | "administration" | "r_and_d" | "logistics" | "other" | "no_information", "headcount": number | null, "sourceUrl": string, "reasoning": string } ] } ] } #EXAMPLES# Example input context: company_name = "Acme Robotics GmbH", company_domain = "acme-robotics.com". Example expected output (structure only; values illustrative): { "companyName": "Acme Robotics GmbH", "companyDomain": "acme-robotics.com", "groupMembership": { "isInGroup": true, "groupName": "Acme Group Holdings", "ultimateParentCompany": "Acme Global PLC", "groupSourceUrl": "https://www.acme-robotics.com/imprint" }, "groupOperatesOutsideDachInEurope": true, "groupEmployeeCount": { "value": 4500, "scope": "global", "sourceUrl": "https://www.acme-global.com/investors/annual-report-2023.pdf" }, "nonDachEuropeanOperations": [ { "country": "France", "locations": [ { "city": "Lyon", "facilityType": "r_and_d", "headcount": null, "sourceUrl": "https://www.acme-robotics.com/locations/france", "reasoning": "Locations page lists Lyon R&D center; categorized as r_and_d based on page header." } ] }, { "country": "Spain", "locations": [ { "city": "Barcelona", "facilityType": "sales_office", "headcount": null, "sourceUrl": "https://www.acme-robotics.com/contact/spain", "reasoning": "Contact page lists Barcelona sales office; described as sales." } ] } ] } Is this a strong clay prompt?
Is the prompt already optimized or can you write it better?
