Back

ZoomInfo

Search company and contact data.

MCP server URL

https://zoominfo.gumstack.com/mcp

Works with

Tools 15

  • Account Research

    Research a company/account/organization using ZoomInfo market data with your CRM and conversation history. Handles both targeted queries ("What's their recent news?") and broad orientation ("What's going on with this account?"). USE THIS WHEN: You need information about a company, organization, or account (their business, news, financials, tech stack, leadership team, recent developments). DO NOT USE FOR: Individual person research (use Contact Research instead). ANSWERS QUESTIONS LIKE: • Company context: What does this company do? Who are their competitors? What's their tech stack? • Leadership: Who's in the C-suite? What's the background of their CTO? • Financials: What's their revenue? Have they raised funding recently? • Recent developments: Any recent news or executive changes? • Relationship status: What's our deal status? Who's on the buying committee? • Engagement overview: How engaged are we? Who are our key contacts? • Qualification: What's our MEDDIC status with this account? DOES NOT HANDLE: • Specific conversation queries ("What exactly did they say about X?") → Use Conversation Intelligence • Detailed engagement timelines or transcripts → Use Engagements • Structured data for export (revenue fields, headcount) → Use Enrich Companies • Raw/paginated signal export (Intent, News, Scoop) for downstream systems or bulk records → Use Enrich Company Signals FOR BEST RESULTS: Construct queries that frame the user's goal and describe what ideal context would look like. The more the query conveys about the situation and desired outcome, the better the response. General orientation: "User wants to understand what's going on with this account. Provide broad situational awareness — who they are, how the relationship stands, what's been happening, and anything notable I should know." Call prep: "User has a meeting tomorrow with their VP of Engineering. Provide context that helps them walk in prepared — understanding of the company, the person, the relationship history, and any active deals or recent developments." Deal review: "User is presenting this account at forecast. Provide context that supports a confident deal narrative — where the deal stands, who's involved, momentum indicators, and any risks." Competitive situation: "User is positioning against Salesforce at this account. Provide context that informs competitive strategy — what they use today, how they evaluate, and what we know about their buying process." Account planning: "User is building a strategic plan for this account. Provide context that supports territory strategy — the opportunity landscape, key relationships, coverage gaps, and growth signals." RETURNS: Market intelligence: company overview, leadership profiles, products/services, financials, competitors, technology stack, recent news. Insight signals: recent trigger events — funding, M&A, earnings, layoffs, partnerships, IPO/divestiture, product launches, leadership and buying-committee changes, hiring anomalies, website visits, and intent spikes (directional indicators). Relationship intelligence: CRM records, buying committee, key contacts, engagement summary, MEDDIC qualification. REQUIRES: zi_company_id (7-8 digit integer, e.g. '344589814'). Use Search Companies if you have a company name without the zoominfo company id. CREDITS: Each query consumes ZoomInfo AI credits. Credit consumption scales with the amount of context retrieved and the complexity of inference performed.

  • Contact Research

    Research a specific person/contact using ZoomInfo market data with your CRM and conversation history. Handles both targeted queries ("What's their background?") and broad orientation ("Who is this person?"). USE THIS WHEN: You need information about an individual person (their role, background, contact details, career history). DO NOT USE FOR: Company/organization research (use Account Research instead). ANSWERS QUESTIONS LIKE: • Professional background: What's their career history? Where did they work before? • Current role: What do they do? What are their responsibilities? • Education & expertise: Where did they go to school? What's their expertise? • Company context: What company do they work for? What does that company do? • Contact info: What's their email? Phone number? LinkedIn? • Recent activities: Any recent news mentions or job changes? • Relationship status: How engaged are they? Are they in our CRM? • Buying involvement: Are they on a buying committee? What's their role in deals? DOES NOT HANDLE: • Specific conversation queries ("What exactly did they say about X?") → Use Conversation Intelligence • Detailed engagement timelines or transcripts → Use Engagements • Structured data for export (title, email, phone) → Use Enrich Contacts FOR BEST RESULTS: Construct queries that frame the user's goal and describe what ideal context would look like. The more the query conveys about the situation and desired outcome, the better the response. General orientation: "User wants to understand who this contact is. Provide broad situational awareness — their background, current role, company, and anything notable I should know." Meeting prep: "User has a meeting tomorrow with this person. Provide context that helps them walk in prepared — understanding of their role, background, recent activities, and our relationship history." Outreach planning: "User is planning to reach out to this contact. Provide context that informs a personalized approach — who they are, what they care about, how they engage, and any warm introduction paths." Champion building: "User wants to build this person into a champion. Provide context that supports relationship development — their influence, interests, engagement patterns, and what motivates them." Stakeholder mapping: "User is mapping stakeholders for a deal. Provide context that clarifies this person's role — their influence, relationships, decision authority, and how they fit into the buying process." RETURNS: Contact intelligence: professional background, current role, education, contact information, company context, recent activities. Relationship intelligence: CRM records, engagement history, buying involvement, communication patterns. REQUIRES: zoominfoContactId (7-9 digit integer, e.g. '123456789'). Use Search Contacts if you have a contact name or email without the zoominfo contact id. CREDITS: Each query consumes ZoomInfo AI credits. Credit consumption scales with the amount of context retrieved and the complexity of inference performed.

  • Conversation Intelligence

    Answers natural-language questions about calls, meetings, and emails, scoped to a single engagement, an account, or a contact. Queries should carry the user's detailed intent and relevant conversation context, including any specified date range of interest (default lookback: 365 days). Requires a meeting or email provider configured in ZoomInfo. USE THIS WHEN: - You need to analyze conversations for objections, commitments, sentiment, next steps, or action items. - You need to surface what was actually said in calls, meetings, or emails — requirements, promises, or concerns. - NOT for CRM fields, firmographics, or market data. ANSWERS QUESTIONS LIKE: Engagement: "What were the customer's main objections on this call?" · "What was agreed that a follow-up should reference?" Account: "What concerns has this account raised recently? What did we promise them?" Contact: "What has this person told us they care about?" SCOPING — pass exactly one ID (else BAD_REQUEST): engagementId — one call, meeting, or email thread in depth. Opaque string from Browse Engagements or this tool's citations; pass verbatim — some email IDs contain literal angle brackets, which are part of the ID; others are opaque tokens. zoominfoCompanyId — synthesize across an account's recent engagements. zoominfoContactId — synthesize across a contact's recent engagements. SAMPLING (IMPORTANT): Account and contact scopes quote from only the 5 most recent meetings and 5 most recent emails (meetings without stored transcripts are excluded). Account scope additionally draws on account-level intelligence synthesized across full history. For deeper coverage, drill into cited engagementIds. ENTRY POINT: Temporal or positional selector ("my most recent call") → enumerate with Browse Engagements (free, deterministic ordering), then pass the engagementId here. Thematic selector ("what concerns have they raised") → start account- or contact-scoped, then ask targeted follow-ups against a cited engagementId. Set includeContent: true (engagementId scope only, default false) only when verbatim text is required — it appends the raw transcript or email body, prefixed "Raw Engagement:"; long transcripts may be truncated; consumes significant context. IMPORTANT: Transcript/email body inclusion is controlled SOLELY by this boolean — never put transcript requests in the query field. If the user asks for a transcript, set this to true and keep the query focused on analytical intent (e.g. summary, key points). RETURNS: Synthesized answer with inline citations to source engagements, opening with a data-check line and ending with a Coverage section (window searched, N of M engagements reviewed, truncation notes) — read Coverage before treating an answer as exhaustive. DOES NOT HANDLE: Exhaustive topic search ("every call mentioning X") — sampling makes topic recall incomplete. Building a full engagement timeline or finding engagement IDs → Browse Engagements. Broad account orientation or company/market research → Account Research. Person background or profile research → Contact Research. Structured firmographic or demographic data for export → Enrich Companies / Enrich Contacts. QUALITY: Transcription, diarization, and inference are automated; attribution errors and omissions are possible — verify critical facts by calling this tool again with the cited engagementId and includeContent=true to retrieve the full transcript. CREDITS: Minimum 9 ZoomInfo AI credits per call, scaling with context retrieved and inference complexity.

  • Enrich Companies

    Enrich companies with full profiles from ZoomInfo: description, industries, company type, ticker, phone, address, founded year, funding, revenue and employee ranges, parent and ultimate parent companies, social media URLs and department headcount. `search_companies` finds companies and returns `companyId`s with a thin record. Pass those `companyId`s here for full profiles, for prospecting, qualification and prioritization. Enrich also works for companies you already know by domain, name or ticker, and for the employer `company.id` on a `search_contacts` result. The default response holds name, description, location, revenue and industries. Add `requiredFields` for anything else, e.g. `["description", "industries", "companyFunding", "foundedYear", "parentName", "phone", "street"]`. Example: "Find fintech companies in NYC and return full profiles for the top 100" ``` 1. lookup(fields: [{ fieldName: "industries", fuzzyMatch: "fintech" }]) → id "software.fintech" 2. search_companies(industryList: ["software.fintech"], metroRegion: "NY - New York City", sort: "-revenue", pageSize: 100) → companyId 55555, 66666, ... 3. enrich_companies({ "companies": [{ "companyId": "55555" }, { "companyId": "66666" }, ...], "requiredFields": ["description", "industries", "companyFunding", "foundedYear"] }) → full profiles; repeat with the next batch until all 100 are done ``` Once you've enriched the right profiles, if `account_research` is available, call it with the `companyId` as `zoominfoCompanyId` to go deeper with a company research agent: news, leadership, competitors, tech stack and your engagement and CRM history. **Identifying companies** For each company, provide at least one of: `companyId` (most accurate), `companyName`, `companyWebsite` or `domain`, `companyTicker`, address information (street, city, state, etc.) or `ipAddress`. Up to 10 companies per call. CREDITS: Records already under management carry no charge; other records use one Data (Bulk) credit each.

  • Enrich Contacts

    Enrich contacts with verified business information from ZoomInfo: corporate email addresses, business direct and mobile phone numbers, LinkedIn and other professional profile URLs, job details, employment history and education. All data returned by this tool is strictly business-related. It does not return personal contact information such as private phone numbers, personal email addresses, or non-professional data of any kind. `search_contacts` finds people and returns `personId`s with a thin record. Pass those `personId`s here to get contact details and full profiles. Enrich also works for people you already know by email, phone, or name and company. The default response holds name, email and employer. Add `requiredFields` for anything else, e.g. `["email", "phone", "mobilePhone", "externalUrls"]` for direct and mobile phones and LinkedIn. Example: "Find the VPs and directors of marketing at Example Corp and get their emails, direct dials and LinkedIn" ``` 1. search_contacts(companyName: "Example Corp", departmentList: ["Marketing"], managementLevelList: ["VP Level Exec", "Director"]) → personId 444, 555, 666, ... 2. enrich_contacts({ "contacts": [{ "personId": "444" }, { "personId": "555" }, { "personId": "666" }, ...], "requiredFields": ["email", "phone", "externalUrls"] }) → verified contact details; repeat with the next batch until all are done ``` **Identifying contacts** For each contact, provide one of these combinations: `personId` (most accurate), `email`, `phone`, `firstName` + `lastName` + `companyName`, `fullName` + `companyName`, `firstName` + `lastName` + `companyId`, or `fullName` + `companyId`. Up to 10 contacts per call. ``` enrich_contacts({ "contacts": [ {"email": "john.smith@company.com"}, {"personId": "12345"}, {"firstName": "Jane", "lastName": "Doe", "companyName": "Tech Corp"} ] }) ``` CREDITS: Records already under management carry no charge; other records use one Data (Bulk) credit each.

  • Lookup

    Lookup reference data - This tool provides standardized parameter values for use in search operations. Using lookup helps ensure accurate search results. **Note: Use the `id` field, not `attributes.name`** • Lookup results contain both `id` and `attributes.name` • Use the `id` value in search parameters • The `attributes.name` is for human readability only • Example: Use `id: "12345"` not `attributes.name: "Vice President"` **Recommended Workflow:** 1. Call lookup to get exact field values 2. Extract the `id` field from each result 3. Use the `id` values in search_contacts or search_companies 4. This helps prevent search failures and ensures accurate results **Single field lookup:** ``` lookup(fields: [{ fieldName: "management-levels" }]) → Get VP options, use result.id ``` **Multiple fields lookup:** ``` lookup(fields: [{ fieldName: "management-levels" }, { fieldName: "metro-regions" }, { fieldName: "industries" }, { fieldName: "employee-count" }]) → Get all options, use result.id values ``` **Fuzzy match filtering:** ``` lookup(fields: [{ fieldName: "tech-vendors", fuzzyMatch: "hubspot" }]) → Returns vendors matching "hubspot", use result.id (e.g., "HubSpot, Inc") lookup(fields: [{ fieldName: "industries", fuzzyMatch: "software" }]) → Returns industries containing "software", use result.id ``` • Use `fuzzyMatch` within each field object to filter results by partial name match (case-insensitive) • Helps find specific items when you don't know the exact name • Example: searching "hub spot" will match "HubSpot, Inc" **Note**: For batch lookups with multiple fields, pass each as a separate object in the `fields` array. - Per-field fuzzyMatch is set on each field object individually (e.g., `[{ fieldName: "metro-regions", fuzzyMatch: "boston" }, { fieldName: "industries" }]` filters field 1, not field 2) - Omit `fuzzyMatch` or set it to null to skip filtering for a specific field **Example: VP-level tech contacts in San Francisco (100-500 employees):** ``` Step 1: lookup(fields: [{ fieldName: "management-levels" }, { fieldName: "metro-regions", fuzzyMatch: "san francisco" }, { fieldName: "industries", fuzzyMatch: "software" }, { fieldName: "employee-count" }]) → Get all options at once Step 2: Extract id values from results Step 3: search_contacts(managementLevel: "<id>", metroRegion: "<id>", industryCodes: "<id>", employeeCount: "<id>") ``` **Example: Companies using specific technology (e.g., "Companies in Boston using HubSpot"):** ``` Step 1: lookup(fields: [{ fieldName: "tech-vendors", fuzzyMatch: "hubspot" }]) → Get vendor id Step 2: lookup(fields: [{ fieldName: "tech-products" }], vendor: "<vendor_id>") → Get HubSpot product ids Step 3: lookup(fields: [{ fieldName: "metro-regions", fuzzyMatch: "boston" }]) → Get metro id Step 4: search_companies(techAttributeTagList: "<product_ids>", metroRegion: "<metro_id>") ``` **Note:** For technology searches, lookup tech-vendors (with fuzzyMatch), then lookup tech-products (with vendor parameter), and use the `id` values. **Common lookup fields:** • **management-levels** - VP, Director, C-Level classifications (for seniority searches) • **metro-regions** - Exact metropolitan area names (for location searches) • **industries** - Industry classifications (for sector searches) • **employee-count** - Company size ranges (for size filters) • **job-functions** - Job function categories • **departments** - Department classifications • **company-types** - Public/private classifications • **revenue-ranges** - Revenue classifications • **funding-round-types** - Funding round classifications (e.g., Seed, Series A) for recentFundingRoundTypes/allFundingRoundTypes filters • **tech-vendors** - Technology vendor names (use fuzzyMatch for partial matching, use before tech-products lookup) • **tech-products** - Technology product IDs (use vendor parameter with exact vendor name from tech-vendors lookup) • **tech-categories** - Technology categories for broader tech stack searches **Note:** Search tools may fail if you use incorrect field values. Consider using lookup and the `id` field for accuracy. **Note:** When looking up hashtags and intent-topics, this tool will return a maximum of 100 records. If more than 100 records exist, the response will include a truncation notice indicating that results were limited. CREDITS: Free to use

  • Search Companies

    Search ZoomInfo for companies by name, domain, industry, location, headcount, revenue, growth, funding, technology and other firmographic filters. Search finds companies and returns their `companyId`s. Once you're happy with your filters, redeem or enrich those companies to fetch full profiles, with attributes such as description, industries, parent companies, funding, founded year, company type and ticker, phone and address. Two ways to fetch full profiles: - `redeem: true`: full profiles for every company on the page, up to 25. - `enrich_companies` with the returned `companyId`s, or domains or names you already have: profiles for the companies you choose. Example: "Full profiles for fintech companies in New York" ``` 1. lookup(fields: [{ fieldName: "industries", fuzzyMatch: "fintech" }]) → id "software.fintech" 2. search_companies(industryList: ["software.fintech"], metroRegion: "NY - New York City", pageSize: 25, redeem: true) ``` Example: "Software companies in San Francisco with 100–500 employees", then "full profiles for the 100 largest" ``` 1. lookup(fields: [{ fieldName: "metro-regions", fuzzyMatch: "san francisco" }]) → "CA - San Francisco" 2. lookup(fields: [{ fieldName: "industries", fuzzyMatch: "software" }]) → ids such as "software" 3. search_companies(metroRegion: "CA - San Francisco", industryList: ["software"], employeeRangeMinimum: 100, employeeRangeMaximum: 500, sort: "-employeeCount", pageSize: 100) → companyId 55555, 66666, ... 4. Redeem: repeat step 3 with pageSize: 25, page: 1, redeem: true, then page: 2, 3 and 4 → 100 full profiles (fewer if meta.totalResults is lower) Or enrich: enrich_companies(companies: [{ companyId: "55555" }, { companyId: "66666" }, ...], requiredFields: ["description", "industries", "companyFunding", "foundedYear", "parentName"]) → full profiles; repeat with the next batch until all 100 are done ``` CREDITS: Search is free. With redeem or `enrich_companies`, records already under management carry no charge; other records use one Data (Bulk) credit each.

  • Search Contacts

    Search ZoomInfo for business contacts by name, company, job title, email, location, management level, department, accuracy score and company attributes. Search finds people and returns their `personId`s. Once you're happy with your filters, redeem or enrich those contacts to fetch full profiles, with attributes such as verified business email, direct and mobile phone numbers, LinkedIn and other profile URLs, job function, position start date and employment history. Two ways to fetch full profiles: - `redeem: true`: full profiles for every contact on the page, up to 25. - `enrich_contacts` with the returned `personId`s: profiles for the contacts you choose. Example: "Emails for the VPs of Marketing at Snowflake" ``` search_contacts(companyName: "Snowflake", managementLevelList: ["VP Level Exec"], departmentList: ["Marketing"], requiredFieldsList: ["email"], pageSize: 25, redeem: true) ``` Example: "VPs at software companies in San Francisco with 100–500 employees", then "get me direct dials for the first 100" ``` 1. lookup(fields: [{ fieldName: "metro-regions", fuzzyMatch: "san francisco" }]) → "CA - San Francisco" 2. lookup(fields: [{ fieldName: "industries", fuzzyMatch: "software" }]) → ids such as "software" 3. search_contacts(managementLevelList: ["VP Level Exec"], metroRegion: "CA - San Francisco", industryList: ["software"], employeeRangeMinimum: 100, employeeRangeMaximum: 500, requiredFieldsList: ["directPhone"], pageSize: 100) → personId 444, 555, ... 4. Redeem: repeat step 3 with pageSize: 25, page: 1, redeem: true, then page: 2, 3 and 4 → 100 full profiles (fewer if meta.totalResults is lower) Or enrich: enrich_contacts(contacts: [{ personId: "444" }, { personId: "555" }, ...], requiredFields: ["email", "phone"]) → verified emails and direct dials; repeat with the next batch until all 100 are done ``` CREDITS: Search is free. With redeem or `enrich_contacts`, records already under management carry no charge; other records use one Data (Bulk) credit each.

  • Search Scoops

    Search for Scoops across ZoomInfo's comprehensive database. Scoops are real-time business intelligence signals that provide timely insights into significant events and changes occurring at companies. **SCOOP FILTERS:** • scoopIds - ZoomInfo unique identifiers for Scoops • scoopTypes - Retrieve scoops based on type (e.g. Earnings, Award, and Partnership). • scoopTopics - Retrieve scoops based on topic (e.g. integration, consolidation and compliance). Accepts a list of IDs from the lookup tool • department - Retrieve scoops based on department (IT, finance, HR and so on). • description - Search for scoops based on description. Accepts a space-separated list of individual words • publishedStartDate - Starting date to search for scoops based on publication date (YYYY-MM-DD format) • publishedEndDate - Ending date to search for scoops based on publication date (YYYY-MM-DD format) • updatedSinceCreation - Setting true will only return scoops that have been updated since publishedStartDate **COMPANY FILTERS:** • companyTicker, companyDescription, companyTypes - Identify or describe specific companies • industryCodes, metroRegions, employeeCount, revenue - Filter by company attributes • techAttributeTagList - Filter by technology stack (use Lookup tech-products for valid values) **CONTACT FILTERS:** • jobTitle, managementLevel, department, jobFunction - Filter by contact role • firstName, lastName, emailAddress - Filter by specific contact **SORT OPTIONS:** • scoopId, originalPublishedDate (default: -originalPublishedDate), description, link, linkText • Prefix with '-' for descending order **USE CASES:** • Find companies with recent leadership changes or expansions • Identify companies that recently received funding or made acquisitions • Track industry-specific business events for timely outreach • Discover companies showing signals of buying intent through business events **RECOMMENDED WORKFLOW:** 1. Use Lookup with fieldName: "scoop-topics" to get valid filter IDs 2. Call search_scoops with the desired filters 3. Use Enrich Scoops for detailed scoop data on a specific company **Example: "Recent leadership changes at fintech companies":** ``` 1. lookup(fields: [{ fieldName: "scoop-topics", fuzzyMatch: "leadership" }]) → Returns topic ids like "leadership-change" 2. lookup(fields: [{ fieldName: "industries", fuzzyMatch: "financial technology" }]) → Returns id: "<industry_id>" (display name: "Financial Technology (Fintech)") 3. search_scoops(scoopTopics: ["leadership-change"], industryCodes: ["<industry_id>"], publishedStartDate: "2026-01-01") ``` **CREDIT USAGE:** This endpoint does not consume any Credits. Each Scoop returned counts towards the customer's Record Limit and a successful response counts towards the Request Limit.

  • Browse Engagements

    Use this tool to retrieve engagements (meetings and emails). Useful for tracking engagement history with external contacts. Supports filtering by date, zoominfoCompanyId and/or zoominfoContactId. The date range can be set with engagementDateStart and engagementDateEnd (defaulting to the last 7 days). The window between engagementDateStart and engagementDateEnd must be no more than 90 days, with engagementDateStart before engagementDateEnd. This tool can only return results if the user's organization has set up integrations between ZoomInfo and at least one of its calendar, email, or meeting providers. Reach out to your account manager for more information. By default, results are returned in chronological order. To get the last few of an item, reverse the sort order with "-chronological". Does not consume data credits.

  • Get Gtm Context

    Returns the authenticated user's GTM context: their role, what their organization offers (products and services), ideal customer profiles, buyer personas, competitors, and strategic priorities. No parameters required — identity is resolved from authentication. This is read-only; the user needs admin permissions to create, update, delete or archive configuration. No ZoomInfo credits consumed. This context shapes how other ZoomInfo tools deliver value. Search, enrichment, and prospecting results are more relevant when filtered and scored against the user's ideal customer profiles, buyer personas, and competitive landscape. Without it, those tools return unqualified results that require manual assessment. Compressed by default (~1,000 tokens). Use detailed=true for expanded profiles. Each entity includes a status field ('active' or 'archived') and an archivedAt timestamp when applicable, so archived configuration is always visible.

  • Get Recommended Contacts

    Use this tool to retrieve a ranked list of top N contact recommendations at a target company based on the users interaction history. Required: - Company ID to search with. If you don't have this yet, use the `search_companies` tool set first. - Use case type for the recommendation. This filters recommendations based on the sales motion, depending on use case, it can the following: - PROSPECTING (capitalized): for recommendations based on similar contacts to those copied, viewed, or exported by the user on the ZoomInfo Platform. The prospecting use-case has cold-start support; when in doubt default to this use case. - DEAL_ACCELERATION: when the use-case calls for recommending contacts similar to those in the set of closed won opportunities in user's tenant's CRM for new business initiatives. - RENEWAL_AND_GROWTH: when the use-case calls for recommending contacts similar to those in the set of closed won opportunities in user's tenant's CRM for renewal initiatives. Contact Recommendations can be used to retrieve a ranked list of people at a target company who are most relevant for a given sales motion (use case), such as prospecting, deal acceleration, or renewal and growth. The recommendations are derived from past user interactions and account activity, and are ranked by a machine learning model. These recommendations can be used to identify appropriate business contacts for research and strategic planning purposes. Compliance Note: Search results are intended for professional research. Ensure all use aligns with your organization's data governance policies. Behind the scenes, the model leverages data such as the user's past contact views, exports, and copies for the PROSPECTING motion, or contacts from a user's CRM related to closed won deals for the DEAL_ACCELERATION motion. The model uses this data to infer which types of people are most likely to drive success for the selected motion. It then finds similar contacts at the target company and scores them using a combination of similarity and propensity signals. The tool returns up to 100 recommended contacts, ORDERED FROM MOST TO LEAST RELEVANT. Each recommendation contains additional metadata (`meta`) that describes the reference person used to form the recommendation. PLEASE USE THIS METADATA TO EXPLAIN WHY THE RECOMMENDED PERSON WAS RECOMMENDED. Each recommendation includes the general similarity score (`score`), and the re-ranking score (`reRankingScore`) which uses several propensity signals (such as contact similarity, contact quality, title boosting, etc.) to refine relevancy. Note that `reRankingScore` returns -1.0 when the ML reranking model did not run. **IMPORTANT CONTEXT**: - The `score` and `reRankingScore` are not directly comparable. The `score` is a general similarity score, while the `reRankingScore` is a re-ranking score that uses several propensity signals (such as user platform interactions, contact similarity, contact quality, title boosting, etc.) to refine relevancy. A `reRankingScore` of -1.0 indicates the reranking model did not run. - The contact recommendations are updated daily. - The contact recommendations are powered by list engagement, CRM data, and propensity signals. - Higher scores does not guarantee responses. **WORKFLOW:** 1. Use the `search_companies` tool to retrieve the Company ID of the target company name, if the Company ID is not provided. 2. Use the Get Contact Recommendations tool to retrieve the contact recommendations. 3. **OPTIONAL** Use the `enrich_contacts` tool to retrieve the contact details. 4. **OPTIONAL** Use the contact details retrieved in step 3 and combine with the metadata retrieved in step 2 to identify appropriate business contacts for research and strategic planning purposes. **COMMON USE CASES:** - Who are the top 10 people to reach out to at Microsoft? (for prospecting) - Who are the top 5 people to reach out to at Apple based on similar successful past deals? - Why is this person recommended for my use case? CREDITS: Free to use