Trade Show Exhibitors & Attendees Finder — who exhibited, who was there, pay per company and per person

Give it the name of a trade show (or its exhibitor list URL) and get the exhibitor list with the people who were probably there: everyone who publicly announced or reported being at the show on LinkedIn (posts and comments) or on the company's own website, each with the evidence and a likelihood score, plus the booth staff found by role. Find who attended, build event attendees lists and exhibitor contacts for trade show leads, and tell it what kind of exhibitors you are looking for so it keeps only those. $0.004 per exhibitor delivered, $0.010 per person with evidence, $0.004 per person found by role.

Built for automation and AI agents: nothing is required, the empty input returns a fictional sample show so you can see the output shape without spending anything, and every result is a plain dataset item with the people nested in it. The third-party Actors it calls on your behalf (Google search, Skython exhibitor list, harvestapi LinkedIn) run under your account at their public prices, within the budget you set (maxSubActorSpendUsd, 10 USD by default): an estimate item comes first and estimateOnly stops there.

What you get

One item per exhibitor, with its people inside:

{
  "type": "company",
  "name": "Northwind Robotics",
  "website": "https://northwind-robotics.example",
  "linkedinUrl": "https://www.linkedin.com/company/northwind-robotics-sample",
  "country": "Germany", "hall": "Hall 4", "stand": "A12",
  "categories": ["Robotics"], "description": "Collaborative robots for packaging lines",
  "companyType": "manufacturer",
  "targetFit": 0.75, "targetFitLevel": "mostly matches", "targetConfidence": 0.9,
  "exhibitorConfirmedBy": "website",
  "webEvidence": [{"url": "https://northwind-robotics.example/news/sample-expo-2026", "title": "Northwind at Sample Expo 2026",
                   "date": "2026-05-20", "excerpt": "Visit us in Hall 4, booth A12...", "nature": "announces_participation"}],
  "peopleEvidenced": 1, "peopleRoleBased": 1,
  "people": [
    {"personName": "Anna Keller", "personLinkedinUrl": "https://www.linkedin.com/in/sample-anna-keller-0000",
     "headline": "Head of Sales at Northwind Robotics", "companyName": "Northwind Robotics",
     "exhibitorMatch": "exhibitor", "isExhibitorStaff": true, "source": "post-author",
     "attendanceScore": 1.0, "attendanceLevel": "certain", "confidence": 0.9, "scoredBy": "jev",
     "roleBasis": {"title": "Head of Sales", "keywords": ["sales"]},
     "evidence": [{"kind": "post", "url": "https://www.linkedin.com/posts/...", "date": "2026-06-05",
                   "excerpt": "What a week at #SampleExpo2026! Thank you to everyone who stopped by booth A12...",
                   "nature": "reports_presence", "onSite": 0.97}]},
    {"personName": "Jonas Weber", "headline": "Marketing Manager at Northwind Robotics", "source": "employee-search",
     "attendanceScore": 0.5, "attendanceLevel": "likely", "roleBasis": {"title": "Marketing Manager", "keywords": ["marketing"]}, "evidence": []}
  ]
}

Also in the dataset: attendees items for the people with evidence who are not staff of an exhibitor (visitors, organizers, speakers, press), in parts of 100; an estimate item first (what the third-party Actors may bill you at most, and what this Actor may charge at most); an event item last (the resolved edition, counts, sub-runs with their cost, warnings). The People view of the dataset (or items?unwind=people in the API) gives one row per person; the key-value store holds SUMMARY (updated after every step) and JUDGEMENTS (every judgement, for audit).

How people are found and scored

• LinkedIn posts and comments: posts found with the show's hashtags (#IMTS2026) and name within the window from 60 days before the show to 30 days after it, with their comments (and reactions when evidenceSources includes reactions). Each post and comment is judged: announces presence, reports presence, mentions only, unrelated. Authors and commenters become people with that evidence.
• Company websites and blogs: the sitemap or the news pages of each targeted exhibitor, read for pages that mention the show; a page that announces or recaps the participation confirms the exhibitor (exhibitorConfirmedBy: website).
• Roles: for each targeted exhibitor with a LinkedIn page, its employees whose current title matches roleKeywords (sales, business development, marketing, CEO, founder, managing director... by default), as booth staff.

Every person gets an attendanceScore from 0 to 1 and its level: no sign (0), weak sign (0.25: mentioned the show or reacted to a post), likely (0.5: announced coming or exhibiting, or a booth-facing role at a confirmed exhibitor), strong (0.75: reported being there), certain (1.0: clearly on site). People under minAttendanceScore (0.3 by default) are not delivered and not charged. Every judgement is made by Jev, a decision model, from the texts themselves; scoredBy says jev, or fallback when the model was unavailable and a keyword rule took over.

Targeting: describe the exhibitors you want in targetCompanies ("manufacturers of CNC machine tools and automation equipment"). Each exhibitor gets a targetFit from 0 to 1 and a companyType (manufacturer, distributor or reseller, service provider, software or data, association, press or institution). Only exhibitors at or above minTargetFit (0.5) are enriched and charged. Without targetCompanies, every exhibitor is kept.

Costs

Who bills	What	Price
This Actor	exhibitor delivered (company-qualified)	$0.004
This Actor	person with evidence (person-evidenced)	$0.010
This Actor	person found by role (person-role-based)	$0.004
apify/google-search-scraper (your account)	resolving the show by name, two result pages	about $0.006
skython/exhibitor-list-scraper (your account)	exhibitor read from the catalog (20 at least)	$0.0045 + $0.01 per run
harvestapi/linkedin-post-search (your account)	post, comment or reaction read	$0.002
harvestapi/linkedin-company-employees (your account)	per company searched, plus per profile	$0.02 + $0.003

The third-party Actors are called with a cap each (Apify imposes at least $0.30 per call, $0.50 for Google search); the sum stays within maxSubActorSpendUsd. The estimate item shows the maximums before anything is spent; estimateOnly: true stops there. Example: 500 exhibitors read, 150 targeted, 600 people with evidence and 600 by role: about $9 for this Actor and $9 to $11 billed by the third-party Actors.

Input

By name (the official site, the exhibitor list and the dates are found with a Google search):

{"eventName": "IMTS 2026", "targetCompanies": "manufacturers of CNC machine tools and automation equipment"}

By exhibitor list URL, with dates:

{"eventName": "Eurosatory 2026", "exhibitorListUrl": "https://www.eurosatory.com/en/catalogue/", "startDate": "2026-06-15", "endDate": "2026-06-19",
 "targetCompanies": "drone and counter-drone makers", "maxCompanies": 300}

From the dataset of an exhibitor list scraper you already ran (or as an integration after it):

{"eventName": "Hannover Messe 2026", "companiesDatasetId": "YOUR_DATASET_ID", "includeRoleBased": false}

LinkedIn only, with your own company list:

{"eventName": "Web Summit 2026", "companies": ["Northwind Robotics", "https://contoso.example"], "evidenceSources": ["posts", "comments"]}

Fields: Trade show (eventName, year, startDate, endDate, eventUrl, exhibitorListUrl, eventLinkedinUrl, hashtags), Exhibitors (companies, companiesDatasetId, targetCompanies, excludeCompanies, minTargetFit, includeOffTargetCompanies, maxCompanies), People (evidenceSources, maxPosts, maxEngagementsPerPost, includeRoleBased, roleKeywords, maxPeoplePerCompany, minAttendanceScore, maxPeople), Budget (maxSubActorSpendUsd, estimateOnly). Aliases: event, tradeShow, name, url, exhibitors, datasetId.

Limits

• The exhibitor catalog is read by Skython's Exhibitor List Scraper, which supports about thirty catalog platforms (Map Your Show, Messe Frankfurt, Koelnmesse, Messe Düsseldorf, Messe Stuttgart, Reed, a2z, Ungerboeck...); tested here on IMTS (Map Your Show). When the listing page is on a platform it does not read (The Battery Show, for instance), the run says so in its warnings and continues in LinkedIn-only mode: people from posts, companies from the posts that speak for an exhibitor. Give companies or companiesDatasetId to add the exhibitors yourself.
• A reaction is weak evidence (level 0.25) and costs $0.002 to read: off by default.
• People named on company web pages are not extracted yet (the page confirms the exhibitor only).
• When the dates are unknown, the whole year is the window; give startDate and endDate for sharper results.
• No e-mail, no phone number.

Data and privacy

Public data only: exhibitor catalogs, public LinkedIn posts, comments and profiles read without any account or cookie, public web pages. No e-mail, phone number, photo or sensitive attribute is collected; nothing is stored outside your own run. Presence is always given as a probability with its evidence, never as a fact, and visitors who are not exhibitor staff appear only through what they wrote publicly. Use the results for B2B prospecting and partnerships under your own legitimate-interest assessment and local law. Removal requests: maxence.fournaux+sherwood@gmail.com.

Related Actors

• Google Maps Scraper API: places with free verdicts and AI qualification.
• Review Analyzer API: sentiment, topics and unanswered complaints of any reviews dataset.