Qualify trade show exhibitors against your company profile
Posted on Aug 29, 2025 • Edited on Aug 29, 2025. Task: The author built an n8n workflow to qualify trade show exhibitors. Bright Data scrapes the exhibitor list, extracting names and URLs. An AI Agent (GPT-5-mini) uses SerpAPI and Bright Data to research each company, evaluating size and tech stance against a specific client profile. Results are scored and appended to a Google Sheet. The author manually reviewed the output to prioritize booth visits.; tools: n8n. Human role: The original post does not say how a person stayed involved.; limitations: The workflow is stateless, evaluating each company independently without cross-referencing. API costs for SerpAPI and Bright Data scale with the number of exhibitors. The author notes that the initial version uses small batches for testing, implying potential latency or cost issues at full scale. Human review is required to validate AI scores before outreach..
Tools
Published on Agent Work Atlas:
Overview
Posted on Aug 29, 2025 • Edited on Aug 29, 2025
Workflow
The author built an n8n workflow to qualify trade show exhibitors. Bright Data scrapes the exhibitor list, extracting names and URLs. An AI Agent (GPT-5-mini) uses SerpAPI and Bright Data to research each company, evaluating size and tech stance against a specific client profile. Results are scored and appended to a Google Sheet. The author manually reviewed the output to prioritize booth visits.
Reported result
The source does not report measured business outcomes, conversion rates, or time savings. It describes the workflow as a functional prototype for lead qualification. No specific metrics on the quality of the generated leads or the efficiency gain compared to manual research are provided in the text.
Human role
The original post does not say how a person stayed involved.
Limitations
The workflow is stateless, evaluating each company independently without cross-referencing. API costs for SerpAPI and Bright Data scale with the number of exhibitors. The author notes that the initial version uses small batches for testing, implying potential latency or cost issues at full scale. Human review is required to validate AI scores before outreach.
Try it
Act as an automation engineer. I need to build an n8n workflow to qualify trade show leads. Inputs: - Trade show exhibitor list URL: [replaceable URL] - My company profile and ideal client criteria: [replaceable text] - Target Google Sheet ID: [replaceable ID] Steps: 1. Use Bright Data Web Scraper to fetch the exhibitor list and extract company name, description, booth number, and website URL. 2. Configure an AI Agent node using GPT-5-mini. 3. Equip the agent with SerpAPI for web search and Bright Data for targeted scraping. 4. Instruct the agent to evaluate each company against my ideal client profile (size, tech interest, modernity) and assign a score (Low/Medium/High) with reasoning. 5. Append the structured results to the Google Sheet. Safeguards: - Run a dry test with only 3 companies first. - Show me the JSON output and the Google Sheet rows for these 3 companies. - Do not process the full list or send any emails until I confirm the scoring logic is accurate. - If the scraper fails on dynamic content, report the error and suggest a fallback method.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.