Research Automation with n8n, Apify, Relevance AI, and Perplexity

Client: AI-Powered Sales Enablement Team

Industry: B2B Sales Intelligence / Marketing Automation

Platform: n8n, Apify, Perplexity API, Relevance AI, OpenAI, Google Sheets, Gmail

Time to Build: 15+ hours

Time Saved: 6–8 hours per week

Consulting Partner: Lumio Consulting

Project Overview / Problem

A client came to me with a time-intensive problem: they needed to prepare personalized reports on individual prospects and their companies for sales enablement. This included reading LinkedIn profiles, recent posts, company background, and TrustPilot reviews—then distilling that data into pain points and actionable insights for their sales team.

Their manual process took 45+ minutes per lead and was inconsistent in quality, depending on who was doing the research.

The client wasn’t trying to replace their SDRs—they just wanted a consistent and scalable way to generate insights and tailor messaging across their outbound campaigns.

Goals of the Automation

The objective was to build a fully automated pipeline that could:

  • Scrape public LinkedIn profile content (posts, job history, education)

  • Enrich company background via Perplexity AI and TrustPilot reviews

  • Generate a tailored profile summary and list of likely pain points

  • Automatically compile and email a clean HTML report to the sales team

Research Automation with n8n Perplexity Relevance.AI Apify OpenAI

Solution Overview / Step-by-Step Breakdown

This automation was built entirely in n8n, with a multi-step process that combined web scraping, AI enrichment, and report generation.

Step 1: Scrape the Prospect Profile

Using Relevance AI, we scraped LinkedIn profile content including:

  • Recent posts

  • Job history

  • Educational background

Step 2: Research the Company

A Perplexity API call was used to summarize the company’s background, extracting:

  • Recent updates

  • Industry positioning

  • Competitive insights

We also used Apify to scrape 1–3 star TrustPilot reviews and parse them into meaningful categories.

Step 3: Analyze & Summarize with GPT

We fed all the scraped and enriched data into OpenAI (GPT-4) to:

  • Write a combined person + company profile

  • Identify likely pain points and suggest tailored solutions

Step 4: Generate the Report

We converted all insights into an HTML-formatted report using a custom template and emailed it directly to the sales team with a subject line including the prospect’s name.

Challenges

A few technical and strategic roadblocks came up during development:

  • Rate-limited sources: LinkedIn and TrustPilot required careful pacing and fallback logic to avoid failed requests.

  • Multi-source data parsing: Normalizing posts, reviews, and education history into structured prompts took significant prompt engineering.

  • Model context limitations: Feeding long scraped data into GPT required chunking and smart summarization to stay within token limits.

  • Email deliverability: HTML formatting needed to be light, readable, and clean to avoid spam filters when sent via Gmail.

Results/Outcome / Time Saved

This automation reduced research time from 45+ minutes per prospect to under 2 minutes, thanks to step-by-step automation and minimal manual review.

In just the first month of use:

  • Time savings: 6–8 hours per week

  • Report quality: more consistent and detailed than human-generated briefs

  • Personalization: pain points and summaries felt custom-built, without actually requiring a writer

Client Feedback

“We used to spend half an hour researching each lead before outreach. Now it’s all there—clean summaries, company context, even pain points we hadn’t thought about. Tom made it easy for us to scale research without losing the human touch.”
— Sales Strategy Lead

Additional Improvements

Since the initial rollout, we’ve added:

  • A Slack integration that posts the report directly into the sales channel

  • A fallback routine that retries failed scrapes or AI calls without breaking the full run

  • The ability to tag reports by region, industry, or campaign for batch analysis later

Tools, Plug-ins, and Platforms Used

  • n8n – Core orchestration and scheduling

  • Relevance AI – LinkedIn profile scraping

  • Perplexity API – Company research summaries

  • Apify – TrustPilot review scraping

  • OpenAI (GPT-4) – Custom summaries and pain point extraction

  • HTML Generator – Dynamic templating for email-ready formatting

  • Gmail API – Email delivery

  • Slack (optional) – Notification and sharing

Interesting in implementing something similar, or in exploring other automations for your business, click here to schedule a quick discovery call.

Previous
Previous

Automating Midjourney Image Generation

Next
Next

Automating Lead Response and Call Routing with n8n and GoHighLevel