AI Trend-Spotting System for Creative Agency

Project Summary

Beng Beng is a creative agency that produces viral video content and develops brand positioning for their clients. Their success depends on understanding what audiences actually think and feel about products, brands, and trends—but manual social media research is slow and incomplete.

We built an AI system that monitors comments and engagement across Instagram, YouTube, TikTok, Reddit, and X (Twitter), then analyzes sentiment and delivers actionable insights in formatted PDF reports with source links.

‍The results: 30% reduction in campaign research time, with deeper insights that inform video production and brand strategy.

How It Works

The creative team submits a research request—a brand name, product category, or trend topic. The AI scrapes relevant conversations across all five platforms, analyzes sentiment patterns, identifies what resonates with audiences, and compiles everything into a comprehensive PDF report. The entire process takes minutes instead of hours.

The Challenge

Creative agencies need to understand market sentiment to make smart decisions: What do people actually think about a brand? What messaging resonates? What pain points keep appearing? Which trends are gaining momentum?

Traditional research means manually scrolling through comments across multiple platforms, trying to identify patterns while missing context. It's time-consuming, incomplete, and hard to scale. Beng Beng's creative director needed a way to get comprehensive market insights quickly, without sacrificing depth or accuracy.

Our Solution

We built an AI-powered research system using n8n that operates through three integrated components:

Multi-Platform MonitoringThe system connects to five major social platforms—Instagram, YouTube, TikTok, Reddit, and X—and scrapes relevant content based on the research query. It's not just collecting posts; it's diving into comments, engagement patterns, and discussion threads where real opinions live.

For a query like "MediaMarkt," the system finds every mention, review, complaint, and discussion across all platforms. For trend research like "loafers in Dubai," it identifies emerging patterns and what people actually care about.

Intelligent Sentiment Analysis - Raw data becomes actionable insights. The AI analyzes collected content to identify:

• Overall sentiment (positive, negative, neutral) with specific examples
• Recurring themes in what people love or hate
• Pain points and desires that appear repeatedly
• Emerging trends and shifting perceptions
• Audience language patterns and terminology

This isn't surface-level analysis—the system connects dots across thousands of comments to find meaningful patterns a human researcher would take days to uncover.

Formatted PDF Reports - All insights get compiled into professional PDF reports with clear sections, data visualizations, and direct links to source content. The creative team gets everything they need to make informed decisions: what the market thinks, why they think it, and proof via actual customer voices.

Reports are designed for immediate action—easy to scan, shareable with clients, and formatted for presentation use.

Real-World Application

When Beng Beng works with clients in competitive markets, they use the system to understand market sentiment before making creative decisions.

For example, if they're working with a shoe brand in Dubai, they can query "loafers Dubai" or "luxury sneakers UAE" to instantly understand:

• What local consumers are saying about these product categories
• Which features and benefits resonate most with the target audience
• What pain points keep appearing in customer discussions
• How competitors are being perceived in the market

Or for a consumer electronics client, they can search "MediaMarkt" to discover:

• Overall brand sentiment and customer satisfaction
• Common complaints and praise patterns
• What customers wish the brand would do differently
• How the brand compares to competitors in customer minds

This intelligence directly informs their creative strategy: which angles to emphasize in videos, how to position the brand, what pain points to address. The result is content that resonates because it's grounded in actual market sentiment, not assumptions.

The Technical Architecture

Built on n8n with specialized scraping and analysis capabilities:

• Custom scrapers for Instagram, YouTube, TikTok, Reddit, and X
• OpenAI for sentiment analysis and insight extraction
• Apify for robust data collection at scale
• PDFEndpoint for professional report generation
• Telegram integration for instant delivery and notifications
• Memory system to track research history and avoid duplication

The modular architecture allows easy expansion—new platforms can be added, analysis frameworks can be refined, and report formats can be customized based on client needs.

The Interface

The system is accessed via Telegram, making it dead simple for the creative team. They send a message with their research query, and the AI handles everything else. Progress updates appear in real-time, and the final PDF report arrives directly in the chat with all sources linked.

No complex dashboards, no learning curve—just natural conversation that triggers powerful research.

The Results

Beng Beng cut campaign research time by 30%, but the bigger win is insight quality. They're making creative decisions based on comprehensive market intelligence instead of gut feel or limited manual research.

Their video content performs better because it addresses what audiences actually care about. Brand positioning resonates because it's informed by thousands of real customer voices. Client campaigns deliver stronger results because the creative strategy is grounded in data.

The creative director now operates at a higher level—spending less time on research drudgery and more time on strategic thinking and creative execution. The AI handles the heavy lifting of market intelligence, delivering the insights that drive better creative work.