AI-Powered B2B Account Scoring System
Built an AI-Powered B2B Account Scoring System — Here's How It Works
One of my recent projects: a fully automated lead scoring pipeline built from scratch using n8n, Perplexity AI, and OpenRouter.
What it does:
- Takes a list of companies from Google Sheets
- Automatically researches each company using live AI web search
- Scores them on a custom 100-point matrix across 6 business signals
- Generates a personalized sales recommendation per account
- Writes all enriched data back to Google Sheets — fully automated
Tech behind it:
- n8n — 4 modular sub-workflows handling orchestration, research, scoring, and output
- Perplexity AI (sonar-pro) — real-time company research
- OpenRouter (Claude 3.5 Haiku) — structured data extraction + recommendation generation
- Google Sheets — input/output layer
Key concepts applied:
- Modular workflow architecture
- API chaining with data persistence between nodes
- Custom scoring logic with classification (Hot / Warm / Cold)
- Error handling and status management
The hardest part wasn't the AI — it was managing clean data flow between chained API calls.
Full project available on GitHub → github.com/aamir7788/n8n-account-scoring-system
Happy to answer any questions or discuss the architecture!
5
3 comments
AI-Powered B2B Account Scoring System
skool.com/ai-automation-society
A community built to master no-code AI automations. Join to learn, discuss, and build the systems that will shape the future of work.
Leaderboard (30-day)
1
+4235
2
+1732
3
🔥
+1375
4
+1319
5
🔥
+636
Powered by