DOGMIS: Building an AI-Powered Stakeholder Intelligence Platform with n8n

Help me Build my Workflow
1

Hello n8n Community,
My name is Dr. Demir Ozdemir.
I am an independent international business development consultant working across APAC, MENA, Europe, and the animal health sector.
I am currently developing an early-stage project called DOGMIS (Distributed Operational Governance & Market Intelligence System).
The goal is to build an AI-assisted platform that combines:

  • Stakeholder relationship management
  • Communication monitoring
  • Market intelligence
  • Business development tracking
  • AI-assisted decision support
    My current workflow includes:
    Gmail → Stakeholder Extraction → Relationship Mapping → AI Analysis → Business Intelligence Dashboard
    I am currently testing Gmail monitoring, stakeholder identification, communication tracking, and automated business intelligence workflows using n8n Cloud.
    As an individual consultant, the project is entirely self-funded and operates with a highly limited budget.
    I would appreciate advice from experienced n8n users:
  1. Is the Starter plan sufficient for this type of project?
  2. At what stage would you recommend moving to self-hosted deployment?
  3. Has anyone built a CRM, stakeholder intelligence, or relationship management system using n8n?
  4. What practical workflow execution limits have you experienced in real-world deployments?
    Any feedback or recommendations would be greatly appreciated.
    Thank you.
    Dr. Demir Ozdemir
    Founder, DOGMIS Project

Describe the problem/error/question

What is the error message (if any)?

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

Information on your n8n setup

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:
3

Hello Dr. Demir

The Starter plan (which includes 2,500 executions per month) is usually more than enough for the development and prototyping phase, but it will likely become a bottleneck once the system goes live because each email will count as a trigger to an execution which will make eat away your quota.

Since your project is self-funded with a limited budget, moving to a self-hosted deployment early in the production stage is highly recommended, especially when you exceed 2,500 executions a month, hosting n8n yourself on a VPS like Hostinger, DigitalOcean, or Hetzner will cost you roughly $5 to $10 a month for unlimited executions.

Yes, this is a very common and powerful use case. But, typically, users do not store the actual CRM data inside n8n, instead, they use n8n as the brain and connect it to a lightweigt, low-cost database or backend, for example, n8n+Airtable or BaseRow.

In real-world, I have experienced LLM payloads, when passing huge, large chunks of texts or massive market research PDFs into AI nodes simultaneously can cause n8n to crash, so at least 2- 4 GB of RAM is recommended.
There is also rate-limiting when Google and your AI providers have strict API rate limits. If your workflow processes 100 emails all at once, the external APIs might block you temporary. Always implement error handling or split data into smaller batches.
The default database for n8n is SQLite. If you stick with n8n Cloud or a small self-hosted instance, SQLite is fine for thousands of logs. But if you keep years of execution history saved on a self-hosted instance, it will slow down. So my tip is Set EXECUTIONS_DATA_PRUNE=true to delete old execution logs automatically

Hope this helps you.