How We Run a Full Marketing Engine on n8n — With Zero Marketing Staff

I built a marketing machine that runs without a marketing team.

Not because I wanted to. Because I had no choice.

When I founded Innovatrix Infotech, I had a 12-person development team, clients to serve, and a company that needed to be found online. I had exactly zero budget for marketing headcount. Hiring a content writer, a social media manager, an email marketing executive, and a lead gen specialist would have cost Rs 3-5 lakh a month before a single rupee of ad spend. That wasn't happening.

So I did what any engineer would do when faced with a problem that looks like a systems problem: I built a system.

This post documents exactly what that system looks like — the workflows, the tools, the logic, and the real numbers after running it for the better part of a year. If you're an agency founder, a solo operator, or anyone trying to market a business without a team, this is the architecture I'd hand you.

---

The Stack

Four tools do the heavy lifting:

• n8n — the automation layer that connects everything (self-hosted on AWS EC2)
• Brevo — email marketing and lead nurture sequences
• ClickUp — content calendar, task management, and lightweight CRM
• Cal.com — discovery call booking with automated pre-call sequences

The entire stack runs on our existing AWS EC2 infrastructure. n8n is self-hosted, which means there are no per-workflow fees, no per-task charges, and no bill that scales with volume. The total monthly tool cost for the entire marketing operation is under Rs 8,000.

For context: equivalent output from human staff would cost Rs 3-5 lakh a month.

---

System 1: The Content Production Pipeline

The foundation is a 130-post blog calendar, planned across 26 weeks, built in ClickUp. Six content types: tutorials, listicles, thought leadership pieces, pricing guides, case studies, and comparison posts.

Here's what happens when a blog task becomes due:

1. An n8n workflow is triggered by the ClickUp task's due date
2. It fetches the task brief from ClickUp via API
3. It pulls the matching writing prompt from a ClickUp document (we have six prompt templates — one per content type)
4. It queries the Pexels API for a relevant featured image and imports it directly into Directus CMS
5. It sends the brief + prompt + image context to Claude AI with precise instructions: voice guidelines, EEAT requirements, internal linking rules, FAQ structure, word count logic
6. The complete draft — with FAQs, internal links, author bio, meta title, meta description — is published to Directus CMS via API and goes live

I then review every post before it's truly final. The AI handles the structural heavy lifting. I add the real opinions, the technical specifics, the client context that only I can provide.

Total time from brief to live post: under 45 minutes. A human writer doing the same work well would take 4-6 hours.

Over 26 weeks, that's roughly 400-500 hours of writing time saved — at zero cost beyond the infrastructure we were already running.

Why This Works for SEO

There's a persistent fear that AI-generated content doesn't rank. It's the wrong framing. Generic AI output doesn't rank. Content with genuine expertise signals, specific metrics, real experience, and technical depth ranks just fine.

Every post in our pipeline is built around EEAT principles:

• Experience: real client stories and outcomes embedded in every post
• Expertise: technical depth from a Head of Engineering background
• Authority: natural mentions of our Shopify Partner, AWS Partner, DPIIT recognition
• Trust: specific numbers — +41% mobile conversion for FloraSoul, Rs 4.2L launch revenue for Baby Forest

The result: 56,500+ impressions in Google Search Console in the first 90 days. CTR is improving as we systematically rewrite titles and meta descriptions for our Page 1 posts.

---

System 2: The Cross-Distribution Engine

A blog post that only lives on your website is a wasted asset. Every post we publish hits five channels automatically — and this is where the n8n workflow canvas does something most people don't expect.

The workflow runs every 5 hours on a schedule trigger. Here's the full node sequence:

1. Fetch Published Posts — pulls all live posts from Directus CMS via GET request
2. Fetch Publish Log — checks which posts have already been distributed (logged back to Directus after each publish)
3. Pick Next Post — selects the oldest un-distributed post from the queue
4. Has Post? Check — if nothing in queue, exits cleanly ("Done - No Posts Left" branch)
5. Generate AI Teasers — passes the post to an AI node that writes four platform-native versions: a developer-focused intro for Dev.to, a clean summary for Hashnode, a hook-first paragraph for LinkedIn, and a tight tweet for Twitter/X
6. Parse Teasers — extracts the four variants into structured data
7. Four parallel conditional branches, each checking if that platform needs a post:
   • Needs Dev.to? → Publish to Dev.to → Log Dev.to Result to Directus
   • Needs Hashnode? → Publish to Hashnode → Log Hashnode Result to Directus
   • Needs LinkedIn? → Post to LinkedIn → Log LinkedIn Result to Directus
   • Needs Twitter? → Post Tweet → Log Twitter Result to Directus

Every result is logged back to Directus CMS. The publish log means nothing gets double-posted, even if the workflow runs multiple times a day. Canonical tags on Dev.to and Hashnode point back to the original URL, preserving SEO authority.

What used to require five separate manual publishing sessions — logging into each platform, reformatting the content, adding the right tags, including the canonical URL, writing a platform-specific hook — now requires zero effort.

If manual cross-posting takes 45 minutes per post across five platforms, and we publish 5 posts a week, that's 3.75 hours a week, 15 hours a month, 180 hours a year. Eliminated.

---

System 3: Lead Capture -> ClickUp CRM

Every inbound signal — contact form submission, Cal.com booking, LinkedIn enquiry — routes into ClickUp via n8n.

The workflow:

1. Webhook receives the lead data (form submission or Cal.com booking payload)
2. n8n extracts name, company, email, service interest, message, and source
3. Creates a task in the ClickUp leads list with full context
4. Assigns priority based on service type and lead source
5. Triggers an immediate confirmation email via Brevo — within 3 minutes of submission

For Cal.com bookings specifically, the workflow goes further. It sends a 3-email pre-call sequence:

• Immediately: Booking confirmation + what to expect
• 24 hours before: Context-gathering email asking the prospect to share their brief, budget range, and timeline
• 2 hours before: Reminder with the meeting link

By the time I join the call, the prospect has been warmed, has received proof of our credentials, and has already shared their context. Discovery calls close at a materially higher rate because neither party is wasting the first 15 minutes on basics.

The number that matters most here: lead response time under 3 minutes, 24/7, including weekends and 2am enquiries from Dubai and Singapore. No lead sits in a form inbox waiting for someone to check their email on Monday morning.

---

System 4: Lead Nurture via Brevo

Not every enquiry converts immediately. Most don't. The buying cycle for a development project is weeks to months — especially in the B2B agency space.

Leads who enquire but don't book a call enter a 21-day nurture sequence in Brevo, enrolled automatically by n8n:

Leads who book at any point are immediately removed from the sequence. No one gets a nurture email after they've already converted.

Leads who complete the 21-day sequence without converting are tagged for a re-engagement campaign 30 days later — a single email, topical, no pressure.

The system doesn't replace relationship-building. It ensures no one falls through the cracks because I was busy shipping code for a client.

---

The Real Numbers

---

What This Is Not

This is not a set-it-and-forget-it system. The AI doesn't run alone. Every post I review. Every lead conversation I personally handle. The automation handles the volume work — the structuring, the distributing, the sequencing, the logging. The human judgment — editorial quality, client relationships, service delivery — stays with me.

The goal isn't to remove humans from marketing. It's to remove humans from the parts of marketing that don't benefit from being human.

Formatting a blog post for Dev.to doesn't benefit from human judgment. Writing the post in the first place — deciding what to say, what angle to take, what experience to draw from — does. So the AI does the former and I do the latter.

---

Can You Build This for Your Business?

Yes. We build this exact architecture for clients.

The core components — content automation, lead capture, CRM integration, email nurture, booking sequences — adapt to virtually any business type. D2C brands, SaaS companies, service businesses, ecommerce operators. If you have an existing tech stack and a clear sense of what your lead journey looks like, we can wire it together.

Most implementations run 2-4 sprints. The result is a marketing system that keeps working whether you're shipping product, serving clients, or asleep.

Want to see what this looks like for your business? Book a discovery call — and yes, the pre-call sequence will run automatically the moment you book.

---

Related Reading

If this post was useful, you might also want to read:

• Our AI Automation services — what we build for clients
• How Zevarly achieved +55% session duration with a Shopify rebuild
• Web development services for D2C brands

---

Rishabh Sethia is the Founder and CEO of Innovatrix Infotech — a DPIIT-recognized Shopify, AWS, and Google Partner agency based in Kolkata. Former Senior Software Engineer and Head of Engineering. He now runs the entire company's marketing through automation infrastructure built on n8n.