The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Why do Make.com scenarios break in production, and what fixes each failure? After 1,200+ automations for 210+ clients, the same seven failures account for almost every rescue job I take on. Ranked by how often I see them.
1. No error route on an external call. An API returns a 429 or a timeout, the scenario stops, and nobody knows until a customer does. Fix: an error handler on every external module, retry with backoff, then a route to an alert with the record and the reason. That is the layer in the image.
2. All-or-nothing batches. One malformed record stops fifty documents. Fix: one record per run, park the failure with its reason, keep going. A US services firm went from 0% to 75% of a bad batch shipping on time this way.
3. Acknowledging too late. TradingView drops a webhook if you do not answer within about 3 seconds. Fix: acknowledge first, process after. That is how alerts reach Telegram in under 2 seconds, 24/7.
4. Ignoring the destination's rate limit. Telegram silently drops messages over roughly 20 a minute per channel. Fix: a per-channel queue and a dedupe key so a retry never posts twice.
5. Retries that write twice. A retried chunk creates a duplicate row or a duplicate email. Fix: idempotency keys on anything that writes, and honour Retry-After headers.
6. Logic spread across a dozen small scenarios. Every rule change touches six places. Fix: routers by case inside one well-structured scenario. A 198-module launch scenario has held 99.9% uptime because every rule lives in one place and every failure is handled where it happens.
7. No handover. The person who built it leaves, and the team is renting a mystery. Fix: documentation of every path and failure mode plus a recorded walkthrough, on an account the client owns.
Case studies for 2, 3 and 6 are on my profile. If one of these seven is happening in your account right now, the 5-day audit on my services page is built to find it.
Prem Patel, Nex Automations. Make.com Level 5 certified, official Make partner, Zapier Certified Expert.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
From my first Zap to Make.com Level 5: six years of building automation that runs businesses
I started six years ago with the smallest kind of automation: a form response landing in a sheet, an invoice that sent itself, a Zap that saved someone ten minutes a day. Client by client on freelance marketplaces, those ten-minute fixes turned into a company, Nex Automations, and into 1,200+ automations running in production today for 210+ clients across 12+ countries.
Somewhere along the way I learned what automation actually is. It is not robots replacing people. It is the invoice cycle of a US transport company dropping from 30 days to 7, a 77% cut, because the paperwork now moves itself. It is a lead reaching the right salesperson in under 2 seconds instead of sitting in an inbox overnight. It is an operations team getting 15 to 20 hours back every week for the work that actually needs a human.
The tools shape the work. Zapier is where most businesses should start, fast to build and hard to break for simple links between apps. Make.com is where the serious systems live, visual scenarios with routing, iterators and real error handling. Self-hosted n8n is where I put AI agents when the data cannot leave the client's server. Knowing all three means the problem picks the tool, not the other way around.
Along the way Make certified me at Level 5, the top of their ladder, and Nex Automations became an official Silver Partner, listed in both directories:
make.com/en/partners-directory/nexautomations
zapier.com/partnerdirectory/partner/nex-automations
Every build still ships the way the first one should have: error handling, logging, alerting and documentation, because an automation only counts if it holds at 2 AM.
Happy to answer any Make or Zapier questions in the comments.🚀
This is the difference between automation and an AI agent. 🤖
A traditional automation says:
“When X happens → do A → then B → then C.”
It’s predictable, reliable, and perfect for repetitive processes. But what happens when the next step isn’t always the same? That’s where an AI agent comes in.
Instead of following one fixed path, an agent can:
→ Understand the goal
→ Analyze the situation
→ Decide what information it needs
→ Choose the right tool
→ Take an action
→ Check the result
→ Decide what to do next
For example, imagine a lead-generation workflow in n8n.
Traditional automation:
New lead → Scrape website → Find email → Add to CRM → Send email
Every lead follows the same path.
Agentic workflow:
New lead → AI Agent
The agent decides:
🔎 Does this company fit the target profile?
🌐 Should I research the website further?
📊 Is there enough information to qualify the lead?
✉️ Should I generate a personalized message?
⚠️ Is human approval needed before outreach?
The workflow becomes goal-driven instead of purely step-driven.
And this doesn't mean replacing automation.
The most powerful systems combine both:
n8n handles the reliable execution.
AI handles decisions where the path isn't always predictable.
That’s what makes an agentic workflow different from simply adding ChatGPT to an automation.
The future isn't: Automation OR AI
It's: Automation + AI agents + the right tools.
Great visual! "Automation does the work, AI agent decides the work" is probably the clearest way to explain this shift to clients.
I built my own lead generation machine because I moved to Denmark with zero network.
I moved to Denmark.
I didn't know anyone.
I had no network.
And I needed to find companies I could sell automation services to.
I could spend hours manually searching for companies, finding emails, cleaning data, and sending messages one by one.
Or I could automate the process.
So I built a complete lead generation pipeline:
→ 🐍 Python scraper connected to Denmark's public government CVR database
→ 📊 Google Sheets synchronization to normalize and deduplicate companies
→ ✉️ Automated cold email workflow with 10 sector-specific templates using n8n
→ 📱 Daily Telegram reports with full pipeline statistics
→ 💼 Assisted LinkedIn outreach: 10 leads per day + a ready-to-send message
And these are the numbers:
📈 19,334 leads in the database
📧 71% email coverage
🏭 10 sectors
🔢 34 industry codes
✉️ 25 automated emails per day
♾️ ~534 days of outreach without repeating a lead
💰 $5/month total cost — basically just the VPS.
The interesting part isn't the scraper.
It's not n8n.
It's not Telegram.
The interesting part is that a process that used to take hours of manual work can now run almost entirely on its own.
And that's exactly what I want to build for companies:
Find repetitive processes → design the system → automate them → measure the results.
In the video, I break down the entire pipeline, piece by piece.
If you're building something similar, let me know which part you'd like me to break down next. 👇
#Automation #AI #LeadGeneration #n8n #Python #ColdEmail #SalesAutomation #B2BAutomation #Denmark #BuildInPublic
Really impressive system. The part that caught my attention is the automated cold email workflow.
Once you start scaling the volume, keeping deliverability consistent across domains and inboxes becomes just as important as the automation itself. Great build!