La sfida
Daily operations were fragmented: new customers were activated manually after each purchase, contacts arrived from different sources with no system to collect and route them, backups were forgotten or done by hand. The automation software used until then charged based on volume — as the business grew, that cost had become unsustainable.
La soluzione
A centralized automation system that handles the entire operational cycle on its own: when a new customer arrives, their access is activated automatically within minutes of purchase. Contacts from campaigns, forms, and referrals flow into a single point, are evaluated, and assigned to the right sales rep via real-time notification. Backups run every night and send an immediate alert if something goes wrong. Everything runs on owned infrastructure: fixed cost, independent of volume, stable for over two years.
Risultati
💰 -99% cost vs previous software
📈 Millions of operations/month at fixed cost
🔁 Customer access: from manual to automatic in minutes
🛡️ Stable in production for 2+ years
Stack tecnico
- n8n self-hosted
- Docker + PostgreSQL
- BullMQ (job queue)
- Cloudflared tunnel
- Portainer + Watchtower
- Telegram Bot API
- Keap / CRM API
The cost of “pay-per-operation” software
When you pay for every operation executed, cost grows with the business. In the beginning it’s negligible. At scale it becomes one of the highest line items in the operating budget — for activities that don’t need a variable cost. A self-hosted system on owned infrastructure has a fixed monthly cost that doesn’t depend on how many operations it runs.
What runs on this system every day
Every new purchase triggers an automatic sequence: access is activated, the welcome message sent, the learning path started — all within minutes, without anyone touching anything. Contacts arriving from campaigns, forms, and referrals are evaluated by source and assigned to the right sales rep with a real-time Telegram notification. Nightly database backups run on their own and send an immediate alert if something goes wrong — not the next day, at the moment it happens.
Reliability in production
An automation system is only useful if it doesn’t break when you need it. Critical operations — access activation, data sync, communications — have automatic retry mechanisms: if the downstream system is temporarily unreachable, the operation is picked up as soon as it’s available again, not lost. Portainer and Watchtower keep containers updated and monitored. No customer has been locked out due to a momentary technical issue in over two years of continuous operation.