48 lines
1.1 KiB
Markdown
48 lines
1.1 KiB
Markdown
# 📅 Day 4 — Implementing Intelligent Alerts
|
|
|
|
## 🎯 Goal
|
|
Make the workflow proactive — detect issues automatically and notify the right people.
|
|
|
|
---
|
|
|
|
## 🧩 Tasks
|
|
|
|
### 1. Branch the Workflow
|
|
- After the Python node, create two branches:
|
|
- One for **reporting/storage**.
|
|
- One for **alerts**.
|
|
|
|
---
|
|
|
|
### 2. Define Alerting Logic
|
|
- Add an **If Node**.
|
|
- Configure the following conditions:
|
|
- `avg_sentiment_score < -0.4`
|
|
- `num_negative_reviews > 5`
|
|
|
|
---
|
|
|
|
### 3. Craft Alert Message
|
|
- Add a **Set Node** on the “True” branch.
|
|
- Example:
|
|
🚨 URGENT: Negative Sentiment Spike!
|
|
Product: {{ $json.product_name }}
|
|
Avg Sentiment: {{ $json.avg_sentiment_score }}
|
|
Please investigate immediately.
|
|
|
|
yaml
|
|
Copy code
|
|
|
|
---
|
|
|
|
### 4. Configure Alert Notification
|
|
- Add a **Slack Node** connected to the alert.
|
|
- Send the alert to a **high-priority channel** like `#ecommerce-alerts`.
|
|
- Optionally, tag users (`@channel`) for instant visibility.
|
|
|
|
---
|
|
|
|
## ✅ Deliverable
|
|
A proactive alerting system that:
|
|
- Detects negative sentiment or unusual review spikes.
|
|
- Sends urgent alerts directly to Slack. |