Scheduled Analysis & Monitoring

Overview

Scheduled Monitoring allows you to configure automated sentiment analysis jobs that run at set intervals (hourly, daily, weekly, or custom cron schedules). The system monitors your chosen topics, evaluates alert conditions, and delivers insights directly to your team via email, Slack, or custom webhooks.

Getting Started

Open Schedule Configuration

Click “Schedule Analysis” from the dashboard header or within a topic detail panel.

Select Topic to Monitor

Enter the topic name or hashtag you want to track. You can add multiple topics using the + button.Example: AI Regulation or #ClimateAction

Set Monitoring Frequency

Choose how often the analysis should run:

Every hour: For breaking news, product launches, or crisis situations
Every 6 hours: For ongoing campaigns or moderate-priority topics
Daily: For trend tracking and weekly reports
Weekly: For long-term brand health monitoring
Custom (cron): Advanced scheduling using cron syntax

Configure Alert Thresholds

Set the conditions that trigger alerts (see Alert Conditions section below).

Choose Delivery Channels

Configure where reports are sent: Email, Slack, or Webhook.

Save & Test

Click “Test Now” to verify your configuration, then “Save Schedule” to activate monitoring.

The Schedule Analysis modal provides comprehensive control:

Topic Selection
Frequency Options
Report Types
Delivery Configuration

// Source: ScheduleAnalysisModal.tsx (lines 68-76)
<div className="space-y-3">
  <h4 className="text-sm font-semibold flex items-center gap-2">
    <Pin className="h-4 w-4 text-primary" /> Topic to Monitor
  </h4>
  <div className="flex gap-2">
    <Input defaultValue="AI Regulation" className="flex-1" />
    <Button variant="outline" size="icon">
      <Plus className="h-4 w-4" />
    </Button>
  </div>
</div>

Features:

Single input field for topic name or hashtag
+ button to add multiple topics to the same schedule
Auto-saves previous selections for quick re-use

// Source: ScheduleAnalysisModal.tsx (lines 78-104)
<RadioGroup defaultValue="6h">
  <RadioGroupItem value="1h" label="Every hour" />
  <RadioGroupItem value="6h" label="Every 6 hours" />
  <RadioGroupItem value="24h" label="Daily" />
  <RadioGroupItem value="7d" label="Weekly" />
  <RadioGroupItem value="custom" label="Custom (cron)" />
</RadioGroup>

Custom Cron Examples:

0 */3 * * * = Every 3 hours
0 9 * * 1-5 = Weekdays at 9 AM
0 0 * * 0 = Every Sunday at midnight

// Source: ScheduleAnalysisModal.tsx (lines 107-130)
<Checkbox id="c1" label="Executive Summary" defaultChecked />
<Checkbox id="c2" label="Full Analysis" defaultChecked />
<Checkbox id="c3" label="Crisis Alerts Only" defaultChecked />
<Checkbox id="c4" label="Trend Comparison" defaultChecked />

Report Types:

Executive Summary: High-level overview (1 page)
Full Analysis: Complete sentiment breakdown with AI insights
Crisis Alerts Only: Only sends when alert thresholds are met
Trend Comparison: Shows how metrics changed since last report

// Source: ScheduleAnalysisModal.tsx (lines 132-151)
<div className="space-y-3">
  <Input 
    icon={<Mail />}
    defaultValue="[email protected]" 
    placeholder="Email addresses (comma-separated)" 
  />
  <Input 
    icon={<MessageSquare />}
    defaultValue="#sentiment-alerts" 
    placeholder="Slack channel" 
  />
  <Input 
    icon={<Webhook />}
    defaultValue="https://api.company.com/webhook" 
    placeholder="Webhook URL" 
  />
</div>

Multi-channel delivery:

Email: Supports multiple addresses (comma-separated)
Slack: Webhook URL or channel name (#sentiment-alerts)
Webhook: POST JSON payload to your API

Alert Conditions & Thresholds

Define precisely when alerts should fire using interactive sliders:

1. Sentiment Drop Threshold

// Source: ScheduleAnalysisModal.tsx (lines 160-167)
<div className="space-y-3">
  <Label className="flex items-center gap-2">
    <TrendingDown className="h-4 w-4 text-destructive" /> 
    Sentiment drops below
  </Label>
  <span className="font-mono">30%</span>
  <Slider defaultValue={[30]} max={100} step={1} />
</div>

Use case: Alert when positive sentiment falls below 30% (indicating majority negative/neutral opinion).

Recommended thresholds:

Conservative: 40% (catches early warnings)
Balanced: 30% (standard crisis threshold)
Aggressive: 20% (only severe crises)

2. Volume Spike Detection

// Source: ScheduleAnalysisModal.tsx (lines 169-176)
<Label className="flex items-center gap-2">
  <Flame className="h-4 w-4 text-warning" /> 
  Volume spikes above (vs avg)
</Label>
<span className="font-mono">150%</span>
<Slider defaultValue={[150]} max={500} step={10} />

Use case: Alert when conversation volume exceeds 150% of the rolling average (viral trend detection). Example: If average hourly volume is 10K mentions, alert triggers at 15K+ mentions.

3. Emotion-Specific Triggers

// Source: ScheduleAnalysisModal.tsx (lines 178-187)
<Label className="flex items-center gap-2">
  <Angry className="h-4 w-4 text-destructive" /> 
  Anger emotion exceeds
</Label>
<span className="font-mono">40%</span>
<Slider defaultValue={[40]} max={100} step={1} />

Use case: Trigger crisis alert when anger surpasses 40% of all emotions. Why anger? Anger is the strongest predictor of:

Boycott movements
Viral backlash
Media amplification
Regulatory action

You can customize this for other emotions. For example, set Fear > 50% for security/privacy topics, or Disgust > 30% for ethical scandals.

4. Viral Momentum Detection

// Source: ScheduleAnalysisModal.tsx (lines 190-194)
<Checkbox id="c5" defaultChecked>
  <TrendingUp className="h-4 w-4" /> 
  Notify if viral momentum detected
</Checkbox>

How it works: Triggers when:

Volume increases 200%+ within 1 hour
Sentiment shifts 30+ points in 6 hours
Topic enters trending algorithms (X, YouTube)

Delivery Channels

Email Delivery

Configuration:

[email protected], [email protected], [email protected]

Email format:

Subject: [SENTi-radar Alert] AI Regulation — Crisis Detected
Body: HTML-formatted report with emotion charts, sentiment timeline, and AI recommendations
Attachments: Optional CSV export

Slack Integration

Configuration:

#sentiment-alerts

https://hooks.slack.com/services/T00000000/B00000000/XXXXXXXXXXXX

Slack message format:

🔴 **Crisis Alert: Global Food Prices**

Sentiment: **Negative** (-45)
Dominant Emotion: **Anger (48%)**
Volume: **156.8K** (+67%)

**Key Concerns**:
• Corporate greed (18.4K mentions)
• Can't afford groceries (14.2K mentions)
• Price controls demanded (9.8K mentions)

**Recommended Action**: Immediate response required — anger at corporate profiteering is escalating.

<https://dashboard.senti-radar.com/topics/food-crisis|View Full Report>

Webhook Integration

Configuration:

https://api.company.com/webhooks/sentiment-alert

Webhook payload (POST request):

{
  "event": "crisis_detected",
  "timestamp": "2026-03-12T14:30:00Z",
  "topic": {
    "id": "4",
    "title": "Global Food Prices",
    "hashtag": "#FoodCrisis",
    "sentiment": "negative",
    "sentiment_score": -45,
    "crisis_level": "high",
    "volatility": 91
  },
  "emotions": [
    { "emotion": "anger", "percentage": 48, "count": 75264 },
    { "emotion": "fear", "percentage": 22, "count": 34496 },
    { "emotion": "sadness", "percentage": 18, "count": 28224 }
  ],
  "volume": {
    "current": 156800,
    "change_pct": 67,
    "trend": "rising"
  },
  "top_phrases": [
    { "phrase": "corporate greed", "count": 18400 },
    { "phrase": "can't afford", "count": 14200 }
  ],
  "alert_triggers": [
    "sentiment_below_threshold",
    "anger_exceeds_40",
    "volume_spike_detected"
  ],
  "report_url": "https://dashboard.senti-radar.com/reports/4-2026-03-12",
  "ai_summary": "Public outrage at corporate profiteering and government inaction. Immediate crisis communication recommended."
}

Webhooks allow you to integrate with PagerDuty, Opsgenie, Jira, or custom incident management systems.

Testing Your Schedule

Before saving, always test your configuration:

// Source: ScheduleAnalysisModal.tsx (lines 48-53)
const handleTest = () => {
  toast({
    title: 'Testing Integration',
    description: 'A test ping has been sent to your configured endpoints.'
  });
};

What happens during a test:

System generates a sample analysis for the selected topic
Sends test messages to all configured channels:
- Email: [TEST] SENTi-radar Configuration Test
- Slack: 🧪 Test Alert — Configuration successful
- Webhook: Includes "test": true flag in payload
Shows success/failure toast notification

If the test fails, check:

Email addresses are valid
Slack webhook URL is correct and not expired
Webhook endpoint returns 200-299 status code
API keys (if required) are configured

Saved Schedules & History

View and manage your active schedules:

// Source: ScheduleAnalysisModal.tsx (lines 202-204)
<Button variant="outline" size="sm">
  <History className="h-4 w-4" /> View History
</Button>

Schedule History includes:

All past executions (timestamp, status)
Alert triggers (which conditions fired)
Delivery status (email sent, webhook 200 OK)
Generated reports (archived for 90 days)
Errors and retry attempts

Use Cases

Product Launch Monitoring

Configuration:

Topic: iPhone 18 Launch
Frequency: Every hour (for first 48 hours)
Alert Thresholds:
- Sentiment < 40%
- Anger > 35%
- Volume spike > 200%
Delivery: Slack #product-team + Email to PR lead
Report Types: Crisis Alerts Only + Trend Comparison

Why this works: Hourly monitoring catches real-time reactions. Crisis Alerts Only prevents noise while catching pricing backlash or feature complaints.

Brand Health Dashboard (Weekly)

Configuration:

Topic: YourBrand (your company name)
Frequency: Weekly (Sunday at 9 AM)
Alert Thresholds:
- Sentiment < 30%
- Disgust > 25% (ethical/quality issues)
Delivery: Email to C-suite + Slack #exec-updates
Report Types: Full Analysis + Trend Comparison

Why this works: Weekly cadence shows long-term trends without alert fatigue. Trend Comparison highlights month-over-month changes.

Crisis-Only Monitoring

Configuration:

Topic: Industry Regulation
Frequency: Every 6 hours
Alert Thresholds:
- Sentiment < 25%
- Anger > 50%
- Fear > 40%
- Volume spike > 150%
Delivery: PagerDuty webhook + Email
Report Types: Crisis Alerts Only

Why this works: Only sends alerts when multiple red flags align (anger + fear + volume spike). Prevents false positives.

Advanced: Combining Conditions

You can create sophisticated alert logic by enabling multiple thresholds: Example: High-Confidence Crisis Alert Trigger only when ALL of these are true:

✅ Sentiment < 30%
✅ Anger > 40%
✅ Volume spike > 150%
✅ Crisis level = “high”

Implementation (pseudo-code):

const shouldAlert = (
  sentiment < 30 &&
  anger > 40 &&
  volumeSpike > 150 &&
  crisisLevel === 'high'
);

This reduces false positives by requiring convergence of multiple signals.

Best Practices

Start Conservative

Begin with sensitive thresholds (e.g., sentiment < 40%, anger > 30%) to catch early warnings. Relax them if you get too many false positives.

Use Crisis Alerts Only for High-Frequency Schedules

If monitoring hourly, enable “Crisis Alerts Only” to avoid inbox/Slack spam. Save “Full Analysis” for daily/weekly reports.

Test Webhooks Thoroughly

Always click “Test Now” before saving. Verify your endpoint receives the payload and handles it correctly.

Archive Old Schedules

Deactivate schedules for completed campaigns or past events to keep your dashboard clean.

Troubleshooting

Alerts Not Firing

Possible causes:

Thresholds too strict: Lower sentiment threshold from 20% → 35%
Topic has low volume: Ensure topic is trending (> 1K mentions/hour)
Schedule paused: Check History to see if it’s active

Email Delivery Failures

Check:

Email addresses are valid (no typos)
Not caught in spam filter (whitelist [email protected])
SMTP quota not exceeded

Webhook Timeouts

Fix:

Ensure endpoint responds within 10 seconds
Return 200-299 status code (even if async processing)
Check firewall/CORS settings

Limitations

Real-time data requirements: Schedules rely on live data from YouTube, X, and Reddit. If API quotas are exceeded or tokens expire, reports may fall back to keyword analysis.Frequency limits: Hourly monitoring can hit rate limits on free API tiers. For sub-hourly monitoring, consider upgrading to paid API plans (Scrape.do, YouTube Data API).No retroactive analysis: Schedules only analyze data from the time they run. If a crisis occurred 3 hours ago but your schedule runs every 6 hours, you’ll miss the peak.

Crisis Detection - Understand what triggers scheduled alerts
Data Export - Export scheduled report data for long-term archival
AI Insights - AI-generated recommendations included in Full Analysis reports

Get Started

Setup & Configuration

Core Features

Data Sources

Guides

Scheduled Analysis & Monitoring

Overview

Getting Started

Alert Conditions & Thresholds

1. Sentiment Drop Threshold

2. Volume Spike Detection

3. Emotion-Specific Triggers

4. Viral Momentum Detection

Delivery Channels

Email Delivery

Slack Integration

Webhook Integration

Testing Your Schedule

Saved Schedules & History

Use Cases

Product Launch Monitoring

Brand Health Dashboard (Weekly)

Crisis-Only Monitoring

Advanced: Combining Conditions

Best Practices

Start Conservative

Use Crisis Alerts Only for High-Frequency Schedules

Test Webhooks Thoroughly

Archive Old Schedules

Troubleshooting

Alerts Not Firing

Email Delivery Failures

Webhook Timeouts

Limitations

Build docs developers (and LLMs) love

Get Started

Setup & Configuration

Core Features

Data Sources

Guides

​Overview

​Getting Started

​Schedule Configuration Modal

​Alert Conditions & Thresholds

​1. Sentiment Drop Threshold

​2. Volume Spike Detection

​3. Emotion-Specific Triggers

​4. Viral Momentum Detection

​Delivery Channels

​Email Delivery

​Slack Integration

​Webhook Integration

​Testing Your Schedule

​Saved Schedules & History

​Use Cases

​Product Launch Monitoring

​Brand Health Dashboard (Weekly)

​Crisis-Only Monitoring

​Advanced: Combining Conditions

​Best Practices

Start Conservative

Use Crisis Alerts Only for High-Frequency Schedules

Test Webhooks Thoroughly

Archive Old Schedules

​Troubleshooting

​Alerts Not Firing

​Email Delivery Failures

​Webhook Timeouts

​Limitations

​Related Features

Build docs developers (and LLMs) love

Overview

Getting Started

Schedule Configuration Modal

Alert Conditions & Thresholds

1. Sentiment Drop Threshold

2. Volume Spike Detection

3. Emotion-Specific Triggers

4. Viral Momentum Detection

Delivery Channels

Email Delivery

Slack Integration

Webhook Integration

Testing Your Schedule

Saved Schedules & History

Use Cases

Product Launch Monitoring

Brand Health Dashboard (Weekly)

Crisis-Only Monitoring

Advanced: Combining Conditions

Best Practices

Troubleshooting

Alerts Not Firing

Email Delivery Failures

Webhook Timeouts

Limitations

Related Features