The engagement score measures how actively your subscribers interact with your emails on a scale of 0-100. This score combines open rates and click-through rates to give you a comprehensive view of email performance.

## Score Calculation

Your engagement score is calculated from two equally weighted components:

### Open Rate Score (50 points maximum)

**Target**: 30% open rate = 50 points
**Formula**: Each 0.6% open rate = 1 point

The open rate measures the percentage of sent emails that were opened by recipients.

**Calculation**: `openRateScore = Math.min((openRate / 0.6), 50)`

**Example**:

- 30% open rate: 30 / 0.6 = **50 points** ✅
- 18% open rate: 18 / 0.6 = **30 points**
- 6% open rate: 6 / 0.6 = **10 points**

### Click Rate Score (50 points maximum)

**Target**: 10% click rate = 50 points
**Formula**: Each 0.2% click rate = 1 point

The click rate measures the percentage of sent emails that resulted in at least one click.

**Calculation**: `clickRateScore = Math.min((clickRate / 0.2), 50)`

**Example**:

- 10% click rate: 10 / 0.2 = **50 points** ✅
- 5% click rate: 5 / 0.2 = **25 points**
- 1% click rate: 1 / 0.2 = **5 points**

### Total Score

**Total Engagement Score = Open Rate Score + Click Rate Score**

**Example Scenarios**:

| Open Rate | Click Rate | Open Score | Click Score | Total Score | Rating       |
| --------- | ---------- | ---------- | ----------- | ----------- | ------------ |
| 30%       | 10%        | 50         | 50          | **100**     | Excellent 🟢 |
| 18%       | 5%         | 30         | 25          | **55**      | Good 🔵      |
| 15%       | 3%         | 25         | 15          | **40**      | Fair 🟡      |
| 6%        | 1%         | 10         | 5           | **15**      | Poor 🔴      |

## How We Track Opens

### Open Tracking Technology

Email opens are tracked using a **1×1 pixel tracking image** embedded in the HTML email. When a recipient opens the email and their email client loads images, this pixel sends a signal back to our servers.

**Technical details**:

- A unique tracking pixel URL is generated for each email
- The pixel is invisible to the recipient (1×1 transparent image)
- When loaded, it triggers a server request that records the open event
- We capture the timestamp, IP address, and user agent

### Unique Opens vs. Total Opens

**Lumail tracks unique opens only**, meaning:

- **First open**: Counted in metrics
- **Subsequent opens**: Marked as duplicate and not counted

This is stored in the `emailOpen` table with an `isDuplicate` flag:

```typescript
const uniqueOpensCount = await prisma.emailOpen.count({
  where: {
    email: { organizationId, createdAt: { gte, lte } },
    isDuplicate: false, // Only count unique opens
  },
});
```

**Why unique opens?**: This prevents inflating your metrics when someone opens an email multiple times.

### Open Tracking Limitations

**Important**: Open tracking is not 100% accurate due to:

1. **Image Blocking**: Some email clients (like Outlook) block images by default
2. **Privacy Features**: Apple Mail Privacy Protection pre-loads images on their servers, not user devices
3. **Plain Text Emails**: Opens cannot be tracked in plain text emails
4. **Preview Panes**: Some clients load images in preview panes, triggering false opens

**Typical accuracy**: ~70-80% of actual opens are tracked

### Apple Mail Privacy Protection Impact

Since iOS 15, Apple Mail pre-loads tracking pixels even if the user never opens the email. This can:

- Inflate open rates for Apple Mail users
- Make timing data unreliable
- Affect A/B testing based on send time

**Our approach**: We still track these opens but recommend focusing on click rates for Apple Mail-heavy lists.

## How We Track Clicks

### Click Tracking Technology

All links in your emails are automatically rewritten to route through our tracking servers.

**Process**:

1. Original link: `https://yoursite.com/blog/post`
2. Rewritten link: `https://lumail.io/c/[unique-id]`
3. User clicks → our server records the click → redirects to original URL

**What we capture**:

- Timestamp of the click
- IP address (for geographic data)
- User agent (device/browser information)
- The specific link clicked
- Which email the click came from

### Unique Clicks vs. Total Clicks

Like opens, **Lumail tracks unique clicks only**:

- **First click on any link**: Counted in metrics
- **Subsequent clicks**: Marked as duplicate and not counted

This is stored in the `emailClick` table:

```typescript
const uniqueClicksCount = await prisma.emailClick.count({
  where: {
    email: { organizationId, createdAt: { gte, lte } },
    isDuplicate: false, // Only count unique clicks
  },
});
```

**Note**: A unique click counts the first click per email, regardless of which link was clicked. Multiple clicks on different links in the same email are still marked as duplicates.

### Click Rate Calculation

Click rate is calculated as **clicks / sent**. Click-to-open (clicks / opens) is a separate metric on campaign reports.

```
clickRate = (uniqueClicks / totalSent) × 100
clickToOpenRate = (uniqueClicks / uniqueOpens) × 100
```

**Example**:

- Sent: 10,000 emails
- Opened: 2,500 (25% open rate)
- Clicked: 200
- **Click rate**: 200 / 10,000 = **2%**
- **Click-to-open**: 200 / 2,500 = **8%**

### Unsubscribe Link Tracking

Unsubscribe links are also click-tracked, but they:

- Don't count toward your engagement score
- Are automatically excluded from click rate calculations
- Are still logged for compliance and analytics

## Understanding the Metrics

### Sent

The total number of emails sent to recipients. This is the baseline for calculating open rate.

**Formula**: `openRate = (uniqueOpens / totalSent) × 100`

### Opened

The number of unique recipients who opened your email at least once.

**Best practice**: If your open rate is low (<15%), consider:

- Testing different subject lines
- Sending at different times of day
- Cleaning your list of inactive subscribers
- Improving sender name recognition

### Clicked

The number of unique recipients who clicked at least one link in your email.

**Best practice**: If your click rate is low (<2%), consider:

- Making CTAs more prominent
- Reducing the number of links (fewer options = more clicks)
- Improving email copy and relevance
- Using more compelling action-oriented language

## Score Interpretation

| Score Range | Status         | Action Required                                           |
| ----------- | -------------- | --------------------------------------------------------- |
| 80-100      | Excellent (🟢) | Your emails are highly engaging. Keep doing what works!   |
| 60-79       | Good (🔵)      | Solid performance. Test improvements to reach excellent.  |
| 40-59       | Fair (🟡)      | Room for improvement. Focus on subject lines and content. |
| 0-39        | Poor (🔴)      | Immediate attention needed. Review content strategy.      |

## Improving Your Score

### Boost Open Rates

1. **Craft compelling subject lines**
   - Use curiosity, urgency, or personalization
   - Keep it under 50 characters
   - A/B test different approaches

2. **Optimize sender name**
   - Use a recognizable person or brand name
   - Avoid "noreply@" addresses
   - Stay consistent

3. **Send at optimal times**
   - Test different days and times
   - Consider your audience's timezone
   - Avoid weekends for B2B

4. **Clean your list regularly**
   - Remove hard bounces immediately
   - Consider removing subscribers with no opens in 6+ months
   - Use re-engagement campaigns before removing

5. **Segment your audience**
   - Send more relevant content to smaller groups
   - Personalize based on interests or behavior
   - Avoid one-size-fits-all campaigns

### Boost Click Rates

1. **Clear call-to-action**
   - One primary CTA per email
   - Make buttons stand out
   - Use action-oriented language ("Get Started", not "Click Here")

2. **Relevant content**
   - Match email content to subscriber interests
   - Deliver on subject line promises
   - Provide immediate value

3. **Optimize email design**
   - Use short paragraphs
   - Include plenty of white space
   - Make links and buttons easy to tap on mobile

4. **Create urgency**
   - Limited-time offers
   - Exclusive content
   - Early-bird access

5. **Test everything**
   - A/B test CTA placement
   - Try different button colors and sizes
   - Test text links vs. buttons

## Industry Benchmarks

For context, here are typical industry benchmarks:

| Metric                | Excellent | Good   | Needs Work |
| --------------------- | --------- | ------ | ---------- |
| Open Rate             | >25%      | 15-25% | <15%    |
| Click Rate (of sent)  | >4%       | 2-4%   | <2%     |
| Engagement Score      | >80       | 60-80  | <60     |

**Note**: Benchmarks vary by industry, audience size, and email type (newsletter vs. promotional). Focus on improving your own metrics over time rather than comparing to others.

## Technical Details

The engagement score is calculated using this formula:

```typescript
function calculateEngagementScore(openRate: number, clickRate: number) {
  const openRateScore = Math.min(openRate / 0.6, 50);
  const clickRateScore = Math.min(clickRate / 0.2, 50);

  return Math.round(openRateScore + clickRateScore);
}
```

Where:

- `openRate` = (unique opens / total sent) × 100
- `clickRate` = (unique clicks / total sent) × 100

This balanced approach ensures both metrics contribute equally to your overall engagement health.

## Related Documentation

- [Deliverability Score](/docs/features/deliverability-score) - Monitor sender reputation
- [Content Deliverability Checker](/docs/features/content-deliverability) - Optimize email content before sending
- [Subscriber Events](/docs/features/subscriber-events) - View open and click events in timeline
- [Variables](/docs/features/variables) - Personalize emails to boost engagement