Visitor value scoring uses AI to predict how valuable each visitor is to your business. Identify high-value prospects in real-time.
What Is Visitor Scoring?
AI assigns a score to each visitor:
| Score Range | Label | Meaning |
|---|---|---|
| 0-20 | Low | Casual browser |
| 21-40 | Medium-Low | Some interest |
| 41-60 | Medium | Engaged visitor |
| 61-80 | High | Strong interest |
| 81-100 | Very High | Likely to convert |
How Scoring Works
Factors Considered
AI analyzes multiple signals:
| Factor | Weight | Example |
|---|---|---|
| Pages viewed | High | 5+ pages |
| Time on site | High | 3+ minutes |
| Return visits | High | Came back 3x |
| Engagement depth | Medium | Scrolled, clicked |
| Content type | Medium | Pricing page |
| Traffic source | Medium | Organic search |
| Goal proximity | High | Started checkout |
Real-Time Updates
Score updates as visitor browses:
Initial: 15 (new visitor)
→ Viewed 3 pages: 28
→ Viewed pricing: 45
→ Started signup: 72
→ Completed signup: 95
Machine Learning
AI learns from your data:
- What behaviors lead to conversion
- Which visitors actually buy
- Patterns unique to your business
Viewing Scores
On the Globe
Visitor colors indicate score:
- Blue: Low value
- Green: Low-medium
- Yellow: Medium
- Orange: High
- Red: Very high
In Session List
Sessions show score:
- Score badge on each session
- Sort by value
- Filter by range
Real-Time View
Live visitors show:
- Current score
- Score trend (rising/falling)
- Key behaviors
Score Details
Visitor Profile
Click a visitor to see:
| Element | Information |
|---|---|
| Current Score | Value out of 100 |
| Score History | How it changed |
| Key Factors | What increased/decreased |
| Prediction | Likelihood to convert |
Factor Breakdown
Score: 72 (High)
Positive factors:
+ Viewed pricing page (+20)
+ Returned 3 times (+15)
+ 4 minutes on site (+10)
+ From organic search (+8)
Negative factors:
- Mobile device (-5)
- High bounce pages (-3)
Confidence Level
AI indicates confidence:
- High: Many data points
- Medium: Some data
- Low: Limited information
Using Scores
Identify Hot Leads
High-value visitors are:
- Most likely to convert
- Worth immediate attention
- Good for personalization
Prioritize Follow-Up
For B2B:
- Focus on high-score companies
- Prioritize sales outreach
- Time follow-ups right
Segment Analysis
Compare segments by score:
- Which sources bring high-value?
- What content attracts them?
- Where do they drop off?
Real-Time Action
Trigger actions on score:
- Chat popup for high value
- Special offers
- Priority support routing
Score-Based Filtering
Filter Sessions
Find specific value ranges:
- Go to Sessions
- Set score filter (e.g., 70-100)
- View high-value sessions
- Analyze behaviors
Filter in Reports
View metrics by value:
- Traffic breakdown by score
- Conversion by score tier
- Revenue by visitor value
Create Segments
Scale PlanSave score-based segments:
- High-value visitors
- Rising stars (increasing score)
- Engaged but not converting
Customizing Scoring
Scale PlanAdjust Factor Weights
Customize what matters:
| Factor | Default | Your Setting |
|---|---|---|
| Pricing page | +20 | +30 |
| Blog visit | +5 | +2 |
| Return visit | +15 | +20 |
| Demo request | +25 | +40 |
Add Custom Factors
Include your signals:
- Specific page visits
- Custom events
- Form interactions
- Feature usage
Industry Templates
Pre-built for:
- E-commerce
- SaaS
- B2B
- Media
Score Triggers
Scale PlanAutomated Actions
Trigger actions on score changes:
When: Score reaches 70
Action: Send Slack notification
Channel: #sales-leads
Message: "High-value visitor on {page}"
Integration Examples
| Score Event | Action |
|---|---|
| Reaches 60 | Add to email sequence |
| Reaches 80 | Notify sales rep |
| Drops below 30 | Exit campaign |
Chat Integration
Route by value:
- High value → Live agent
- Medium → Bot + escalation
- Low → Self-service
Score Analytics
Score Distribution
See overall distribution:
Low (0-40): 45%
Medium (41-60): 35%
High (61-100): 20%
Score by Source
Compare traffic quality:
| Source | Avg Score | High-Value % |
|---|---|---|
| Organic | 52 | 28% |
| Direct | 48 | 22% |
| Referral | 61 | 35% |
| Paid | 44 | 18% |
| Social | 35 | 12% |
Conversion by Score
Validate scoring accuracy:
| Score Range | Conversion Rate |
|---|---|
| 0-20 | 0.5% |
| 21-40 | 1.2% |
| 41-60 | 3.8% |
| 61-80 | 9.5% |
| 81-100 | 22% |
Improving Score Accuracy
Train the Model
Help AI learn:
- Mark actual conversions
- Indicate false positives
- Provide conversion values
Regular Review
Periodically check:
- Are high scores converting?
- Are low scores being missed?
- What behaviors matter?
Feedback Loop
AI improves when:
- Conversions are tracked
- Values are accurate
- Behaviors are measured
Score Privacy
What's Stored
Score data includes:
- Aggregated behavior
- Derived score
- Factor breakdown
What's Not Stored
Privacy protected:
- Personal identity (unless provided)
- Sensitive behaviors
- Off-site activity
Visitor Control
Visitors can:
- Not be scored (privacy mode)
- Request data deletion
- Opt out of tracking
Best Practices
Act on Scores
Don't just observe:
- Set up notifications
- Create workflows
- Personalize experience
Validate Regularly
Check accuracy:
- Compare scores to conversions
- Adjust weights as needed
- Review false positives
Segment Strategically
Use scores to:
- Prioritize support
- Target advertising
- Customize content
Troubleshooting
Scores Seem Wrong
If scores don't match expectations:
- Check tracked behaviors
- Review factor weights
- Validate conversions tracked
All Scores Similar
If no differentiation:
- Need more data
- Add more factors
- Wait for learning period
Scores Not Updating
If scores stale:
- Verify tracking active
- Check real-time data
- Confirm session continuation