# Intent Journeys

Visualize how users progress from curiosity to conversion.

## What Are Intent Journeys?

An **intent journey** is the complete conversational path a user takes from their first query to booking or conversion. Unlike traditional analytics that only show "User clicked → User converted", intent journeys reveal the entire story.

### Traditional Analytics View

```
User clicked your link → User booked
```

**What you see**: 1 click, 1 conversion\
**What you miss**: Everything that happened before the click

### Intent Journey View

```
Turn 1: "Coffee shops in Portland" (Low intent)
Turn 2: "Coffee shops with outdoor seating" (Medium intent)
Turn 3: "Artisan Coffee Roasters hours" (High intent)
Turn 4: "Book table at Artisan Coffee" (Conversion)
```

**What you see**: Complete journey with intent progression\
**What you gain**: Understanding of how users make decisions

## Why This Matters

### The AI Conversation Difference

Microsoft research shows AI search sessions have **22% more chat turns** than traditional search. Users don't just search once and click - they have conversations that evolve.

**Key findings**:

* Average AI search: 4.2 turns before conversion
* Average traditional search: 1.3 clicks before conversion
* AI users arrive **3x more qualified**
* Journey length predicts conversion likelihood

### Real Business Impact

Businesses using intent journey analytics see:

* **40% better lead qualification** (focus on high-intent users)
* **25% shorter sales cycles** (understand where users are in journey)
* **2x better conversion rates** (optimize for journey stages)
* **Predictive insights** (know who's likely to convert)

## Understanding Intent Progression

### Intent Strength Levels

**Low Intent (0-39 points)**

* Early exploration phase
* General questions
* No urgency signals
* Example: "What coffee shops are in Portland?"

**Medium Intent (40-69 points)**

* Active research phase
* Comparing options
* Asking specific questions
* Example: "Compare Artisan Coffee and Stumptown"

**High Intent (70-100 points)**

* Ready to take action
* Asking about availability, pricing, booking
* Urgency signals present
* Example: "Book table at Artisan Coffee tonight"

### How Intent is Scored

Our AI analyzes multiple signals to calculate intent scores:

**High-Intent Keywords** (25 points each):

* Action words: book, reserve, schedule, buy, order
* Urgency: now, today, tonight, urgent, need
* Contact: call, visit, hire, contact

**Medium-Intent Keywords** (15 points each):

* Research: price, cost, how much, available
* Comparison: compare, best, recommend, versus
* Validation: review, rating, testimonial

**Contextual Factors**:

* Previous queries in session: +5 points each
* Session duration > 5 minutes: +10 points
* Citations received: +8 points each
* Comparisons appeared in: +12 points each
* Location indicators ("near me"): +15 points

**Example Calculation**:

```
Query: "book table at italian restaurant tonight"

Keywords:
- "book" = 25 points (high-intent action)
- "tonight" = 25 points (urgency)

Context:
- 2 previous queries = 10 points
- 1 citation = 8 points
- Session duration 7 min = 10 points

Total: 78 points = HIGH INTENT
```

## Journey Visualization

### Timeline View

Your dashboard shows every turn in the conversation:

```
Journey #12345 - Session started Dec 7, 2025 at 2:15 PM

Turn 1 (2:15 PM) - Query
"Coffee shops in Portland"
Intent: LOW (25 points)
→ Your business: Impression

Turn 2 (2:17 PM) - Refinement
"Coffee shops with outdoor seating and wifi"
Intent: MEDIUM (45 points)
→ Your business: Citation (secondary)

Turn 3 (2:19 PM) - Refinement
"Artisan Coffee Roasters menu and prices"
Intent: MEDIUM (62 points)
→ Your business: Citation (primary)

Turn 4 (2:21 PM) - Refinement
"Book table at Artisan Coffee for 4 people today"
Intent: HIGH (88 points)
→ Your business: CONVERSION ✓

Journey Metrics:
- Total turns: 4
- Duration: 6 minutes
- Intent evolution: Increasing
- Peak intent: 88
- Converted: Yes
```

### Intent Progression Chart

Visual representation of how intent evolved:

**Pattern**: Steadily Increasing (High conversion probability: 80%)

## Journey Patterns

### Pattern Types

**1. Steadily Increasing** (Best for conversion)

* Intent grows with each turn
* User getting more specific
* Conversion probability: **80%**

**2. Spike Pattern** (Sudden decision)

* Low intent, then sudden jump
* User found what they needed
* Conversion probability: **60%**

**3. Volatile Pattern** (Exploring)

* Intent goes up and down
* User comparing many options
* Conversion probability: **30%**

**4. Decreasing Pattern** (Lost interest)

* Intent drops over time
* User not finding what they need
* Conversion probability: **10%**

## Query Refinement Analysis

### Refinement Types

**Specification** - Adding more details

* "restaurants" → "italian restaurants"
* "coffee" → "coffee with outdoor seating"
* Indicates: User knows what they want

**Clarification** - Seeking understanding

* "Artisan Coffee" → "Artisan Coffee hours"
* "menu" → "vegetarian menu options"
* Indicates: User gathering information

**Comparison** - Evaluating options

* "Artisan Coffee" → "Artisan Coffee vs Stumptown"
* "best coffee" → "compare coffee shops"
* Indicates: User in decision phase

**Location** - Adding location context

* "coffee" → "coffee near me"
* "restaurants" → "restaurants in downtown Portland"
* Indicates: User ready to visit

**Timing** - Adding time constraints

* "restaurants" → "restaurants open now"
* "book table" → "book table tonight"
* Indicates: High urgency, ready to act

### Refinement Patterns

**Narrowing** (60% conversion rate)

* User getting more specific with each query
* Adding constraints and details
* Strong buying signal

Example:

```
"restaurants"
→ "italian restaurants"
→ "italian restaurants with outdoor seating"
→ "italian restaurants open now"
```

**Deciding** (80% conversion rate)

* Moving from research to action
* Intent score jumps significantly
* Strongest conversion signal

Example:

```
"coffee shops"
→ "Artisan Coffee reviews"
→ "book table at Artisan Coffee"
```

**Broadening** (20% conversion rate)

* User expanding search
* May not find what they need
* Lower conversion probability

Example:

```
"Artisan Coffee"
→ "coffee shops in Portland"
→ "cafes and restaurants"
```

**Exploring** (30% conversion rate)

* General exploration
* No clear direction
* Moderate conversion probability

Example:

```
"coffee"
→ "best coffee"
→ "coffee shops with wifi"
→ "coffee or tea"
```

## Your Dashboard

### Overview Tab

**Key Metrics**:

```
Total Journeys:        1,247
Converted Journeys:    456 (36.6%)
Avg Intent Score:      52.3
Avg Journey Length:    4.2 turns
Avg Duration:          6.8 minutes

Intent Evolution:
- Increasing:  67% (835 journeys)
- Stable:      18% (224 journeys)
- Volatile:    10% (125 journeys)
- Decreasing:   5% (63 journeys)

Intent Distribution:
- High Intent:    234 journeys (18.8%)
- Medium Intent:  678 journeys (54.4%)
- Low Intent:     335 journeys (26.8%)
```

### Individual Journey View

Click any journey to see:

* Complete timeline with all turns
* Intent score at each turn
* Your touchpoints (impressions, citations)
* Refinement pattern analysis
* Conversion outcome
* Journey metrics

**Example**:

```
Journey #12345
Started: Dec 7, 2025 at 2:15 PM
Duration: 6 minutes
Turns: 4
Pattern: Steadily Increasing
Converted: Yes ✓

Timeline:
[Turn 1] Query: "coffee shops" (Intent: 25)
  → Impression in summary

[Turn 2] Refinement: "coffee with outdoor seating" (Intent: 45)
  → Citation (secondary placement)

[Turn 3] Refinement: "Artisan Coffee menu" (Intent: 62)
  → Citation (primary placement)

[Turn 4] Refinement: "book table at Artisan Coffee" (Intent: 88)
  → CONVERSION: Booking created

Touchpoint Impact:
- Impression: +5 intent points
- Secondary citation: +8 intent points
- Primary citation: +15 intent points
```

### Pattern Analysis Tab

See aggregated patterns across all journeys:

```
Refinement Patterns (Last 30 days)

Narrowing:    456 journeys (36.6%)
  Avg conversion rate: 60%
  Avg turns: 3.8
  Most common: Specification → Location → Timing

Deciding:     234 journeys (18.8%)
  Avg conversion rate: 80%
  Avg turns: 2.9
  Most common: Research → Action

Exploring:    389 journeys (31.2%)
  Avg conversion rate: 30%
  Avg turns: 5.2
  Most common: Comparison → Clarification

Broadening:   168 journeys (13.5%)
  Avg conversion rate: 20%
  Avg turns: 4.5
  Most common: Specification → Broadening
```

## How to Improve Journey Outcomes

### Increase Intent Progression

✅ **Do**:

* Provide clear, specific information
* Answer common questions proactively
* Make booking/contact easy
* Show availability and pricing upfront
* Respond quickly to inquiries

❌ **Don't**:

* Hide important information
* Use vague descriptions
* Make users work to find details
* Ignore common questions
* Delay responses

### Optimize for High-Intent Users

**Identify high-intent signals**:

* Queries with action words (book, reserve, call)
* Timing indicators (today, tonight, now)
* Specific questions (hours, prices, availability)

**Respond appropriately**:

* Prioritize high-intent leads
* Offer immediate booking options
* Provide direct contact methods
* Show real-time availability
* Send follow-up messages

### Reduce Journey Friction

**Common friction points**:

* Missing information (hours, prices, location)
* Unclear booking process
* No availability shown
* Slow response times
* Complicated requirements

**Solutions**:

* Complete your profile 100%
* Enable instant booking
* Show real-time availability
* Set up auto-responses
* Simplify requirements

## Advanced Insights

### Conversion Prediction

Based on journey patterns, we can predict conversion likelihood:

**High Probability (70-100%)**:

* Steadily increasing intent
* 3-5 turns
* High-intent keywords present
* Multiple citations received
* Narrowing or deciding pattern

**Medium Probability (30-69%)**:

* Stable or slightly increasing intent
* 2-4 turns
* Medium-intent keywords
* At least one citation
* Exploring pattern

**Low Probability (0-29%)**:

* Decreasing or volatile intent
* 1-2 or 6+ turns
* Low-intent keywords only
* No citations
* Broadening pattern

### Journey Efficiency

**Efficient journeys** (3-4 turns, high conversion):

* User finds what they need quickly
* Clear intent progression
* Your content answers their questions
* Smooth path to conversion

**Inefficient journeys** (6+ turns, low conversion):

* User struggling to find information
* Intent not progressing
* Missing key information
* Friction in conversion path

**Optimize for efficiency**:

* Provide comprehensive information
* Answer questions proactively
* Make booking frictionless
* Show availability clearly
* Enable instant actions

### Touchpoint Impact

Track how each touchpoint affects intent:

```
Touchpoint Impact Analysis

Impression in summary:
  Avg intent increase: +5 points
  Conversion lift: 1.2x

Secondary citation:
  Avg intent increase: +8 points
  Conversion lift: 1.5x

Primary citation:
  Avg intent increase: +15 points
  Conversion lift: 2.3x

Comparison appearance:
  Avg intent increase: +12 points
  Conversion lift: 1.8x

Click to profile:
  Avg intent increase: +20 points
  Conversion lift: 3.1x
```

**Insight**: Primary citations have the biggest impact on intent progression. Focus on earning primary placements.

## API Access

Access intent journey data programmatically:

```bash
GET /api/v1/business/analytics/intent-progression
```

**Parameters**:

```json
{
  "businessId": "business-123",
  "startDate": "2025-11-01",
  "endDate": "2025-11-30",
  "journeyId": "journey-456", // Optional: specific journey
  "includePatterns": true, // Include refinement patterns
  "includeScoring": true // Include intent scoring examples
}
```

**Response**:

```json
{
  "success": true,
  "data": {
    "businessId": "business-123",
    "period": {
      "startDate": "2025-11-01",
      "endDate": "2025-11-30"
    },
    "summary": {
      "totalJourneys": 1247,
      "convertedJourneys": 456,
      "conversionRate": 36.6,
      "avgIntentScore": 52.3,
      "avgJourneyLength": 4.2,
      "avgTurns": 4.2,
      "evolutionDistribution": {
        "increasing": 835,
        "decreasing": 63,
        "stable": 224,
        "volatile": 125
      },
      "strengthDistribution": {
        "high": 234,
        "medium": 678,
        "low": 335
      }
    },
    "journeys": [
      {
        "journeyId": "journey-456",
        "sessionId": "session-789",
        "timeline": [
          {
            "timestamp": "2025-12-07T14:15:00Z",
            "turnNumber": 1,
            "event": "query",
            "content": "coffee shops in Portland",
            "intentStrength": "low",
            "context": "Initial exploration"
          }
        ],
        "intentProgression": [
          {
            "turnNumber": 1,
            "intentStrength": "low",
            "intentScore": 25,
            "confidence": 0.85,
            "signals": ["general query", "no urgency"]
          }
        ],
        "metrics": {
          "totalTurns": 4,
          "journeyLength": 360,
          "intentEvolution": "increasing",
          "peakIntentScore": 88,
          "avgIntentScore": 55,
          "conversionLikelihood": 85
        }
      }
    ]
  }
}
```

[View API reference →](https://amistan.gitbook.io/aidp-docs/for-developers/api-reference/analytics)

## Real-World Examples

### Example 1: Perfect Journey

```
Journey #45678
Pattern: Steadily Increasing → Converted ✓

T1: "restaurants in Seattle" (Intent: 20)
  → Impression

T2: "italian restaurants in Seattle" (Intent: 35)
  → Citation (secondary)

T3: "The Walrus and the Carpenter reservations" (Intent: 72)
  → Citation (primary)

T4: "book table for 2 tonight" (Intent: 92)
  → CONVERSION

Why it worked:
✓ Clear intent progression
✓ User found specific business
✓ Information readily available
✓ Easy booking process
```

### Example 2: Lost Opportunity

```
Journey #45679
Pattern: Volatile → Not Converted ✗

T1: "seafood restaurants" (Intent: 30)
  → Impression

T2: "best seafood in Seattle" (Intent: 45)
  → Citation (tertiary)

T3: "seafood restaurant prices" (Intent: 55)
  → No mention (competitor cited)

T4: "cheap seafood" (Intent: 35)
  → No mention

T5: "food delivery" (Intent: 20)
  → Lost to competitor

Why it failed:
✗ Pricing information not clear
✗ Lost to competitor in T3
✗ User broadened search
✗ Intent decreased over time
```

### Example 3: Quick Converter

```
Journey #45680
Pattern: Spike → Converted ✓

T1: "coffee near me" (Intent: 25)
  → Impression

T2: "book table at Artisan Coffee now" (Intent: 95)
  → CONVERSION

Why it worked:
✓ User already knew the business
✓ High urgency signal
✓ Instant booking available
✓ Minimal friction
```

## FAQs

**Q: How accurate is intent scoring?**\
A: Our algorithm has 85% accuracy in predicting conversions based on intent patterns. It improves over time as we collect more data.

**Q: Can I see journeys that didn't convert?**\
A: Yes! Understanding why users don't convert is just as valuable. Look for patterns in non-converting journeys.

**Q: How long are typical journeys?**\
A: Average is 4.2 turns over 6-8 minutes. But this varies by industry and intent level.

**Q: What's a good conversion rate?**\
A: Industry average is 30-40% for AI search journeys. Higher than traditional search (5-10%) because users arrive more qualified.

**Q: Can I export journey data?**\
A: Yes! Use the API or export from your dashboard (CSV, JSON).

**Q: How often is data updated?**\
A: Real-time. Journeys are tracked as they happen.

**Q: What if a user has multiple sessions?**\
A: Each session is tracked separately. We can link sessions by user ID if they're logged in.

## Next Steps

* [Learn About Attribution](https://amistan.gitbook.io/aidp-docs/for-business-owners/analytics/attribution)
* [View Upstream Metrics](https://amistan.gitbook.io/aidp-docs/for-business-owners/analytics/upstream-metrics)
* [See Competitive Benchmarks](https://amistan.gitbook.io/aidp-docs/for-business-owners/analytics/benchmarking)
* [Optimize Your Content](https://amistan.gitbook.io/aidp-docs/for-business-owners/profile-setup/exclusive-content)

***

**Questions?** Contact <support@aidp.dev> or join our [GitHub Discussions](https://github.com/aidp/platform/discussions)