Virtual Demonstration
Identifying Mentions of Product/Service Demonstrations
The "Virtual Demonstration" tracker detects phrases indicating that a product or service demo—often conducted via screen sharing, online platform, or live guided walk-through—is happening or planned. By capturing these references, teams gain insights into how often demos occur, which features are showcased, and how customers respond to visual presentations.
Who Benefits:
- Project Managers & Team Leads: Understand how frequently and effectively representatives use demos, informing training and workflow improvements.
- Sales & CX Teams: Identify when prospects are shown product capabilities in real-time, helping tailor follow-ups and assess demo outcomes.
- Business Managers & Strategists: Spot trends in demo usage to refine product presentation strategies, tools, and messaging.
- Compliance Officers: Ensure demos adhere to guidelines, that no non-compliant functionality is shown, and that promised features align with reality.
Value Proposition:
- Enhanced Engagement: Tracking demo occurrences helps gauge how customers interact with visual presentations, impacting conversions and satisfaction.
- Operational Efficiency: Structured JSON output integrates into dashboards, enabling analysis of demo frequency, effectiveness, and needed enhancements.
- Continuous Improvement: Insights from demos guide refinements to presentation materials, product prototypes, and representative training.
Data Dictionary
Objective:
Detect phrases in the conversation that indicate a virtual product/service demonstration—such as “let me show you,” “I’ll share my screen,” “I can walk you through the interface,” or “we’ll do a live demo.” The tracker returns occurrences of matched phrases, along with timestamps and confidence, enabling teams to understand how demos factor into their engagements.
Proposed JSON Schema:
{
"trackers": [
{
"name": "Virtual_Demonstration",
"matches": [
{
"text": "<string>",
"timestamp": "<ISO 8601 or relative time>",
"confidence": <number> // 0.0 to 1.0
}
]
}
]
}
Data Dictionary:
Field | Description | Example |
---|---|---|
trackers | Array of tracker objects. | [ {...} ] |
trackers[].name | The tracker’s name. | "Virtual_Demonstration" |
trackers[].matches | Array of matches for phrases indicating a virtual demo. | [ {...}, {...} ] |
matches[].text | The matched phrase representing a virtual demonstration reference. | "Let me share my screen and show you the interface." |
matches[].timestamp | When the phrase occurred (ISO 8601 or relative time). | "2024-07-10T10:00:00Z" or "00:02:30" |
matches[].confidence | Confidence (0.0 to 1.0) indicating how closely the phrase aligns with a virtual demo mention. | 0.9 |
If No Matches Found:
If no virtual demonstration references are detected, return an empty matches
array for "Virtual_Demonstration" or omit the tracker. If no trackers match at all, return:
{
"trackers": []
}
Determining Confidence & Thresholds
How the Virtual_Demonstration Tracker Works:
- Seed the tracker with phrases related to demos (e.g., "I'll share my screen," "let's do a live demo," "I can show you the product interface," "I'll walk you through a quick demonstration").
- The system generalizes these phrases to find semantically related statements in the transcript.
- Confidence indicates how closely the detected phrase aligns with conducting or planning a virtual demonstration.
Confidence & Calibration:
- Initial Testing: Validate on known calls to confirm that detected phrases truly represent a virtual demo scenario.
- Refine Phrases: Add or remove seed phrases as tools, platforms, and demonstration styles evolve.
- Thresholding: Downstream processes can filter out low-confidence matches if necessary.
- Stakeholder Feedback: Involve sales leads, PMs, or compliance teams to refine definitions and ensure relevant demos are captured.
Prompt Construction & Instructions
Role Specification & Reiteration:
- Present the system as a “highly experienced assistant” identifying virtual demo mentions.
- Reiterate instructions: return only JSON, follow schema, no extra commentary.
No Hallucination:
- Only detect virtual demos if the transcript supports it.
- If uncertain, assign a lower confidence or omit matches.
Strict Formatting:
- Return only JSON with
trackers
. - If none found, return empty
trackers
.
Example Prompt for Implementation
System Message (Role: System):
"You are a highly experienced assistant that identifies 'Virtual_Demonstration' mentions in a conversation. You have:
- A transcript
- A 'Virtual_Demonstration' tracker defined by seed phrases indicating product/service demos.
Your task:
- Detect phrases suggesting a live or planned virtual demo.
- Return
text
,timestamp
, andconfidence
. - Return only JSON as defined. No extra commentary.
Instructions & Variables (example):
{
"tracker": {
"name": "Virtual_Demonstration",
"phrases": [
"share my screen",
"walk you through the interface",
"live demo",
"let me show you how this works"
]
}
}
User Message (Role: User):
"Analyze the following transcript and detect 'Virtual_Demonstration' mentions:
[TRANSCRIPT_JSON][TRANSCRIPT_JSON]"
Expected Output Format:
{
"trackers": [
{
"name": "Virtual_Demonstration",
"matches": [
{
"text": "Let me share my screen so you can see how the feature works.",
"timestamp": "2024-07-10T10:00:00Z",
"confidence": 0.9
}
]
}
]
}
Updated about 2 months ago