Written By William Bowen
Last updated 15 days ago
Intro
The Sales Useful Information Agent passively analyzes conversation history to extract key details that can enhance future sales interactions, such as personal preferences, buying intent, or potential upsell opportunities.
It automatically updates the contact card in Clerk Chat, ensuring sales teams have rich, contextual insights at their fingertips.
Use cases
Personalized Upselling: The agent detects when a customer mentions specific preferences or plans and logs them for future upsell opportunities.
Example: A customer says they’re bringing their family to a football game, so the agent notes this for later offers on family ticket packages or merchandise.Identifying Buying Signals: It captures subtle cues that indicate purchasing intent, helping sales teams follow up strategically.
Example: A customer mentions they’re “thinking about upgrading” their software plan, so the agent logs this as a sales opportunity for proactive outreach.Cross-Selling Opportunities: The agent recognizes mentions of related needs and suggests complementary products.
Example: A customer asks about booking a hotel for a conference, so the agent logs this and recommends transportation services in a follow-up.Tracking Competitor Mentions: It flags when customers reference competitors, giving sales teams insight into their alternatives.
Example: A client mentions they’re also considering a rival’s service, so the agent records this for future competitive positioning.Capturing Key Decision Makers: The agent extracts information about stakeholders involved in the buying decision.
Example: A customer casually mentions their CFO needs to approve the deal, so the agent updates the contact card with this crucial detail.Recognizing Budget Discussions: It identifies when a customer shares budget constraints or willingness to spend.
Example: A client says they have a $10,000 budget for marketing software, so the agent logs this for customized pricing recommendations.Highlighting Urgency and Timelines: It captures time-sensitive needs to help sales teams prioritize leads effectively.
Example: A customer says they need a solution “by the end of the month,” so the agent records this and flags it as a high-priority follow-up.Recording Product Preferences: It notes specific product features or options a customer is interested in for targeted recommendations.
Example: A client repeatedly asks about AI-driven automation features, so the agent logs this to tailor future demos or pitches.Detecting Pain Points: The agent captures customer frustrations and challenges to refine sales approaches.
Example: A prospect complains about slow customer support with their current provider, so the agent logs this as a potential selling angle.Identifying Renewal & Expansion Opportunities: It tracks signals that a customer may be ready to renew or expand their usage.
Example: A client mentions their team is growing and may need more seats in the future, so the agent logs this for proactive upsell outreach.
Screenshot of agent

Updates contact card:

Template
Example{
"$schema": "https://web-api.clerk.chat/pipeline-schema",
"name": "Sales - useful infomation",
"nodes": [
{
"type": "tool_runner",
"name": "Update contact card",
"triggeredBy": [
"unreadMessage.fromAny"
],
"responseType": "json",
"nodeConfig": {
"type": "contact_tools",
"config": {},
"enabledTools": [
"set_attribute"
],
"tools": {
"set_attribute": {
"paramValues": {
"attributeName": {
"type": "fixed",
"value": "Useful information"
},
"attributeValue": {
"path": "useful_information",
"type": "pipeline-variable"
}
}
}
}
}
},
{
"type": "ai_bot",
"name": "Extract any useful info",
"triggeredBy": [
"unreadMessage.fromAny"
],
"responseType": "json",
"nodeConfig": {
"modelProvider": "openai",
"modelVersion": null,
"maxTokens": null,
"temperature": null,
"variables": {},
"prompt": "Read the conversation history. \n\nYour job is to read the conversation history, and extract any useful information about the client from the conversation.\n\nUseful is anything that a sales rep could use in a conversation with the client that would be helpful or impress them. \n\nSome examples: \n1. **Personalized Upselling**: The agent detects when a customer mentions specific preferences or plans and logs them for future upsell opportunities. \n *Example:* A customer says they’re bringing their family to a football game, so the agent notes this for later offers on family ticket packages or merchandise. \n\n2. **Identifying Buying Signals**: It captures subtle cues that indicate purchasing intent, helping sales teams follow up strategically. \n *Example:* A customer mentions they’re “thinking about upgrading” their software plan, so the agent logs this as a sales opportunity for proactive outreach. \n\n3. **Cross-Selling Opportunities**: The agent recognizes mentions of related needs and suggests complementary products. \n *Example:* A customer asks about booking a hotel for a conference, so the agent logs this and recommends transportation services in a follow-up. \n\n4. **Tracking Competitor Mentions**: It flags when customers reference competitors, giving sales teams insight into their alternatives. \n *Example:* A client mentions they’re also considering a rival’s service, so the agent records this for future competitive positioning. \n\n5. **Capturing Key Decision Makers**: The agent extracts information about stakeholders involved in the buying decision. \n *Example:* A customer casually mentions their CFO needs to approve the deal, so the agent updates the contact card with this crucial detail. \n\n6. **Recognizing Budget Discussions**: It identifies when a customer shares budget constraints or willingness to spend. \n *Example:* A client says they have a $10,000 budget for marketing software, so the agent logs this for customized pricing recommendations. \n\n7. **Highlighting Urgency and Timelines**: It captures time-sensitive needs to help sales teams prioritize leads effectively. \n *Example:* A customer says they need a solution “by the end of the month,” so the agent records this and flags it as a high-priority follow-up. \n\n8. **Recording Product Preferences**: It notes specific product features or options a customer is interested in for targeted recommendations. \n *Example:* A client repeatedly asks about AI-driven automation features, so the agent logs this to tailor future demos or pitches. \n\n9. **Detecting Pain Points**: The agent captures customer frustrations and challenges to refine sales approaches. \n *Example:* A prospect complains about slow customer support with their current provider, so the agent logs this as a potential selling angle. \n\n10. **Identifying Renewal & Expansion Opportunities**: It tracks signals that a customer may be ready to renew or expand their usage. \n *Example:* A client mentions their team is growing and may need more seats in the future, so the agent logs this for proactive upsell outreach.\n\nThis should be maximum 100 words.",
"promptSections": [],
"sectionTemplates": {},
"responseSchema": {
"type": "object",
"required": [
"useful_information"
],
"properties": {
"useful_information": {
"type": "string"
}
}
},
"opts": {
"sendStructuredConvo": true
}
}
},
{
"type": "trigger",
"name": "Start",
"triggeredBy": [
"userMessage"
],
"responseType": "json"
}
],
"edges": [
{
"name": null,
"filters": [],
"sourceNode": "Start",
"destinationNode": "Extract any useful info"
},
{
"name": null,
"filters": [],
"sourceNode": "Extract any useful info",
"destinationNode": "Update contact card"
}
]
}