AI Overview
Thread uses OpenAI’s GPT models to generate personalized content for emails, handoff Q&A, forms, and more. AI drafts save CSMs hours while maintaining quality and personalization. [SCREENSHOT: AI-generated email draft in review interface] Caption: AI generates drafts based on customer context and template guidanceWhat AI Generates
Action Drafts
Email Content
Subject lines and email bodies with customer-specific details
Form Intro Text
Instructions and context for form collection
Slack Messages
Channel updates and notifications
Handoff Q&A
Answers to handoff questions from CRM data
What AI Doesn’t Generate
Form Fields
Pre-configured in templates by admins
Document Types
Pre-defined in action configuration
Meeting Agendas
Template-based, not AI-generated (yet)
Portal UI
Configured through Thread’s interfaces
How AI Generation Works
The AI Pipeline
Context Gathering
Thread collects all relevant context:
- Account data (company, contacts, ARR, close date)
- Handoff Q&A
- Template configuration
- Previous actions and responses
- Milestone information
Prompt Construction
Thread builds a structured prompt with:
- System instructions (tone, format, constraints)
- Customer context
- Action purpose and template
- Variable placeholders
- Output format requirements
OpenAI API Call
Thread sends prompt to OpenAI GPT-4
- Uses structured output mode
- Enforces JSON schema
- Includes retry logic for failures
Response Processing
Thread processes AI response:
- Validates format
- Interpolates variables
- Applies final formatting
- Stores as draft
AI Context Sources
Account Data
- Company name for personalization
- Industry for relevant examples
- Contact name and role for appropriate tone
- ARR for understanding customer tier
- Notes for specific context
Handoff Intelligence
- Why customer bought (use case)
- What success looks like (goals)
- Potential challenges (risks)
- Who’s involved (stakeholders)
- Technical context (environment)
Template Configuration
- Purpose of this action
- Stage in customer journey
- Expected tone and format
- Required elements to include
Previous Actions
- Established relationship context
- What’s been discussed already
- Customer engagement level
- Appropriate follow-up tone
Prompt Engineering
System Prompt
Thread uses a carefully crafted system prompt:User Prompt (Context)
Then adds customer-specific context:Output Format
AI responds with structured JSON:Regeneration
When to Regenerate
Draft Misses the Mark
Draft Misses the Mark
AI generated something generic or off-topicSolution: Regenerate with specific instructions
Wrong Tone
Wrong Tone
Too formal, too casual, or inconsistent with brandSolution: Regenerate with tone guidance
Missing Key Details
Missing Key Details
Didn’t mention important contextSolution: Regenerate with “Include [specific detail]”
Too Long/Short
Too Long/Short
Doesn’t match desired lengthSolution: Regenerate with length constraint
How to Regenerate
- Click “Regenerate” in action review
- Add instructions in text field:
- “Make it more casual”
- “Add urgency about Q3 deadline”
- “Focus on technical integration aspects”
- “Keep under 150 words”
- “Reference their Salesforce CRM specifically”
- Click “Regenerate”
- AI generates new draft with your guidance
- Review new version
AI Quality Factors
High-Quality Drafts
When you get great drafts:- ✅ Rich customer data in account
- ✅ Detailed handoff Q&A completed
- ✅ Clear action purpose in template
- ✅ Good content template with examples
- ✅ Relevant previous action context
- ✅ Specific, not generic account notes
Lower-Quality Drafts
When drafts are generic:- ❌ Minimal account data
- ❌ Skipped or sparse handoff
- ❌ Vague action descriptions
- ❌ Generic content templates
- ❌ No previous context
- ❌ Empty account notes
Improving AI Performance
1. Enrich Account Data
Before AI generates drafts:- During Account Creation
- During Handoff
- Ongoing
- Complete all optional fields
- Add detailed notes
- Include CRM data in raw payload
- Paste relevant conversations
2. Optimize Templates
Action descriptions:- ❌ “Send email”
- ✅ “Welcome email introducing CSM, setting expectations, and providing portal access”
- ❌ “Email customer about status”
- ✅ Structured template with sections, examples, and variables
- Add notes in template: “This email should reference the customer’s use case and address any risks mentioned in handoff”
3. Provide Feedback
Help improve AI over time:- Note patterns in edits (always adding industry examples?)
- Track which prompts work well
- Share feedback with team
- Request template improvements from admins
AI Limitations
What AI Can’t Do
Read Your Mind
AI doesn’t know unstated customer preferences or internal politics
Remember Calls
AI doesn’t have access to verbal conversations unless you document them
Predict Future
AI can’t anticipate customer reactions or upcoming issues
Replace Human Touch
AI drafts need your personalization and relationship context
When to Override AI
Always edit or rewrite when:- Customer has unique situation not captured in data
- Sensitive topic requiring human nuance
- Relationship dynamic AI doesn’t understand
- Brand voice needs adjustment
- Humor or empathy required
- Complex negotiation or escalation
Think of AI as your first draft writer, not your final copy editor. You’re the expert—AI just saves you time on the initial structure.
Best Practices
Always Review
Never execute AI drafts without reading them
Add Personal Touch
Include customer-specific details only you know
Use Regeneration
Don’t settle for mediocre—guide AI to better output
Enrich Context
Better input data = better AI output
Track Patterns
Note what works and share with team
Provide Feedback
Help improve prompts and templates over time
