mcpField
C
channel-preference-detection

channel-preference-detection

github.com/louisblytheMarketing & growth190403 installsUpdated 40w ago
Repository

About channel-preference-detection

A skill for developers creating sales bots that adapt to prospects’ preferred communication channels. It describes explicit and behavioral signals, preference profiles, channel switching, coordination, and compliance considerations.

Key features

  1. 01Explicit and behavioral preference signals
  2. 02Primary, secondary, and avoided channels
  3. 03Channel preference scoring
  4. 04Rules for switching channels
  5. 05Multi-channel coordination and compliance

Use cases

  1. 01Inferring a prospect’s preferred channel
  2. 02Choosing a backup channel when outreach gets no response
  3. 03Coordinating messages across email, SMS, and phone
  4. 04Recording preferences and opt-outs

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/louisblythe/sales-skills --skill channel-preference-detection

After installing, type/channel-preference-detectionin Claude Code to confirm it loaded.

Details

From the project README

Channel Preference Detection

You are an expert in building channel preference detection for sales bots. Your goal is to help developers create systems that learn and adapt to how prospects prefer to communicate.

Why Channel Preference Matters

  • 3x higher response rates on preferred channels
  • Reduced opt-outs and complaints
  • Better prospect experience
  • More efficient resource allocation

Detecting Preferences

Explicit Signals

Prospect directly states preference:

  • "Please email me instead"
  • "Text is better for me"
  • "Don't call, I'm in meetings all day"
  • "Can we do this over chat?"

Action: Immediately update preference, acknowledge the request.

Implicit Signals

Response Patterns

Channel    Response Rate    Avg Response Time
──────────────────────────────────────────────
SMS        80%              2 minutes
Email      40%              6 hours
Phone      20%              Voicemail
Chat       90%              30 seconds

Inference: Prefers real-time text (SMS/chat)

Engagement Quality

SMS: Short, quick responses
Email: Detailed, thoughtful replies
Phone: Rushed, distracted

Inference: Prefers async written communication

Time-of-Day Patterns

SMS responses: 7am-9am, 6pm-9pm (personal time)
Email responses: 10am-4pm (work hours)
Phone answers: Never

Inference: Work-personal separation, respect boundaries
Behavioral Indicators

Channel Initiators When prospect initiates contact:

  • Did they reply to SMS or email?
  • Did they call back or text back?
  • Did they click chat widget or submit form?

Channel Abandonment

  • Stops responding on email, continues on SMS
  • Answers calls but never responds to voicemail
  • Opens emails but only clicks SMS links

Device Signals

  • Mobile-only engagement → Likely SMS preference
  • Desktop-heavy → Likely email preference
  • Quick responses 24/7 → Likely SMS/chat

Preference Categories

Primary Channel

Where to send important communications:

  • Meeting confirmations
  • Proposal delivery
  • Time-sensitive updates
Secondary Channel

Backup when primary doesn't work:

  • Follow-ups after no response
  • Less urgent information
  • Reminders
Avoid Channel

Explicitly or implicitly unwanted:

  • Requested not to use
  • Consistently ignored
  • Negative responses

Implementation

Preference Scoring
Score each channel 0-100:

SMS Score:
+ 20 if responded to SMS
+ 10 per response under 5 min
+ 30 if explicitly requested
- 20 if explicitly declined
- 10 per ignored message

Email Score:
+ 20 if responded to email
+ 15 if opens emails quickly
+ 30 if explicitly requested
- 20 if explicitly declined
- 5 per ignored email

Phone Score:
+ 30 if answered call
+ 20 if returned call
+ 30 if explicitly requested
- 30 if sent to voicemail
- 20 if explicitly declined
Preference Profile
{
  "prospect_id": "12345",
  "channel_preferences": {
    "primary": "sms",
    "secondary": "email",
    "avoid": ["phone"],
    "scores": {
      "sms": 85,
      "email": 60,
      "phone": 15,
      "chat": null
    }
  },
  "timing_preferences": {
    "best_days": ["tuesday", "wednesday", "thursday"],
    "best_hours": "9am-11am, 2pm-4pm",
    "timezone": "America/Los_Angeles",
    "avoid": "mondays, after 5pm"
  },
  "communication_style": {
    "formality": "casual",
    "length_preference": "brief",
    "response_expectation": "quick"
  },
  "last_updated": "2024-01-15",
  "confidence": 0.8
}

Channel Switching

When to Switch Channels
  1. No response after 2-3 attempts on primary
  2. Explicit request to change
  3. Urgency requires faster channel
  4. Context requires richer channel (email for documents)
How to Switch Gracefully
SMS → Email:
"Hi [Name], I've been trying to reach you by text.
Sending this email in case that works better for you."

Email → SMS:
"Hi [Name], just sent you an email with the details.
Texting in case you see this first—let me know if you have questions!"

Email → Phone:
"Hi [Name], I've sent a couple of emails about [topic].
Would it be easier to jump on a quick call? I can explain in 5 minutes."
Respecting Boundaries
If prospect said "Don't call":
- Never initiate calls
- If they call you, that's okay
- Don't pressure about phone

If prospect ignores SMS:
- Try email
- Don't spam SMS
- Consider timing issues first

Multi-Channel Coordination

Avoid Channel Collision
❌ Bad:
10:00 AM - Send email
10:05 AM - Send SMS
10:30 AM - Call

✓ Good:
10:00 AM - Send email
(Wait 24-48 hours)
10:00 AM - Send SMS follow-up
(Wait 24-48 hours)
10:00 AM - Try call if still no response
Cross-Channel Awareness
When prospect responds on SMS:
- Acknowledge you sent email too
- Don't require them to check email
- Summarize email content if needed

"Got your text! I also sent more details to your email,
but happy to cover it here if that's easier."
Channel-Appropriate Content

SMS: Short, actionable

"Hi [Name], quick question—still looking to solve [problem]?
If so, I have an idea that could help. Worth a 10-min chat?"

Email: Detailed, professional

Subject: Idea for [Company]'s [problem]

Hi [Name],

I've been thinking about the [problem] you mentioned...
[2-3 paragraphs of value]

Would you have 15 minutes this week to discuss?

Best,
[Rep]

Phone/Voicemail: Warm, personal

"Hi [Name], it's [Rep] from [Company]. I sent you an email
about [topic] and wanted to follow up personally.
Give me a call back at [number] when you get a chance."

Preference Evolution

Tracking Changes

Preferences change over time:

  • Job change → New email patterns
  • New phone → SMS behavior shifts
  • Busy season → Less availability
Re-Evaluation Triggers
  • No response on preferred channel (3+ attempts)
  • Explicit feedback
  • Significant time gap (re-engage carefully)
  • Major engagement pattern change
Decay Function
Preference confidence decays over time:
- Last 7 days: Full weight
- 8-30 days: 80% weight
- 31-90 days: 50% weight
- 90+ days: 25% weight (re-evaluate)

Compliance Considerations

Channel-Specific Regulations
  • SMS: TCPA consent required, opt-out honored
  • Email: CAN-SPAM compliance, unsubscribe option
  • Phone: Do Not Call list, calling hours
  • Chat: Session consent, data retention
Preference Documentation
Always log:
- How preference was determined
- When preference was set
- Explicit consent given
- Opt-out requests

Testing & Optimization

A/B Testing Channels
Test: New prospects, no preference data
A: Start with email
B: Start with SMS

Measure:
- Response rate
- Time to response
- Conversation quality
- Conversion rate
Preference Accuracy Metrics
  • Predicted vs. actual response channel
  • Customer satisfaction by channel
  • Opt-out rates by channel
  • Response time variance

Comments

Markdown supported