Custom-Development of the EnergyPal Inbound AI Triage Agents to capture, triage, qualify, route, and schedule inbound callers that are currently being missed across multiple parts of the EnergyPal customer journey.
Based on our Zoom call, EnergyPal is currently receiving several hundred inbound calls per month across different functions, and missing approximately 70–80% of them.
These missed calls include new leads, returning prospects, assigned opportunities, quote-stage leads, and post-sale support inquiries.
What We'll Do (Phase 1)
Build an AI inbound receptionist/triage agent for missed and overflow calls
Route callers through 4–5 sub-agent flows based on intent and customer status
Use EnergyPal's custom CRM/API to look up caller records and personalize conversations
Qualify new inbound leads and book callbacks or advisor appointments
Create cases or support records for post-sale issues
Log all outcomes and dispositions back into EnergyPal's system
Expected Outcomes
Capture all of the 70–80% of inbound calls currently being missed
Reduce lead leakage from callers who are ready to speak but cannot reach a human
Improve funnel movement by qualifying and booking eligible prospects
Reduce manual follow-up burden on the team
Provide a scalable inbound triage layer that can be expanded over time
AYB AI TESTIMONIALS
HEAR WHAT OTHERS ARE SAYING…
Norton Insurance, United Kingdom.
INSURANCE INDUSTRY
BEN WHITE - HEAD OF OPERATIONS
It's been a pleasure working with AYB AI. We had already started looking into how AI can improve productivity & performance and their solution certainly did both of those things.
We have been able to integrate their voice agent in to our lead chasing which has had a huge positive effect. Getting our prospects to answer the phone was our biggest issue (even when they request call backs on our website).
After introducing the voice sales agent, our contact rate has improved from 40% to over 60% - which has exceeded our hopes and targets!
⭐⭐⭐⭐⭐
Advocate Rights Center, Dallas, Texas.
LEGAL INDUSTRY
ROBERT WILSON - FOUNDER
It’s been a great pleasure working with AYB AI.
They are very knowledgeable and experienced in AI-generated communications.
We consider AYB AI to be a valued & trusted partner.
⭐⭐⭐⭐⭐
Blip Payments, Ireland.
FINANCIAL SERVICES INDUSTRY
HUGO URIEN - FOUNDER
Working with AYB AI has been the definition of tailored professionalism. They have had a personal touch from day one and are very quick to respond and make changes as needed.
AYB AI developed several solutions for us that have helped our business become more systematic and efficient.
They take the time to really understand the goal and offer solutions to get there.
I plan to continue to work with AYB AI and would recommend them to anyone who is looking to optimize systems in their business.
⭐⭐⭐⭐⭐
Preferred Fertility Concierge, Atlanta, Georgia.
HEALTHCARE INDUSTRY
TAKERA MITCHELL - FOUNDER
AYB AI exceeded my every expectation and went above and beyond to come up with a solution for my somewhat complex request.
Not only was the service thorough and timely but the customer service was superb.
AYB ensured my needs were understood and followed up with me consistently throughout the entire process.
I highly recommend AYB for any of your automation needs!
⭐⭐⭐⭐⭐
Return On Investment
70–80%
Calls Currently Missed
Of several hundred inbound calls per month going unanswered
$0
Revenue from Missed Calls
Every missed call is a lost opportunity with zero recovery
5x
Higher Intent
Inbound callers have higher intent than cold outbound prospects
The Cost of Missed Calls
Lost revenue & lead leakage - prospects lost forever
Delayed follow-up erodes momentum
Prospects lose interest and engage competitors
Sales reps spend more time chasing instead of converting
Post-sale support issues go unstructured or untracked causing frustrated customers
Return On Investment With AI
All inbound calls answered instantly
More qualified solar consultations booked
Lower operational costs vs. hiring human employees
Better routing for existing leads and assigned prospects
Reduced manual chasing from sales reps
Cleaner data and outcome tracking inside EnergyPal's CRM
Goals & Current Context
Your Goals (From Zoom Call)
Start with inbound triage - inbound is the operational current gap
Immediately answer all overflow calls across multiple EnergyPal functions
Triage callers based on where they are in the customer journey
Qualify & book eligible prospects with an advisor or sales rep
Create cases for post-sale support issues
Reduce missed opportunities from calls that go unanswered
Keep the experience fast, natural, aligned with EnergyPal's brand & processes
Today's Reality
EnergyPal is already staffed for outbound. Inbound is not staffed to the same level. Several hundred inbound calls come in over the course of a month - and EnergyPal estimates that 70–80% of inbound calls are missed.
Inbound Caller Types
Leads EnergyPal has never spoken to
Leads EnergyPal called who are now calling back
Prospects already assigned to a sales rep
Prospects with quotes already in the system
Existing opportunities that need a follow-up booked
Post-sale customers needing support or case creation
The Problem
EnergyPal is missing hundreds of inbound calls each month (approx. 70–80%) - spanning the full funnel from new leads to post-sale support.
Current Inbound Challenges
Missed calls span the full funnel: new leads, returning prospects, assigned opportunities, quote-stage leads, and post-sale support
Sales reps may be on other calls when prospects call back
Some callers are high-intent but do not get routed quickly enough
Support-related calls may not be structured into trackable cases quickly enough
The company is not staffed around inbound the same way it is staffed around outbound
Why This Matters Now
Inbound callers often have higher intent than cold or outbound-only prospects
If a prospect calls back and gets no answer, they lose momentum, resulting in lost revenue
If a prospect is already assigned to a rep, a missed call can slow down the deal
If an existing customer calls with an issue, a missed response creates frustration
Missed calls create operational noise - the team must manually chase, call back, and reconstruct context later
In short: EnergyPal is already generating inbound demand. The opportunity is to stop letting those calls leak out of the funnel, increasing revenue & customer satisfaction.
Scope of Work - Inbound AI Only
✅ In-Scope (Phase 1)
AI inbound triage agents for missed and overflow inbound calls
General agent to capture intent for routing.
5 specific sub-agent flows based on caller type and intent
CRM/API lookup before or during the call to personalize the conversation
New lead qualification flow
Returning lead / callback flow
Assigned prospect / rep follow-up booking flow
Existing opportunity / quote-stage booking flow
Post-sale support triage and case creation flow
Appointment/callback booking where appropriate
Outcome tracking and CRM disposition updates
Initial controlled rollout with all calls manually reviewed for optimization
🚫 Out-of-Scope (Phase 1)
Outbound AI calling
Complex technical support resolution beyond intake, triage, and case creation
Building a fully custom private-server voice stack before the first production test in Retell
Knowledge base-heavy support calls/automation unless scoped later
This proposal focuses exclusively on Inbound AI. Outbound AI and advanced support automation can be scoped in future phases.
Solution Design
Channels: Inbound phone calls routed to the AI agent when calls are missed, overflowed, or directed to a specific inbound AI number.
Voice AI Core
Voice-enabled AI triage agent with front-end intent detection and sub-agent routing based on caller type and CRM lookup
Instant Answer
Answer overflow calls immediately, ask why the caller is calling, and look up the caller in EnergyPal's custom CRM
Smart Routing
Identify whether the caller is new, qualified, assigned to a rep, quote-stage, post-sale, or unknown - then route to the right sub-flow agent
Book & Log
Book appointments or callbacks, create support cases, log transcripts, outcomes, and dispositions - automatically
Performance Tracking
Flag callers who request a human, tune interruption sensitivity, and monitor performance via dashboards
Sub-Agent 1: New Lead / Never Spoken To
New Inbound Lead
For callers interested in going solar who are not found in the CRM or have not yet been qualified.
01
Confirm Identity & Contact Details
Confirm the caller's name and contact information to begin the record
02
Capture Solar Interest
Ask whether they are interested in solar, battery storage, backup power, lowering bills, or comparing options
03
Qualify the Property
Confirm property address, home ownership, property type, utility provider, and average monthly electricity bill
04
Gauge Readiness
Ask whether they are interested in solar, battery, or both - and whether they are just researching or ready to speak with an advisor
05
Book & Confirm
Book a free Zoom consultation with an EnergyPal advisor if qualified, send SMS & email confirmation and reminders, and log all qualification answers and booking outcomes
Sub-Agent 2: Returning Lead / Called Back After Outreach
Returning Prospect
For prospects EnergyPal has already contacted who are now calling back.
What Makes This Flow Different
This caller already exists in the CRM.
The AI avoids repeating the full qualification flow and focuses only on moving the lead forward efficiently - respecting the caller's time and prior engagement.
The AI Will:
Recognize the caller if they exist in the CRM
Confirm they are calling back about solar options
Avoid repeating the full qualification flow where unnecessary
Ask only the minimum missing questions needed to move the lead forward
Book the next available advisor appointment or callback
Send the caller confirmation & reminders via SMS & email
Log the call outcome and update the lead status
Sub-Agent 3: Assigned Prospect / Rep Busy
Assigned to Rep
For leads already assigned to a sales rep where the rep is unavailable or on another call.
Identify the Assigned Rep
Pull the assigned rep's name and details from CRM data to personalize the interaction immediately
Acknowledge the Relationship
Use a personalized intro: "I see you've already been working with [Rep Name] on your solar options." - making the caller feel recognized
Offer to Book a Follow-Up
Present available times if calendar availability is accessible and offer to book a follow-up directly with the assigned rep
Confirm & Notify
Confirm the booking, send confirmation and reminder messages to the caller, and log the callback request while notifying the assigned rep
Sub-Agent 4: Existing Opportunity / Quote-Stage Lead
Quote Stage
For prospects who already have quotes or active opportunities in the system.
Key Principle
This caller is already deep in the funnel.
The AI avoids full re-qualification and focuses on understanding what the caller needs help with - then gets them booked with the right person as quickly as possible.
The AI Will:
Identify the opportunity stage from CRM data
Avoid full qualification unless required
Ask what the caller needs help with
Book the caller with the assigned advisor or rep
Confirm the booking and send confirmation/reminder messages
Log the callback request and notify the assigned rep
Sub-Agent 5: Post-Sale / Support Inquiry
Post-Sale Support
For customers further down the line who need help after the sale.
01
Confirm Customer Identity
Verify the caller is an existing customer using CRM lookup by phone number or other identifier
02
Capture the Issue
Ask what issue or question they are calling about and categorize the support request appropriately
03
Create a Support Case
Create a support case in EnergyPal's system, capturing urgency, preferred callback time, and all relevant details
04
Provide a Clear Next Step
Book a callback, escalate or flag urgent cases based on rules EnergyPal defines, and send confirmation messages
05
Notify Support Team
Alert the support team with the case details, urgency level, and caller information for prompt follow-up
Integrations & Data
EnergyPal Custom CRM
Look up caller by phone number or other identifier
Retrieve caller status, lead stage, assigned rep, appointment status, quote status, or support status
Determine the correct sub-flow
Update dispositions and outcomes after the call
Create callbacks, appointments, or support cases
Log transcript summaries and key fields
Calendar / Scheduling
Advisor consultations
Rep follow-ups and callback appointments
Rescheduling and support callbacks
Rep/advisor availability (team basis if needed)
Time zone validation for TCPA compliance
Booking confirmations and reminders via SMS/email
Telephony / Call Routing
Forward all missed/overflow calls to the AI
Route a controlled daily volume at the start
Route by business function or caller type if available
Use one general agent that triages into sub-agent flows
Conversation Design & Optimization
Our Zoom call highlighted several important production requirements that will shape how the agent is built and tuned.
Shorter Answers
Tighter, more concise responses. The agent will be tuned to avoid long explanations unless the caller asks for more detail.
No Unnecessary Questions
The agent will not over-qualify callers who are already in the system or already qualified - respecting their time and prior engagement.
Interruption Sensitivity
The agent can be tuned to be more interruptible so it does not continue speaking over callers. This will be optimized during testing.
Voice Tuning
We can test different voices, add light background noise, and tune delivery to feel more natural - avoiding the "too perfect" AI sound.
Time Zone Logic
Because EnergyPal serves the whole U.S., booking must account for customer location and time zone for TCPA compliance.
Fallback Behavior
If a caller asks for a live person, the agent can attempt transfer, book the next available callback, route to the assigned rep/team, or flag for human follow-up.
Production Optimization: During the early deployment period, we will manually review every call recording, transcript, and outcome to improve prompts, pacing, routing, and conversion logic.
Implementation Plan & Timeline
Total Estimated Duration: 3–4 weeks for initial build and launch, assuming no delays with access, API setup, calendar rules, and EnergyPal-side approvals.
1
Phase 1 - Kickoff & Access
Confirm inbound call categories, sub-agent flows, success metrics, and routing plan. Provision CRM/API access, booking rules, calendars, time zone logic, and disposition fields. Collect 5–10 historical call recordings per flow.
2
Phase 2 - Agent Build
Build the general inbound triage agent and all 5 sub-agent flows: new lead qualification, returning lead callback, assigned rep follow-up, existing opportunity booking, and post-sale support case creation. Configure voice, tone, and response rules.
3
Phase 3 - Integrations & Routing
Connect CRM/API lookup, calendar/booking logic, call outcome logging and dispositions, and configure inbound call forwarding or test routing.
4
Phase 4 - UAT & Controlled Pilot
Test with EnergyPal internal team. Route a controlled volume of missed/overflow calls (optionally 50–100 calls/day or all missed calls). Review calls and refine flows.
5
Phase 5 - Go-Live & Optimization
Expand routing once performance is validated. Manually review call recordings and transcripts. Optimize response length, routing logic, booking outcomes, and escalation rules. Track all performance metrics.
Deliverables
1
Inbound AI Triage Agent
Fully developed EnergyPal Inbound AI Triage Agent with general inbound triage agent and flows
2
5 Sub-Agents & Logic Flows
Complete sub-agents and logic flows for EnergyPal's main inbound call categories — new lead, returning lead, assigned prospect, quote-stage, and post-sale
3
CRM/API Integration
CRM/API lookup and update logic for caller identification, status retrieval, and disposition logging
4
Calendar Booking Logic
Calendar booking logic for advisor, rep, and callback scheduling with time zone validation and SMS/email confirmations
5
Support Case Creation
Support case creation logic for post-sale callers with urgency capture and team notification
6
Reporting & Optimization
Call recording/transcript review process, disposition and reporting structure, go-live support, and ongoing optimization
Pricing & Terms (USD)
One-Time Setup Fee
$9,995
($1,999 per sub-agent & system of automation)
Custom development of 5 agents
Inbound triage flow design
CRM/API integration
Calendar integration
Testing and pilot deployment
Full EnergyPal ownership rights of all AI agents & systems (pending receipt of final support payment after 6 months)
Payment Milestones: 50% upfront | 50% halfway through development | First monthly maintenance fee due before go-live
Monthly Maintenance
$1,500 p/mo
For the first 6 months
Support and reporting
Manually reviewing calls
Prompt tuning and routing improvements
Minor updates and error handling
System monitoring and monthly optimization
After 6 months: move to a lower monthly fee of $395/month for support, or a break/fix relationship with no monthly fee.
AI Usage & Billing
Usage billed at cost with no markup, directly from the multiple platforms used to develop this solution.
$0.14
Per Minute (Low)
Minimum AI voice minute rate depending on final stack and configuration
$0.17
Per Minute (High)
Estimated AI voice minute rate - exact rate confirmed after agent is developed and tested
$200
Optional N8N Setup
One-time fee for AYB AI to set up EnergyPal's own N8N account self-hosted on their server for backend automation
Monthly Reporting Will Include:
Total inbound calls handled
Total AI minutes used
Calls triaged
Calls booked
Support cases created
Human-transfer / callback requests
Missed-call recovery rate
All other relevant performance metrics
Compliance, Privacy & Guardrails
Inbound Compliance
Because this proposal focuses on inbound calls, TCPA outbound dialing restrictions are not the primary issue for Phase 1. Callers are initiating contact with EnergyPal.
Privacy & Data Handling
The AI will only collect information required to triage, qualify, book, or create a case
All data captured will be logged into EnergyPal's approved systems
Call recordings and transcripts can be retained according to EnergyPal's preferred retention policy
AI Guardrails
The agent will not make guarantees about solar savings, incentives, or federal credits
The agent will not provide final quote/pricing advice
The agent will not overstep into advisor-level recommendations
The agent will route complex or sensitive questions to a human
The agent will identify opt-out / "do not contact" requests via SMS and update systems accordingly
The agent will flag callers who do not want to speak with AI and offer a human callback path
Support & SLA
AYB-AI Support Hours: Monday–Friday, 9am–5pm EST via Zoom, Slack, WhatsApp or email.
P1 - Production Issue
Inbound agent unavailable or critical failure
Response within 2-4 business hours
P2 - Performance Issue
Routing issue or degraded call handling performance
Same business day response
P3 - Minor Change
Script or prompt update, minor configuration adjustment
Within 2 business days
Ongoing Optimization Includes:
Script adjustments based on manually listening to calls
Flow refinements based on EnergyPal feedback
Interruption sensitivity tuning and voice/pacing optimization
Booking and disposition improvements and reporting reviews
Assumptions, Risks & Mitigations
Assumptions
EnergyPal provides CRM/API access and documentation
EnergyPal provides example historical call recordings or transcripts for each main inbound call type