Email automation dashboard
Back to Case Studies
SignsSA
SignsSAManufacturingScale-up

Intelligent Email Response System

An AI agent that qualifies leads using company knowledge, getting to "no" faster so your team focuses on ideal clients

Key Outcomes

01

Human-like pacing

Implemented a strategic delay mechanism, and work hours to make AI responses feel natural and less robotic.

02

Context-aware knowledge

The system learned the company's business by scraping the entire website and internal documentation.

03

Getting to the no

The AI quickly disqualifies 50% of enquiries, allowing the team to focus on ideal clients who value quality.

The Challenge

Lost in the noise

The sheer volume of enquiries meant that quality leads were being lost amongst enquiries that were never going anywhere. Sales staff spent far too much time replying to clients who would never value the company's service or quality.

Without an effective qualification process, the team couldn't differentiate between serious prospects and those who would never proceed due to unrealistic budget expectations - particularly for complex services like full vehicle wraps.

The inability to quickly "get to the no" meant valuable time and energy was wasted on leads that would never convert, while ideal clients weren't getting the attention they deserved.

High-value opportunities buried under low-quality leads that consumed staff time.

Our Approach

Company Knowledge Base and AI Processing

The Kipanga Email Intelligence Protocol

Kipanga scraped the client's entire website to build a vector knowledge base. Every product, service, and FAQ was indexed and made searchable by the AI.

We integrated this with Outlook, and the AI drafts a response but holds it in a queue for a dynamic amount of time to mimic human behavior before sending. This is based on the time of day and the day of the week, as well as the business hours.

The delay was crucial - instant AI responses feel robotic and often get worse reply rates. The dynamic delay hits the sweet spot of being responsive but believable.

Phase 01

Knowledge Capture

Website scraping and vector database creation

Phase 02

AI Training

Vector database with company-specific context

Phase 03

Integration

Outlook webhook and AI processing

Phase 04

Optimization

Delay tuning and response quality improvement

The Solution

Digital Sales Agent

An intelligent email handler that uses the company's pricing guide to quickly qualify or disqualify leads. The system "gets to the no" on 50% of enquiries, allowing the sales team to focus exclusively on ideal clients and high-value opportunities.

For qualified leads, the AI pre-qualifies them with budget and timeline questions before routing to the team. This ensures every conversation that reaches a human is worth having.

The system respects business hours (7am-4:30pm) and implements strategic delays to make responses feel natural rather than robotic. Managers get flagged immediately when a high-value lead comes in.

Lead qualification & disqualification
Pricing guide integration
Dynamic natural delay
Business hour scheduling
Website-based knowledge base
Manager escalation routing

Results

Measurable Business Impact

Cost savings
$100k+
Annual Savings
Capacity increased
160+
Hours Saved
Quick "no" responses
50%
Leads Disqualified
Remaining leads
100%
Pre-qualified

Technology

Built With

Backend

Python
Python
Node.js
Node.js

Infrastructure

n8n
n8n
Pipedream
Pipedream

AI/ML

OpenAI
OpenAI
Gemini
Gemini

"Guy and the team have made AI & automation easy to understand, removed the 'tech talk' and been a breeze to deal with. The automated email responses and client research tool is going to be extremely valuable in terms of time savings, better staff utilization and cleaner client service. Couldn't recommend these guys enough."

Andrew Lawson

Andrew Lawson

Owner, SignsSA

Start Your Project

Ready to achieve similar results?

Let's discuss how we can help transform your business with technology.