Operator proof · ChatSac · Existing public case study

One conversation carrying the whole record

A multi-API platform orchestrating CRM, ERP and WhatsApp for retail service operations. Built, operated and grown by the founder of isonew, and included here as the engineering record behind how we build agents.

Start the AI Opportunity Diagnostic

Founder and operator proof. Ronan Pinho founded and built ChatSac and ran it as an operating company. It is not an isonew client.

Service line: AI Agents & Automation

Context

What the operation looked like

Retail networks were running customer service across WhatsApp while their commercial and inventory truth lived in separate CRM and ERP systems. Agents worked with three screens and answered from memory. ChatSac was built to make one conversation carry the whole record.

Constraint
Messaging volume is unforgiving: a customer waiting in a chat notices every second of lookup time. The platform had to orchestrate across systems that were never designed to be queried inside a conversation, without introducing a new place for data to go stale.

Decisions

What was decided, and why

01
Treat WhatsApp as an operational surface, not a marketing channel, so every interaction had to resolve against the systems of record.
02
Orchestrate rather than replace: CRM and ERP stayed authoritative, and the platform composed their data at conversation time.
03
Keep a human agent in the loop with assisted responses instead of a fully autonomous bot, because a wrong commercial answer costs more than a slow one.
04
Design for multi-tenant deployment early, because the same orchestration pattern repeated across store networks.

System built

What exists, module by module

Conversation orchestration layer

WhatsApp interactions resolved against CRM and ERP records in the flow of the conversation.

Multi-API integration

Connections to commercial, inventory and customer systems with retry and failure handling.

Agent assist surface

Suggested responses and record context presented to the human agent handling the conversation.

Multi-store deployment model

One platform pattern rolled out across a store network rather than rebuilt per location.

Built with

OpenAI

Agent orchestration

CRM

ERP

PostgreSQL

Node

Docker

Evidence

Figures, with scope and source

Average handle time, after deployment

One 180-store retail network, down from 8 minutes · Existing isonew.digital public case study

ClaimScopeSource

3,000+ customers served through the platform
ScopeCumulative, as published
SourceExisting isonew.digital public case study
93% retention
ScopeCustomer base, period as published
SourceExisting isonew.digital public case study
150% year over year growth for three consecutive years
ScopePlatform revenue growth
SourceExisting isonew.digital public case study
Up to $100K ACV
ScopeLargest annual contract value band
SourceExisting isonew.digital public case study
180-store deployment, average handle time from 8 minutes to 40 seconds
ScopeOne retail network, average handle time before and after deployment
SourceExisting isonew.digital public case study

Support automation panel

Rolling 12 weeks · refreshed daily

Live

Tickets auto-resolved

First reply

0swas 6h 12m

Monthly cost avoided

Share of tickets closed without a human

last 12 weeks

Escalated to a human32%
Wrong answer rate0.7%

Sample

Support Automation Runbook

Escalation rules, confidence thresholds and the rollback path.

8 pp

Download the sample (PDF)

The artifact handed over on this engagement. Client figures are replaced with representative ones, the structure is unchanged.

These are the only figures we publish for ChatSac. Each one carries its scope and its source, and no additional metric has been added or extrapolated.

Ronan doesn't sell theory. He runs the framework on his own companies first, then ships it.

Matt Ham · Founder & CEO, ChurnDefense · Existing isonew.digital public case study

Current state

Where it stands today

ChatSac continues to operate as an independent company. It is included here because it is the engineering record behind isonew's approach to agents, orchestration and production operations.

Next expansion
For isonew clients, the reusable part is the orchestration pattern: conversation surfaces resolving against systems of record with a human in the loop.

What this does not claim
The figures above were published by ChatSac with scope attached. isonew claims the engineering record behind them, not a promise that another company will see the same numbers.

Written and reviewed by

Ronan Pinho

Founder of isonew and of ChatSac, co-founder and CRO of ChurnDefense. Over ten years building and operating technology companies. Based in Apex, North Carolina. Works in English, Portuguese and Spanish.

Last reviewed 2026-08-14

Keep going

Other write-ups and the commercials

Same format, different relationship — plus what work like this costs.

Operator proof · Live · Operating companyChurnDefense — Product, billing and support signals turned into one risk score and a prescribed action.A behavioural intelligence layer for SaaS retention: product, payment and support data unified into a real-time risk score and a prescriptive playbook for every account a CS team manages.Open

Client systems · Signed · In deliveryVerdeOps — Field operations for a Triangle landscaper, built beside the existing crew.A custom operating system for a Triangle landscaping business, replacing scheduling, crew coordination and job records spread across spreadsheets and messaging.Open

Client systems · Signed · In deliveryHilo — Atelier counter and production record, migrated without stopping the floor.A custom operating system for a Triangle tailoring atelier, including migration of the existing Talus workflow into an owned system.Open

One next step, and it is a paid one on purpose

The AI Opportunity Diagnostic is a fixed-scope engagement. You leave with a ranked plan you can act on with us or without us.

Start the AI Opportunity Diagnostic