Deployment Planning
The Expertflow CX solution can be deployed On-Premise, in the Cloud by the customer/partner, or hosted by Expertflow. The key difference in hardware sizing is who manages the underlying infrastructure.
💻 Hardware Sizing Quick Reference
Use the table below to quickly identify the necessary hardware documentation based on the customer's scale.
Customer Scale / Goal | Key Requirement | Resource Guide |
Proof of Concept (PoC) | Minimum resources for functional validation (up to 10 agents); no HA. | |
Small-to-Mid Enterprise | Production system (up to 100 agents); requires HA. | |
Large Enterprise | High-volume production (500+ agents); requires HA/DR. | |
Partner Hosting Own Cloud | Infrastructure sizing and per-tenant hardware increments. | |
Partner Selling Expertflow Cloud | Capacity-based licensing model; no hardware management. |
Expertflow CX Solution Components
Expertflow CX is a unified platform composed of mandatory core components and optional, modular add-ons (deployed via Helm sub-charts).
Mandatory Core Platform Components
These components form the foundation of the solution and are required for all deployments (voice, digital, or blended). They are deployed on a Linux/Kubernetes cluster.
Component | Description | Business Function |
CX-Core | Channel Manager, Routing Engine, Agent Manager, Activities. | Unified Agent State, Skill-Based Routing, Interaction Persistence, and Agent Desktop API. |
Data Pipeline Orchestrator | Apache Airflow engine for ETL (Extract, Transform, Load). | Moves transactional data from NoSQL persistence to reporting SQL database and executes custom data integration jobs. |
CX Analyzer | Apache Superset based reporting interface. | Real-time and Historical Reporting for business insights. |
Voice & Conversational Components (Included/Optional)
Component | Description | Capability | Deployment |
Media Server | FreeSWITCH-based server for processing voice, traditional IVR, and secure recording. | Handles Inbound/Outbound Voice, IVR, and Interaction Recording. | Deployed as a Linux System Service. |
Jambonz | Voice gateway enabling voice stream forking. | Enables advanced Voicebots/Conversational IVR via integration with third-party AI engines. | Optional Add-on (Helm Sub-chart). |
Conversation Studio | Node-RED low-code environment. | Visual design tool for building inbound/outbound conversation flows (IVR, Voicebot). | Optional Add-on (Helm Sub-chart). |
💼 Business Application Add-ons (Optional)
Component | Description | Capability | Deployment |
WFM (Workforce Management) | Backend and Frontend application for resource planning. | Forecasting, Scheduling, and adherence tracking. | Optional Add-on (Helm Sub-chart). |
Surveys and Campaigns | Backend for outbound calls/SMS and post-interaction surveys. | Automated Outbound Dialing, Secure Interaction Surveys, and Campaign Management. | Optional Add-on (Helm Sub-chart). |
Digital Channels | Separate sub-charts for WhatsApp, Viber, Telegram, etc. | Provides out-of-the-box integration for all supported social/digital media channels. | Optional Add-on (Helm Sub-chart). |
🚀 Deployment Options & Sizing Strategy
CX supports three flexible deployment options. Select the customer's scenario below to find the correct hardware guide. See CX Sizing Methodology for sizing guidelines.
Customer-Managed Deployment (On-Premise / Customer Cloud)
In this model, the customer provides and manages all necessary hardware and infrastructure components (Linux OS, Kubernetes cluster, physical or virtual servers, and redundancy measures).
Customer Profile | Responsibility | Recommended Hardware Guide |
|---|---|---|
PoC (Proof of Concept) | Customer provides and manages all hardware and infrastructure (Linux/Kubernetes). | |
Small-to-Mid CC (Up to 100 Agents) | Customer provides and manages all hardware and infrastructure (Linux/Kubernetes). | |
Large Enterprise (500+ Agents) | Customer provides high-availability (HA) infrastructure with dedicated DR planning. |
Partner-Managed Deployment (Partner Hosted Cloud)
This model is for Multi-Tenant Service Providers where the Partner owns and manages the underlying infrastructure, providing isolated resources for each tenant they host.
Customer Profile | Responsibility | Recommended Hardware Guide |
|---|---|---|
Multi-Tenant Service Provider | Partner owns and manages the infrastructure, providing isolated resources per tenant. |
Expertflow-Managed Deployment (Expertflow Cloud Hosted)
In this model, the Partner sells tenants that are hosted on Expertflow's cloud infrastructure. Expertflow manages all hardware, scaling, and platform maintenance.
Customer Profile | Responsibility | Recommended Hardware Guide |
|---|---|---|
Partner sells tenants on Expertflow's hosted platform. | Expertflow manages all hardware, scaling, and platform maintenance. |
Hybrid CCaaS Model (Expertflow Cloud + Regional Data)
This specialized model is for customers with strict data residency requirements. Expertflow hosts the cloud applications, but the customer retains the data layer (Databases, ETL, AI Services) in their premises/region.
Customer Profile | Responsibility | Recommended Hardware Guide |
|---|---|---|
High-Compliance Customer | Customer provides and manages the infrastructure for all data components (Databases, ETL, AI Services). Expertflow manages the application layer. |
Key Integrations
The platform integrates with major contact center and enterprise systems:
Contact Center: Cisco UCCE, UCCX
CRM: Integrates via available APIs.
AI Engines: Supports third-party TTS (Text-to-Speech), ASR (Automatic Speech Recognition), NLU (Natural Language Understanding), and LLMs (e.g., LLAMA) for advanced agent assist and autonomous bots.