Powerful Features
Everything you need to deliver exceptional customer support, built-in and ready to use.
Core Features
Essential CRM functionality that every team needs.
Ticket Management
Full-featured ticket system with unique UID tracking, priority levels, and smart assignment.
- Unique Ticket UID (TCK-XXXXXXXX)
- Priority & SLA tracking
- Smart ticket assignment
- Real-time status updates
- Search by any field including UID
Multi-tenant Support
Manage multiple clients or departments from a single installation with complete data isolation.
- Tenant isolation
- Role-based access control
- Custom branding per tenant
- Centralized administration
Analytics Dashboard
Real-time insights into team performance, ticket metrics, and customer satisfaction.
- Response time metrics
- Resolution rates
- Agent performance tracking
- Custom report builder
Pipeline Management
Visual pipeline for tracking leads and deals through your sales process.
- Drag-and-drop interface
- Custom pipeline stages
- Deal value tracking
- Win/loss analysis
Quality Assurance System Exclusive
Our groundbreaking QA system automatically validates ticket closures and ensures resolution quality. No other CRM has this built-in.
Closure Validation
Automatically validates that tickets are closed with proper resolution proof from the customer.
Proof Detection
Scans messages for resolution keywords to confirm the customer was satisfied.
Premature Closure Flag
Flags tickets closed without proof for QA review and agent coaching.
Agent Metrics
Track closure quality scores per agent for performance reviews.
Resolution Proof Keywords Detected:
Competitor Comparison - QA Features
Survey System
Comprehensive customer feedback with three survey types, including the unique Customer Effort Score.
CSAT (1-5)
Customer Satisfaction score measuring overall experience with your support team.
- Classic 1-5 scale
- Auto-trigger on resolution
- Trend analysis
NPS (0-10)
Net Promoter Score measuring customer loyalty and likelihood to recommend.
- Standard 0-10 scale
- Promoter/Detractor classification
- Benchmark tracking
CES (1-7) Unique
Customer Effort Score measuring how easy it was to resolve the issue. Critical for churn prediction.
- 1-7 ease-of-resolution scale
- Churn prediction indicator
- Process improvement insights
Configurable Surveys
Fully customizable survey settings per tenant. Choose which survey type and when to trigger.
- Per-tenant settings
- Keyword-based triggers
- Custom branding
How We Solve the Problems That Matter
Most tools bolt on "compliance" as a checkbox. We built these into the engine, because the teams we serve — multi-center operations across Nigeria and Africa — can't afford the alternative. Underneath the trust layer sits the work itself: turning conversations into resolved tickets, and turning interest into revenue.
Privacy you can prove, not just promise
The problem: a privacy policy is words until a regulator asks for proof. Our answer is enforcement, not prose. Marketing over email, SMS, or WhatsApp is blocked the moment a contact opts out of that specific channel — and every consent grant, withdrawal, and data request is written to an immutable record. Customers can object or request their data through a public center and we act within the NDPA's 30-day window, with no penalty.
- Independent email / SMS / WhatsApp opt-out
- Signed, tamper-proof links (consent & surveys)
- 30-day Data Subject Request handling
An audit trail that holds up
The problem: "who changed this, and from which center?" is the first question in any incident. Our answer: every significant action is logged with the actor, their center/region, the requester's IP, and a before/after of what changed — and the trail is append-only, so it can't be quietly rewritten after the fact.
- Actor + center + IP on every entry
- Before/after change capture
- Auditor-scoped, exportable, immutable
Multi-center by design, not by patch
The problem: a Lagos hub and an Abuja service center should never see each other's customers by accident. Our answer: a region → hub → service-center model where agents see only what they own, supervisors see only their team, and customers see only their own data — enforced server-side, not in the UI.
- Region / hub / service-center hierarchy
- Supervisor-scoped team visibility
- Server-side scoping (no client trust)
Every channel, one customer
The problem: a customer who emails, texts, and WhatsApps is treated as three different people. Our answer: one customer record across email, SMS, and WhatsApp, worked from a single console — with each channel's consent respected independently, so opting out of SMS never silences an essential service message.
- Unified email + SMS + WhatsApp
- Provider-agnostic gateways (Termii, Sendchamp…)
- Per-channel consent, no cross-contamination
Support that actually closes the loop
The problem: tickets slip between channels and agents, SLA breaches go unnoticed, and "resolved" often means "never checked." Our answer is a single ticket spine — every email, SMS, WhatsApp, and widget message becomes one traceable ticket with an SLA timer, owner, and full history. Agents escalate to a supervisor and transfer to a colleague without losing context; QA re-checks closures; and the same audit trail above records who did what.
- Unified inbox across every channel
- SLA timers + supervisor escalation
- QA on closure, zero context loss on transfer
Lead generation that converts, not just collects
The problem: leads are captured and then leak — lost between the form, the rep, and the close. Our answer is one pipeline from first touch to won. Leads flow in from the web widget, surveys, and imports into a Kanban board (New → Contacted → Qualified → Proposal → Negotiation → Won/Lost), get assigned to the right agent, and convert into contacts and tickets without re-keying. Supervisors see only their team's pipeline; customers only ever see their own journey.
- Capture → Kanban pipeline → close
- Lead-to-ticket conversion, no re-entry
- Team-scoped pipeline, server-enforced
Data that expires on schedule
The problem: keeping everything forever is itself a compliance risk. Our answer is storage limitation you control: per-data-type retention policies (customers, tickets, leads, feedback, survey responses, audit logs) with a default action of anonymisation so referential integrity and the audit trail survive. Rules run from the same job engine as the rest of the system and are configured by admins in one place — never hardcoded.
- Per-type retention windows (NDPA s.32)
- Anonymise vs. delete, your choice
- Runs on the unified job engine
A breach plan that beats the 72-hour clock — with a human in the loop
The problem: a breach is a security event, not just a customer-data form — and the NDPA expects a person, not a script, to decide what gets sent. Our answer is a breach register that treats every incident (unauthorised access, malware, ransomware, leak, lost device, insider, phishing, misconfiguration) as a cyber event with affected systems and data subjects. Declaring one starts the 72-hour NITDA clock and pre-fills the NDPA breach checklist; a human reviews and completes it, then manually sends the NITDA and data-subject notices. Nothing goes out automatically — the system only reminds you when the clock is running out.
- 72-hour NITDA clock, surfaced + audited
- NDPA breach checklist (system/DPO pre-filled, human-confirmed)
- Human-sent NITDA + subject notices (never automatic)
Live by default — no special infrastructure
The problem: live dashboards usually need a dedicated real-time server that many hosts
won't allow. Our answer is automated live regions over plain HTTPS — no special runtime.
The in-app notification bell refreshes every 15 seconds, and the operations
console's autonomous engine re-checks SLAs, queue spikes and breaches every
2 minutes; scheduled jobs (retention, breach reminders, webhook delivery)
run automatically every 15 minutes while the console is open, or via your
host's scheduled tasks pointing at cron.php. Every outbound webhook payload is
signed (HMAC) so your systems can verify it came from us.
- Live notifications every 15s — no dedicated real-time server required
- Autonomous engine console sync every 2 min
- Signed outbound webhooks (HMAC)
AI-Powered Features
A memory-backed AI that learns your business, drafts replies, and hands off to a human the moment it should.
AI Copilot Reply Drafting
Agents click Draft with AI in any ticket. The reply is grounded in your knowledge base and the customer's long-term memory, then the agent reviews and edits before sending — the human stays in the loop.
Long-Term Memory (L0–L3)
Every conversation is captured as L0, then distilled into atomic facts (L1), scenarios (L2) and a customer persona (L3). Replies are recalled with blended keyword + vector search, so the AI remembers context across tickets.
AI Help Widget + Human Takeover
Embed a chat widget on any site. The AI answers from your knowledge and memory; if it can't, the customer asks for a human and an agent takes over — the chat becomes a managed ticket instantly.
AI Training & Curation
Teach the AI directly: add facts, or paste a policy document and let it distil durable knowledge. Review, correct, and erase memories (including on data-subject requests) from one panel.
Per-Tenant Memory Quotas
Super-admins cap how much long-term AI memory each organization may store — a built-in control for the NDPA storage-limitation principle — with automatic pruning when a tenant hits its limit.
Bring-Your-Own AI Keys
Each tenant supplies its own AI provider keys. Embeddings run through OpenAI-compatible, NVIDIA NIM, or a local Ollama instance — your data and models stay under your control.
External AI Help Widget
Drop a self-service assistant onto any website — and let a human step in the moment it's needed.
One-Line Embed
Add the widget to any site with a single iframe — no build step, works on any platform.
<iframe src="https://your-crm.com/widget.php?tenant=1"
style="position:fixed;bottom:20px;right:20px;width:380px;height:600px;border:0;border-radius:16px;"
title="Support"></iframe>
AI That Knows Your Business
The assistant answers from your knowledge base and the customer's long-term memory, and asks to connect to a human when it can't help — never inventing policy.
Human Takeover, Built In
When a customer requests a person, agents see it in the Live Inbox, take over, and the chat becomes a linked ticket — the customer keeps talking in the same window.
Runs Anywhere
Standard HTTPS with automatic refresh — no dedicated real-time server required, so it deploys cleanly on ordinary shared hosting.
Ready to Get Started?
Join thousands of teams using NexusX CRM to deliver exceptional support.