RMT Engineering Logo
Chat and omnichannel messaging

Omnichannel Chat

Clients choice to choose the preferred communication channel to reach to the contact center, the entire history of conversation is available to the agent as a complete flow.

  • Same or different agent - same knowledge
  • One customer - across channels
  • Web widget in one snippet
We reply within one business day. No newsletter.
8 Chat channels in one queue
1 Universal inbox and customer key
76 Report definitions across channels
One inbox WhatsApp, web, social and email

Omnichannel Integrated .

A Client may prefer to use the convenient channel of choice. A conversation may start with the phone call, move to a WhatsApp Chat or SMS or Telegram, Or ChatBOT or Website or though Social Media every source is integrated to understand the chain of communication and process the response to the Client. The Client never feels disconnected and finds continuity in communications. No need to repeat the problem, reiterating past conversations but straightway driving the communication with a solutioning approach. Moving between channels is not a handover to another system.

One customer One channel.

A separate tool per channel means a separate queue per channel. The same customer ends up at the front of one queue and the back of another, waiting twice for the same answer, and nobody on the floor notices until the complaint arrives on a third channel. One inbox is not a cosmetic merge. It is what makes ownership, service levels and reporting mean the same thing everywhere.

How OptiML Omnichannel works

The path a conversation takes through Omnichannel Chat

Eight chat channels converging through channel adapters into one conversation model, and out to the agent, the desk and analytics WhatsApp Web chat SMS Email Messenger Instagram Telegram Discord Channel adapters One conversation model unified customer key · phone, email, your identifiers The same AI agent Agent Desk Analytics and quality

Every channel enters through its own adapter and lands on the same conversation, keyed to the same customer. The agent, the desk and the reporting see one thing, not eight.

By the numbers

Few Facts - at a glance

  • 8+

    Chat channels on one conversation model — WhatsApp, Web, SMS, Messenger, Instagram, Telegram, Discord, Email, Social media

  • 1

    All-in-One inbox with shared routing, Memory and Ownership across all the Channels

  • 76

    Report definitions covering every channel, so the numbers add up across the Channels

The process

How It Works

The four stages, end to end

  1. Channel adapters

    Each channel connects through its own adapter. WhatsApp Business Cloud API, the Web widget, an SMS provider, a Mailbox, a Social account. Whatever is native to that channel survives the trip. A template stays a template, a story reply stays a story reply, and an email keeps the thread it was replying to instead of arriving is isolation.

  2. Identity resolution

    The unified customer key matches an inbound number, an email address or one of your own identifiers against an existing record. An anonymous web session is stitched to a known customer the moment they identify themselves. That usually happens mid-sentence, not at a login screen.

  3. One conversation

    Earlier contacts are retrieved into the agent's context before it replies. Nobody explains the situation twice. When a human is needed the thread moves to the universal inbox with its history attached, and the person picking it up reads that history instead of asking for it again.

  4. One set of numbers

    Ownership, service levels, tags and sentiment are recorded on the conversation rather than on the channel. The 76 report definitions read from that record. Volumes and handle times add up across channels rather than sitting in eight separate tools that each count a contact slightly differently and each produce a monthly deck nobody quite trusts.

Features

Every capability you need in one module

1. The universal inbox

Voice callbacks, WhatsApp threads, web chats and emails in one list. Filter by owner, channel, tag, sentiment or SLA risk. Assign a thread, transfer it, drop an @mention on whoever handled this customer last time, leave an internal note the customer never sees, or fire a macro with the personalisation tokens already filled in. Ownership locks, so two agents never end up answering the same customer at once.

  • Assign, transfer, mention
  • Filters for owner, channel, tag, sentiment and SLA risk
  • A collision lock, so two people cannot reply to the same thread at the same moment
  • Personal and team macros, with personalisation tokens filled from the record

See the agent workspace the inbox sits in

Illustrative interface. Names and settings are sample data, not a performance claim.

2. WhatsApp, properly

WhatsApp Business Cloud API, with template management inside the platform, interactive buttons and lists, media and commerce flows that read a Commerce catalogue. Inbound replies land in the same queue as everything else. Outbound campaigns are held to the same consent and as per DNC rules and other start/stop rule applied.

Business Cloud API Template management Interactive buttons and lists Commerce flows and WooCommerce catalogues Consent and DNC shared with the dialer

See how consent and DNC are enforced on outbound

3. The web widget

One snippet to embed. The same widget ships as an NPM package, a React component or a sandboxed iframe. Themeable through presets or CSS variables, voice-capable, 14 locales with right-to-left rendering for Arabic and Hebrew, and around 254 KB gzipped. Cookie consent, Do-Not-Track and data-deletion controls ship inside it.

  • One script tag
  • NPM package, React component or sandboxed iframe when a snippet will not do
  • Themeable by preset or CSS variables, and voice-capable
  • 14 locales, with right-to-left for Arabic and Hebrew
  • Cookie consent, Do-Not-Track and data deletion built in

That is the whole install

The widget points at the same agent that answers the phone, with the same knowledge and the same guardrails behind it. Swap the snippet for the NPM package or the React component when you want the conversation inside your own application shell. 

<!-- Drop this on any page. That is the whole install. -->
<script
  src="https://cdn.[domain]/widget.js"
  data-agent="support-en-in"
  data-channel="web"
  data-voice="enabled"
  defer></script>

Talk to our team about embedding the widget

Illustrative interface. The page and the exchange are sample content.

4. Proactive, not just reactive

Trigger a message on cart abandonment, churn risk, a support event or any signal you can define. Visitor identification stitches an anonymous session to a known customer and carries their traits into the conversation: lifetime value, churn risk, plan tier. So the opening line names what the person is actually doing, instead of "how can I help you today".

Cart abandonment Churn risk Support events Any custom signal Traits carried into the conversation

See how triggers and journeys are built

5. Cross-channel continuity

A unified customer key ties phone, email and your own identifiers together. When someone who called yesterday messages today, their history is retrieved and injected into the agent's context before it answers. No transcript search after the fact. No "let me explain the situation again".

  • Same record, whichever channel they arrive on
  • One key across phone, email and your own identifiers
  • History retrieved into context, not hunted down by hand while the customer waits

One customer, four contacts

Voice · yesterday 14:02
Asked about the policy renewal date WhatsApp · today 14:41
Asked for the document again — sent, no repetition Email · 02 Aug
Invoice request, resolved by the AI agent Web chat · 12 Jul
Renewed the annual plan

Illustrative timeline. Dates and events are sample data.

6. Multi-brand and white-label

One tenant, many brands. Per-brand themes, multiple origins, white-label domains. Separate knowledge and separate agents behind each one. Multi-brand exists for BPOs running several client programmes off one roster, and for enterprises that carry more than one customer-facing brand and would rather not stand up a separate deployment behind each of them. The brands stay apart for the customer. The roster stays shared for you.

Per-brand themes Multiple origins White-label domains Separate knowledge and agents per brand

Read how tenants, hosting and data residency are set up

Use Cases

Omnichannel Chat- Value Creation delivers value

Cross-channel continuity

A WhatsApp message finishes yesterday's call

Life Insurance

Scenario

A policyholder rings at two in the afternoon about a renewal date, then messages on WhatsApp that evening asking for the document itself. The customer key matches the mobile number to the call record, the afternoon conversation is retrieved into the agent's context before it replies, and the document goes out in the very next message without anyone asking which policy they mean.

Outcome

The second contact is one message long instead of a re-explanation. Both sit on one conversation record, so quality review and reporting see a single customer rather than two.

A mid-size outsourcing provider

Three client programmes on one tenant

A mid-size outsourcing provider

Scenario

Three clients, three brands, three sets of knowledge, three white-label domains, one operations team. Each brand gets its own widget theme, its own origins and its own agents. The supervisor still sees one queue, and can move people between programmes when the afternoon volume moves.

Outcome

A new client programme launches on the existing tenant instead of a new deployment. Per-brand reporting stays separate. Capacity planning stays shared.

Proactive triggers and visitor identification

A stalled checkout gets a message, not a survey

Retail Establishment

Scenario

A returning shopper loads the payment step and stops. Visitor identification has already stitched the session to their record, so the cart-abandonment trigger fires with order history and lifetime value in context, and the opening message names the actual basket rather than offering a generic greeting.

Outcome

The conversation starts where the customer got stuck. If it needs a person, it lands in the same queue as WhatsApp and email, owned by whoever is free.

At a glance

Specification

The numbers and limits, without the sales copy

Specification for Omnichannel Chat
Specification Detail
Channels WhatsApp Web Chat SMS Messenger Instagram Telegram Discord EMail
Widget delivery Script tag NPM React component Sandboxed iframe
Widget size ~254 KB zipped
Widget locales 14, with RTL support for Arabic and Hebrew
Widget privacy controls Cookie consent Do-Not-Track Data deletion
File uploads 20 MB, Images, PDF and Documents with inline preview
Identity Unified customer key across Phone, Email and Custom identifiers
Collaboration Assignment ownership @mentions internal notes macros
Inbox filters Owner channel tag sentiment SLA risk
Proactive triggers Cart abandonment churn risk support events custom signals
Multi-brand Per-brand themes, multiple origins, white-label domains, separate knowledge and agents
SMS providers Twilio Telnyx Brevo
Reporting 76 report definitions, covering every channel on one conversation model
FAQ

Questions,
answered

What teams ask us before they roll out Omnichannel Chat — how it works, what it needs from your side, and what happens when it gets something wrong

Still not sure?

Talk to a specialist and get a straight answer.

Ask our team

Yes. Each channel in OptiML Omnichannel Chat is its own adapter, so most teams open with WhatsApp and the web widget and switch the rest on as they need them. SMS runs through Twilio, Telnyx or Brevo, and Messenger, Instagram, Telegram and Discord are enabled per tenant. Because the conversation model and the customer key never change, adding a channel does not re-cut the reports you already built.

Omnichannel Chat runs WhatsApp on the Business Cloud API and keeps template management inside the platform, so approved templates keep working, along with interactive buttons, lists and rich media. Inbound replies route into the same queue as web chat and email. Outbound WhatsApp campaigns are held to the same consent and DNC rules the outbound dialer enforces, which means one opt-out covers both.

One script tag on the page. If a snippet is not how your front end works, the OptiML web widget also ships as an NPM package, a React component or a sandboxed iframe. It is around 254 KB gzipped, themeable by preset or CSS variables, and available in 14 locales with right-to-left rendering for Arabic and Hebrew.

The thread moves into the universal inbox with its history intact and is assigned or claimed by a human agent, with ownership locking so two people do not answer the same customer at once. Internal notes and @mentions stay on your side of the conversation. Omnichannel Chat records the handover and the reason for it on the conversation, so quality management can pull the whole thing up afterwards.

Cookie consent, Do-Not-Track handling and data-deletion controls ship inside the widget, so a visitor can decline tracking and still use chat. Attachments up to 20 MB, whether images, PDFs or documents, are stored against the conversation record rather than the channel, with inline preview in the inbox. Which region the data lives in is a hosting decision you make when the tenant is set up. Omnichannel Chat inherits that choice rather than making it for you.

Keep exploring

Additional supporting modules

Native modules that share the same record, knowledge and governance

Platform module

Voice AI Agents

The ninth way in. Real-time phone conversations that write to the same conversation record as every chat channel.

Explore
Platform module

Agent Desk

Where a human picks the thread up: call controls, customer 360, a grounded copilot and wrap-up written for them.

Explore
Platform module

Ticketing & Case Management

When a conversation becomes a case. Native tickets, 11 ITSM connectors and a universal adapter for whatever else you run.

Explore
Connected solutions

Where Omnichannel Chat is used

The Solutions pages that lean on this module, and what it looks like once it is configured for a particular floor, job title or job to be done.

17 solutions built on this module

By industry

Sector floors that run on it

8

By role

Job titles that answer for it

3

By use case

Jobs it is put to work on

6
Browse every solution
Talk to a specialist

Watch a WhatsApp thread finish a phone call

Give us one phone call and one WhatsApp thread from the same customer, and tell us which channels you actually run. We will land them on a single conversation record so you can see what the second agent opens with.

  • 30 minutes
  • Your channels, one conversation record
  • No sandbox tour

Book your slot

Leave your email and our team will come back to you within one business day.

or reach us directly
Call our team Email us

Your details stay private. We never share them.

This website uses cookies.

Cookies are small text files that allow us to create the best browsing experience for you on our site. By continuing to use this website or clicking "Accept & Close", you are agreeing to our use of cookies. To understand how we use cookies or how to manage them, please see our cookies policy.

Ask OptiML

Powered by RMT Engineering