Skip to content

Solution

Customer service that answers, resolves and follows up

Queues with real SLA targets, an IVR that lets callers finish things themselves, one inbox for voice, WhatsApp, SMS and email, help for the agent while the call is live, and a survey afterwards that turns a bad score into a case with a name on it.

14-day free trial · no credit card required

The DialerBee queues page for a demo tenant: the Collections, Sales, Renewals and Support queues, each with its routing strategy, priority, maximum wait and SLA target, above the cards for active queues, agents, active calls and mailboxes
Built by BroadNet, 22 years in telecom 11 languages, Arabic dialect-aware BYOC, your carriers Compliance-supporting controls

Quick answer

What is customer service in DialerBee? It is the inbound side of the platform assembled for a support team. ACD queues route by FIFO, round-robin or longest-idle, each with its own priority, max wait, wrap-up seconds, recording mode and an SLA target with a breach threshold; callers hear their position and estimated wait. The IVR builder handles the contacts that never needed a person: check or manage a booking, request a callback, hear queue status, choose a language, look the caller up in the CRM. The omnichannel inbox puts voice, WhatsApp, SMS and email threads in one list with visible states and assignment to an agent or a queue, with a screen-pop and Customer 360 by phone number. Live agent assist builds a summary during the call, reads Arabic-aware sentiment, and suggests a response and a disposition the agent confirms. Afterwards, surveys ask CSAT or NPS on five channels and route detractors to a recovery queue as cases with an owner and an outcome, while the wallboard and the queue, inbound and agent activity reports show what actually happened. Agents handle inbound queues and outbound campaigns from one desktop; automatic blending is planned.

What you get

Everything in one workspace

ACD queues

Each queue picks the next agent by FIFO, round-robin or longest-idle, with its own priority, max wait, wrap-up seconds and recording mode.

SLA target and threshold

Every queue carries the answer target it is measured against and the threshold that counts as a breach, so service level is a setting rather than an opinion.

Position and estimated wait

Callers can hear where they are in the queue and how long it is likely to be, with music on hold in between. People wait longer when they know.

DID routing

Route each number to a queue, a user, an IVR flow, a campaign or an AI voice agent, in bulk when a batch of numbers arrives, with CSV export of the inventory.

IVR self-service

The IVR builder covers caller self-service: check or manage a booking, request a callback, hear queue status, choose a language, CRM lookup mid-call.

Business hours and after-hours

Business-hours and holiday calendars per tenant, enforced by schedule gates, with a text-to-speech after-hours message and an SMS fallback template.

One conversation list

Voice, WhatsApp, SMS and email threads in the same inbox, each new, queued, assigned, in progress or waiting on customer, assigned to an agent or a queue.

Customer 360 on the pop

The record opens with the call: profile, history, notes and a CRM tab, keyed on the number, with BroadNet CRM enrich, contact create and disposition push.

Live agent assist

A summary that builds during the call, Arabic-aware live sentiment, a suggested response the agent can edit or ignore, and a suggested disposition they confirm.

Surveys with recovery cases

CSAT and NPS asked by voice IVR, AI voice, WhatsApp, SMS or a short link, and detractors routed to a queue as cases with a name and an outcome.

Wallboard and reports

Queue depth and agent states live on the wallboard, then queue performance, inbound call, agent activity and service-level reports behind it.

Listen, whisper, barge

A supervisor can listen to an inbound call, whisper to the agent, barge in, or force a logout, while coaching rules surface prompts during the shift.

The queue

Service level is a setting, not a hope

Support starts with queues that behave. Each one picks the next agent by FIFO, round-robin or longest-idle, and carries its own priority, max wait, wrap-up time and recording mode, so a billing queue and a complaints queue do not have to be run the same way. Skills-based routing is planned.

Every queue also carries an SLA target and the threshold that counts as a breach. That is what turns service level into something the wallboard can show live and the queue performance report can settle afterwards.

  • FIFO, round-robin and longest-idle routing today; skills-based routing is planned.
  • Priority, max wait, wrap-up seconds and recording mode set per queue
  • SLA target and breach threshold per queue, live on the wallboard
  • Position and estimated wait announced to the caller, with music on hold
A queue's routing settings in DialerBee: the Agents, Routing, SLA, Overflow and Media tabs open on the Collections queue, with its priority level, its routing strategy, its business-hours schedule, the caller ID used for queue callbacks and the skills it requires

Before an agent is needed

The calls that never needed a person

A large share of support contacts are a customer checking or changing something they could have handled themselves. The IVR builder has nodes for exactly those: check a booking, manage a booking, request a callback and keep your place, hear queue status, choose a language, or have the CRM looked up mid-call.

Out of hours the same flows still answer. Business-hours and holiday calendars are set per tenant and enforced by schedule gates, and the caller hears a text-to-speech message in the voice you chose, with an SMS fallback template so they know they were heard.

  • Caller self-service nodes for bookings, callbacks, queue status and language choice
  • CRM lookup during the call, so the flow knows who is calling
  • Business-hours and holiday calendars per tenant, enforced by schedule gates
  • After-hours text-to-speech message with a chosen voice, plus an SMS fallback template
The call handling settings in DialerBee for a demo tenant: the no-answer announcement written in Arabic with the voice that speaks it, and the missed-call SMS fallback with its sender number and message template

One inbox

Work the customer, not the channel

Voice, WhatsApp, SMS and email threads sit in one conversation list. Each conversation is new, queued, assigned, in progress or waiting on customer, and belongs either to a named agent or back to a queue, so nothing sits in a thread nobody owns. Inbound email into the inbox is planned.

The customer record opens with the call. Customer 360 is keyed on the phone number, with profile, history, notes and a CRM tab, and BroadNet CRM enrich, contact create and disposition push behind it. Agents handle inbound queues and outbound campaigns from one desktop; automatic blending is planned.

  • Voice, WhatsApp, SMS and email in one conversation list with visible states
  • Screen-pop with Customer 360 by phone number, and a tenant webform in an iframe
  • Live summary, Arabic-aware sentiment, suggested response and suggested disposition
  • Transcripts redacted, searchable, retained on your policy, with legal hold and export
The DialerBee browser-based agent desktop, showing the three-column layout with contact information, WebRTC call controls and an embedded CRM panel, plus live transcription and one-click dispositions

Afterwards

A bad score becomes a case with a name on it

The survey is the part most support operations do badly. In DialerBee a published survey can be asked by voice IVR, by the AI voice runtime, over WhatsApp, over SMS or on a web page behind a short link, and the published version is locked, so this quarter's CSAT means what last quarter's meant.

Then the detractors and low scores are routed to a queue as recovery cases, assigned to a person and closed with an outcome. Alongside them sit the reports: queue performance, inbound calls, agent activity and the insights centre, with exports in CSV, Excel and PDF that keep Arabic readable.

  • CSAT, NPS, yes/no, consent and open text, asked on five channels
  • Detractors routed to a queue as cases, assigned and closed with an outcome
  • Analytics that state their denominators, plus scheduled daily, weekly or monthly reports
  • Agent activity and utilisation reports today; forecasting and scheduling are planned.
The reports screen in the DialerBee admin console for the Northwind Demo tenant: fifteen standard report cards covering campaign summary, agent performance, CDR, compliance audit, cost analysis, quality metrics, call recordings, abandoned calls, queue performance, DNC compliance, inbound and outbound calls, agent utilization, lead conversion and supervisor overview, above a date-range selector and a generate report button

How it compares

Support stitched from separate toolsDialerBee customer service
Where the channels liveA phone system, a chat tool and a shared mailbox, each with its own historyVoice, WhatsApp, SMS and email in one conversation list with shared states and assignment
What the agent sees on an inbound callA number, then a search in another windowA screen-pop with Customer 360 by phone number: profile, history, notes and a CRM tab
Service levelCounted afterwards from an exportAn SLA target and breach threshold on every queue, live on the wallboard
Out of hoursThe call rings outSchedule gates on business hours and holidays, a text-to-speech message in your voice, and an SMS fallback template
What happens to a bad scoreA row in a spreadsheetA recovery case in a queue, assigned to a person and closed with an outcome
LanguagesWhatever the phone system's prompts support11 languages across the platform, Arabic dialect-aware, with Arabic that survives the export

Frequently asked questions

What is the DialerBee customer service solution?+

It is the inbound side of the platform assembled for a support operation: ACD queues with SLA targets, an IVR builder with caller self-service, an omnichannel inbox covering voice, WhatsApp, SMS and email, live agent assist during the call, surveys with recovery cases afterwards, and the wallboard and reports over all of it.

How are calls routed to agents?+

FIFO, round-robin and longest-idle routing today; skills-based routing is planned. Each queue also has its own priority, max wait, wrap-up seconds and recording mode, so different queues can behave differently without a separate system.

Can we run support and outbound campaigns with the same team?+

Agents handle inbound queues and outbound campaigns from one desktop; automatic blending is planned. That means the same agent, the same reports and the same compliance controls, with the mix decided by supervisors rather than by an algorithm.

Which channels are in the inbox?+

Voice, WhatsApp, SMS and email threads in one conversation list, each in a visible state and assigned to an agent or a queue. Email is sent from your own per-tenant SMTP. Inbound email into the inbox is planned.

What does the agent get help with during the call?+

A summary that builds while the call is running, live sentiment that is Arabic-aware, a suggested response the agent can use, edit or ignore, and a suggested disposition they confirm. Transcription runs on demand, and transcripts are redacted, searchable and retained on your policy.

What happens outside business hours?+

Business-hours and holiday calendars are set per tenant and enforced by schedule gates. Outside them the caller hears a text-to-speech message in a voice you chose and previewed, and an SMS fallback template goes out, so a closed office is not a ring-out.

How do we measure the service we are giving?+

Queue depth, queue stats and agent states are live on the wallboard, and behind it sit queue performance, inbound call, agent activity and service-level reports, plus the insights centre. Exports come as CSV, Excel or a PDF with your tenant branding, with Arabic rendered correctly.

Does the survey do anything beyond producing a number?+

Yes. Detractors and low scores are routed to a queue as recovery cases, each assigned to a person and closed with a recorded outcome, so a bad score produces a follow-up you can audit rather than only a chart.

See your own support flow on a demo tenant

A 30-minute walkthrough on a demo tenant, in English or Arabic.