How to Create a WhatsApp Team Inbox by Department

If you haven’t connected WhatsApp Business API yet, start with our full setup guide. This post assumes that part is done, your team is already replying from a shared inbox, and you’ve hit the next problem: everyone is in the same flat inbox, and Sales, Support, and Billing conversations are all mixed together.

The fix isn’t a second WhatsApp number, it’s splitting your one connected number into multiple team inboxes inside the same platform, each with its own routing, its own queue, and its own reporting.

When to split into team inboxes

A single shared inbox works fine until one of these becomes true:

• Agents from different departments are all pulling from the same unassigned queue and have to read every message to know if it’s theirs
• You want to report on Sales response time separately from Support response time
• Different teams need different SLAs, quick replies, or escalation rules
• New hires in one department are seeing (and occasionally answering) conversations meant for another

If none of those apply yet, stick with one inbox, splitting too early just adds routing overhead you don’t need. If they do apply, department-based team inboxes are the natural next step.

Step 1: Create your team inboxes

In Resolvenyx, go to Settings → Inboxes → Team Inboxesand create one for each department, Sales, Support, and Billing being the most common split. A team inbox is not a second WhatsApp number: it’s a filtered view and queue that sits on top of your existing connected number.

Name each one clearly and assign it a colour, agents will see this as a quick visual cue when switching between queues during the day.

Step 2: Set up department-based routing rules

With team inboxes created, the next question is how an incoming message ends up in the right one. Auto-routing rules handle this without any manual sorting:

• Keyword-based routing: messages containing “refund”, “invoice”, or “payment” route to Billing; “price”, “quote”, or “buy” route to Sales
• First-contact vs existing-contact routing: new numbers messaging for the first time route to Sales, contacts with an open ticket route to Support
• Time-of-day routing: route to an on-call department outside standard hours

Rules run in order, so put your most specific conditions first and a catch-all (usually routing anything unmatched into a general queue for manual triage) last.

Step 3: Assign agents to their team inbox

Under Settings → Users, assign each agent to one or more team inboxes. An agent only sees the queue(s) they’re assigned to by default, which is what actually solves the “reading everyone else’s conversations” problem. Agents who legitimately need visibility across departments, team leads, for example, can be added to multiple team inboxes.

From here, the same assignment mechanics you already use apply per team inbox: manual claiming, auto-routing, and workload balancing all work within a team inbox’s own unassigned queue, not across the whole account.

Step 4: Report on each team separately

Once conversations are flowing into the right team inbox, response times, resolution rates, and volume all become measurable per department instead of blended into one account-wide number. This is usually the moment splitting into team inboxes actually pays off: Support can be judged against a support SLA, and Sales against a sales one, without one dragging the other’s numbers around.

Review the split every quarter or so. Teams change shape, and it’s common to add a fourth team inbox (Onboarding, VIP, or a specific product line) once the first three prove useful.

A quick example

A 12-person e-commerce support team we’ve seen this pattern with split into three team inboxes: Support (general order/delivery questions), Billing (refunds and payment issues), and VIP (a label-triggered route for their top spend tier). Keyword rules handled about 70% of routing automatically; the rest landed in a general queue where a team lead triaged manually each morning. Within a month, average first response time dropped because agents were no longer scanning through department-mismatched conversations to find their own.