Message Inbox Opening and Closing Checklist: What to Review at the Start and End of the Support Day

What this checklist is for

A WhatsApp inbox without an opening and closing routine always breaks in the same place: nobody knows who answered what, conversations from yesterday are left open, and the first message of the morning gets answered two hours late because whoever opened didn't check what was left pending from the previous shift. The problem isn't the tool, it's that there's no written procedure that anyone can follow the same way, even the person covering a day off.

This checklist is meant to be printed or copied into a spreadsheet and checked off. Each item says what to review and how to know it's done right. If an item doesn't apply to your operation, you skip it; you don't replace it with an invented one. The idea is for it to work whether one person is handling it or there are five agents across three shifts.

Stage 1: shift opening

Opening isn't turning on the computer. It's reconstructing the state of the inbox before touching a single new message. If the person opening doesn't understand what was left pending, they'll answer wrong or duplicate responses that were already given.

• Review the list of unanswered conversations from the last 24 hours. OK sign: each one has an assigned owner or a sent reply. If there's one with neither, it gets handled before opening the rest.
• Read the handover from the previous shift. OK sign: you know which conversations were left waiting for something specific (a customer detail, a confirmation, a quote) and not just 'pending'.
• Verify that out-of-hours auto-replies are active if applicable. OK sign: the message someone receives outside business hours is the one your business wants to send, not an old one that was left loaded.
• Confirm that the WhatsApp channel is connected and receiving. OK sign: a test message comes in or you see new activity in the inbox; if nothing comes in, check before assuming it's 'quiet'.
• Check that the active templates and campaigns are the ones for the day. OK sign: there's no old campaign still firing messages you no longer want to send.
• Review tags and new contacts from the previous day. OK sign: the contacts who wrote in were tagged so you can segment them later.

A common mistake at opening is starting with the newest message. The correct order is the reverse: first the old unanswered ones, then the new ones. A customer who wrote last night at 23:00 and still has no reply is more urgent than one who wrote five minutes ago, even if the second one is at the top of the list.

Stage 2: during the shift (controls that prevent the closing disaster)

Closing isn't fixed at closing. It's fixed during the day. These are the controls that, if you skip them, will blow up at 19:55 when you want to leave.

• Every conversation that gets answered is assigned to someone. Sign of OK: if tomorrow someone asks 'who saw this?', there's an answer.
• Conversations waiting on something from the customer are marked as such, not as generic pending items. Sign of OK: at the end of the shift you can tell 'waiting for them to send me the ID' apart from 'I didn't answer it'.
• AI-suggested responses that are used are reviewed before sending. Sign of OK: no response went out without a human approving it. If your tool suggests and approves with one click, that click is the control.
• Incoming voice messages are transcribed or listened to, not left for later. Sign of OK: no voice message is left unprocessed at the end of the day.
• Tags are applied in the moment, not at the end. Sign of OK: if someone else takes over the conversation mid-day, they understand the status without asking.

Stage 3: end of shift

Closing has a single goal: that whoever opens tomorrow doesn't have to guess anything. If the handoff is clear, opening takes five minutes. If not, forty are lost.

• List the open conversations waiting for a response from the business. Sign of OK: they're all there, none is missing, and each one has a note about what's expected.
• List the ones waiting on something from the customer. Sign of OK: they're separated from the previous ones, so tomorrow they aren't answered again for no reason.
• Write down the handoff: what happened today, what's still pending, what needs follow-up. Sign of OK: someone who wasn't on the shift understands the status by reading three lines.
• Check that there are no unassigned conversations. Sign of OK: zero orphan conversations. If there are any, they're assigned before closing.
• Confirm that the day's campaigns were sent or rescheduled. Sign of OK: no campaign is left half-sent that will fire off on its own tomorrow.
• Note down the technical problems or unusual cases of the day. Sign of OK: if something failed (a message that didn't go out, a rejected template, a down channel), it's recorded for review, not lost in someone's memory.

How to know an item is really done (and not just checked off)

A checklist rushed through doesn't work. The difference between 'I reviewed it' and 'it's done' is that the second one has an observable signal. These are the signals that leave no room for interpretation: either they're there or they're not.

• A conversation is 'properly closed' when the next action is defined and assigned: someone has to do something, or the customer has to reply. 'Pending' is not a status, it's an excuse.
• A handover is 'properly written' when someone who wasn't on the shift can pick it up without asking questions. Test it: if you have to explain something verbally, the handover is incomplete.
• A label is 'properly applied' when it helps you find the conversation later. If you label everything as 'inquiry', it's useless.
• A channel is 'properly connected' when you receive messages, not when you set it up. Yesterday's configuration doesn't guarantee today's reception.
• A campaign is 'properly closed' when you know how many messages went out and how many are left. If you don't know, it's not closed.

What to do when something fails

No checklist is 100% met. What matters is what you do when an item doesn't come back OK. These are the typical cases and what to do in each one. How long it takes to resolve depends on each case and on whether the problem is in your configuration or on Meta's side; there is no guaranteed timeframe.

Shift handover: the minimum format

The handover is the part of the checklist that gets skipped the most and costs the most. A minimum handover has three fields and doesn't need more: what happened today (one line), what's waiting for a reply from the business (list), and what needs follow-up tomorrow (list). If your team writes paragraphs, nobody reads them. If they write three lines with the essentials, it gets used.

• What happened today: incidents, volume spikes, campaigns sent. One line.
• Waiting for a response from the business: conversations where the customer has already done their part and ours is pending.
• Follow-up tomorrow: conversations that need to be picked up again even if the customer hasn't written.
• Nothing else. If the handoff grows, it stops being read.

How to measure whether the routine is working

A checklist without measurement becomes an empty ritual. To know whether the opening and closing are working, look at three things over the course of the week: how many conversations are left unanswered from one day to the next, how many times someone had to ask out loud what was happening with a conversation, and how many technical incidents were logged versus how many were resolved. If the conversations that cross over from one day to the next go down and the verbal questions go down, the routine is working.

You don't need a new tool to measure this. A spreadsheet with three columns is enough. What does help is for the inbox to have the minimum features so the checklist can actually be executed: conversation assignment, tags, internal notes, and visibility into what was left unanswered. If your inbox doesn't allow assigning or tagging, the checklist becomes hard to follow no matter how much discipline you put into it.