Skip to main content
Card class: Non-HeroCategory: Conversation Intelligence
The conversations your team has hit snooze on. Deferred work that will land back in the queue when it wakes.

At a glance

Snoozed Conversations is a Conversation Intelligence metric tracked from Intercom data. It is a live count of conversations currently in the snoozed state — paused by an agent to resurface at a set time (waiting on a customer reply, a stock restock, a refund clearing). For the support lead at Blitz it is the hidden half of the queue: snoozed work is invisible in the open backlog but will reappear. A large or growing snooze pile can mask a backlog problem and create a future wave of re-opens.
What it countsConversations whose state is snoozed right now, from the Intercom conversations endpoint. Each snoozed conversation has a snoozed_until timestamp; when it passes, the conversation wakes back to open.
Sample typeReal-time data from Intercom (conversations filtered to state = snoozed).
Why it mattersSnooze is a legitimate tool — but it also lets a queue look cleaner than it is. Conversations snoozed and forgotten become aged, unhappy customers. A snooze count that keeps climbing usually means agents are deferring rather than resolving, and every snoozed item is a future addition to Open Backlog Trend when it wakes.
Reading the valueRead the level and its trend. A modest, stable count (genuinely waiting on customers or restocks) is healthy. A rising count, or a count large relative to your open queue, is a flag — it can hide a backlog and predict a re-open spike when a batch of snoozes wakes at once.
Currencynumber
Time windowRT
Alert trigger
Sentiment keynull
Rolesowner, operations

Calculation

Calculated automatically from your Intercom data. Vortex IQ queries the conversations endpoint in real time for conversations whose current state is snoozed and counts them. Snoozed conversations are deliberately excluded from the open backlog while they sleep — they hold a snoozed_until time and automatically return to open (waking) when that time passes, at which point they rejoin the open count. This card is a point-in-time snapshot, not a rolling window. See the worked example below for a typical reading.

Worked example

A representative reading of Snoozed Conversations for Blitz on Intercom. On a normal day the card reads ~12 snoozed — customers asked to confirm a size, or orders snoozed until a restock lands. The support lead is relaxed: open backlog reads 28, snoozed is a small slice. Then over a fortnight snoozed climbs to 60 while open backlog stays at 28. The lead spots the pattern: agents are snoozing “where is my order” conversations to next week rather than resolving them, because a supplier shipment is late. The open queue looks healthy but 60 customers are simply parked. When the shipment date slips and 40 of those snoozes wake on the same morning, Open Backlog Trend jumps and Reopen Rate ticks up. The founder, forewarned by the snooze pile, has already staffed for the wave. For deeper investigation, use Vortex Mind to see which topics are being snoozed; for natural-language exploration, ask Ask Viq “how many snoozed conversations wake in the next 24 hours?”.

Sibling cards merchants should reference together

CardWhy merchants reach for it
ic_backlog_trendSnoozed work re-enters this backlog when it wakes — read them as one queue.
ic_open_conversationsThe live open count that a large snooze pile can artificially flatter.
ic_alert_oldest_openA repeatedly re-snoozed conversation often becomes your oldest open once it wakes.
ic_reopen_rateA batch of snoozes waking at once shows up as reopens here.
ic_volume_trendContext for whether snoozing is rising because intake rose or because resolution stalled.

Reconciling against the vendor’s own dashboard

Where to look in Intercom’s own dashboard: In the Intercom Inbox, filter to the “Snoozed” state to see the live list and count. This is the most direct match for this card. Reports does not surface a standalone snoozed total, so the Inbox filter is your reconciliation point. Why the Vortex IQ value may legitimately differ:
ReasonDirectionWhat to do
Snapshot timing. A snoozed conversation can wake between the Vortex IQ refresh and the moment you read the Inbox.VariableRefresh both views together; expect small drift on a busy workspace.
Inbox scope. The Inbox “Snoozed” filter may be scoped to a specific team or admin; the card counts the whole workspace.Inbox lowerClear team/assignee filters in the Inbox before comparing.
Just-woken conversations. A conversation that woke seconds ago is open in one view and may still show snoozed in a stale cache of the other.MarginalAllow for refresh lag near snoozed_until boundaries.
Cross-connector reconciliation: snoozes tied to “waiting on restock” overlap with the commerce inventory signal — when a stock topic is heavily snoozed, check Complaints on Out-of-Stock SKUs. For divergence investigations, use Vortex Mind.

Known limitations / merchant FAQs

Q: How often does Snoozed Conversations update? This is a real-time card; it reflects the live snoozed count at the last refresh, typically within a minute or two of the workspace state. Q: Why does my Intercom Inbox show a different number? Usually snapshot timing (conversations waking between reads) or an Inbox filter scoped to one team. Clear filters and refresh both views to reconcile. Q: Do snoozed conversations count in my open backlog? No. They are excluded while snoozed and rejoin the open count when they wake. That is exactly why a big snooze pile can hide a backlog — watch this card alongside Open Backlog Trend. Q: Can I customise the alert threshold? This card ships without a threshold, but you can add a sensitivity rule per profile in the Sensitivity tab — for example, alert when snoozed exceeds a share of your open queue.

Tracked live in Vortex IQ Nerve Centre

Snoozed Conversations is one of hundreds of KPI pulses Vortex IQ tracks across Intercom and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. Start for free or book a demo to see this metric running on your own data.