The size of your open queue, day by day. The line you never want to see drifting up.
At a glance
Open Backlog Trend is a Conversation Intelligence metric tracked from Intercom data. It plots the number of conversations still in an open state at the end of each day over the trailing 30 days, so the support lead and founder at Blitz can see whether unresolved work is accumulating or draining. Where New vs Closed shows the daily flow, this card shows the level — the standing pool of customers still waiting. A rising line is the clearest sign that intake is outrunning throughput, and it leads first-response and CSAT decline by days.
| What it counts | The count of conversations in the open state at each day’s close over the trailing 30 days. A conversation enters the count when created (or reopened) and leaves it when closed or snoozed. Snoozed conversations sit outside backlog until they wake. |
| Sample type | Backend API data from Intercom (conversations state reconstructed from created_at, statistics.last_close_at, and snooze events), refreshed on the standard data refresh. |
| Why it matters | Backlog is the inventory of customer dissatisfaction. Every conversation in it is someone waiting. A steadily rising trend means waits will lengthen and SLA breaches will follow, regardless of how fast individual agents reply. Founders treat a climbing backlog as a staffing or process signal, not a one-off. |
| Reading the value | Read the direction and the floor. A flat line — even a high flat line — means you are at equilibrium. A line that never returns to its floor after spikes means each incident leaves residue. A >20% week-over-week rise is the configured warning that the queue is structurally growing. |
| Currency | number |
| Time window | 30D |
| Alert trigger | rise >20% WoW |
| Sentiment key | conversation_backlog |
| Roles | owner, operations |
Calculation
Calculated automatically from your Intercom data. Vortex IQ reconstructs the open-conversation count at the end of each day across the trailing 30 days. A conversation is counted as open on a given day if it was created (or reopened) on or before that day and had not yet been closed or snoozed by that day’s close, usingcreated_at, conversation.statistics.last_close_at, and snooze/unsnooze events. Snoozed conversations are excluded for the duration of the snooze and re-enter the count when they wake. The week-over-week comparison drives the threshold: a >20% rise versus the prior week fires the backlog-growing alert. See the worked example below for a typical reading.
Worked example
A representative reading of Open Backlog Trend for Blitz on Intercom. The backlog line hovers around 30 open conversations all month — a comfortable floor for a two-agent team. After a delayed shipment from a supplier, “where is my order” conversations pour in; the line climbs from 30 to 78 over a week. That is a 160% rise, well past the 20% WoW trigger, and the alert fires. The support lead reads the line: it is not spiking and recovering, it is stepping up and holding — residue, not a blip. New vs Closed confirms closed is trailing new; Top Topics (Tags) confirms the cluster isshipping-delay. The founder posts a proactive status note to deflect new contacts and adds weekend cover. Over ten days the line walks back down toward 30 as the team out-closes intake. For deeper investigation, use Vortex Mind to trace the rise to its driving topic; for natural-language exploration, ask Ask Viq “what’s in our open backlog right now?”.
Sibling cards merchants should reference together
| Card | Why merchants reach for it |
|---|---|
ic_new_vs_closed | The flow that feeds this level — the daily reason backlog rises or falls. |
ic_open_conversations | The live, right-now value of the same queue the trend tracks over time. |
ic_snoozed_count | The conversations parked outside backlog that will re-enter it when they wake. |
ic_alert_oldest_open | Surfaces the worst-aged item hiding inside a growing backlog. |
ic_workload_by_team | Tells you whether capacity exists to drain the backlog or whether you must add it. |
Reconciling against the vendor’s own dashboard
Where to look in Intercom’s own dashboard: Intercom does not expose a point-in-time historical backlog chart directly — the Inbox shows the live open count today, and Reports → Conversations shows flow (new and closed), not the standing level. The closest native equivalent is to track the live “Open” Inbox count daily, or to derive backlog from the new-minus-closed running total in Reports. Vortex IQ reconstructs the daily level for you, which is why this trend has no single Intercom view to match against directly. Why the Vortex IQ value may legitimately differ:| Reason | Direction | What to do |
|---|---|---|
| Snooze treatment. Vortex IQ excludes snoozed conversations from backlog; the Intercom Inbox “Open” filter may include or hide snoozed depending on the view. | Variable | Check whether your Inbox count includes snoozed; compare against Snoozed Conversations. |
| Reconstruction vs live. Vortex IQ rebuilds historical state from timestamps; rare gaps in webhook/event history can shift a past day’s level slightly. | Marginal | Trust the live Open Conversations card for “right now”; use this trend for direction. |
| Time zone. End-of-day boundaries use your reporting time zone. | Marginal | Align time zones before comparing a specific day’s level. |