The shape of your inbound support demand — how many conversations shoppers started each day over the last 30 days.
At a glance
Conversation Volume Trend is a Conversation Intelligence metric tracked from Intercom data. It plots how many new conversations were created per day across a rolling 30-day window, so the support lead and founder at Blitz can see whether demand is climbing, flat, or spiking. As an area chart it is built to be read at a glance — the curve tells you the trend, not a single number. Pair it with New vs Closed and Open Backlog Trend below to see whether your team is keeping pace with the inbound you see here.
| What it counts | New conversations created per day, bucketed by created_at, over the trailing 30 days. Each conversation from the Intercom conversations endpoint counts once on the day it was opened — regardless of channel, assignee, or current state. |
| Sample type | Backend API data from Intercom (conversations search by created_at), refreshed on the standard data refresh. |
| Why it matters | Volume is the denominator behind almost every other support metric. A rising trend warns you to staff up before first-response and backlog cards slip; a falling trend can signal a deflection win (better help docs) or a tracking gap. Founders read it as a proxy for customer friction and seasonal load. |
| Reading the value | Read the slope, not the dot. A steady baseline with predictable weekday/weekend rhythm is healthy. A sustained climb means demand is outpacing your plan; a sharp one-day spike usually traces to an incident (failed payments, a stockout, a shipping delay) — confirm against Volume Spike and the cross-channel cards. |
| Currency | number |
| Time window | 30D |
| Alert trigger | — |
| Sentiment key | null |
| Roles | owner, operations |
Calculation
Calculated automatically from your Intercom data. Vortex IQ queries theconversations endpoint for every conversation with created_at inside the trailing 30 days, then groups the count by calendar day in your reporting time zone. Each day becomes one point on the area chart. Conversations are counted by creation date only — reopened, merged, or reassigned conversations are not double-counted, and current state (open, closed, snoozed) does not affect the trend. See the worked example below for a typical reading.
Worked example
A representative reading of Conversation Volume Trend for Blitz, a sports retailer running support on Intercom. Across a normal fortnight the curve sits around 40-55 new conversations on weekdays and dips to ~20 on weekends — a healthy, predictable rhythm. Then on a Tuesday it jumps to 130. Reading the slope, the support lead sees this is a one-day spike, not a new baseline. Cross-referencing Volume by Channel shows the surge came in via email, and Top Topics (Tags) shows a cluster ofpayment-failed and order-stuck tags — pointing at a checkout incident rather than organic growth. The founder uses the trend to justify a temporary extra agent for the day, then watches New vs Closed to confirm the team clears the spike rather than letting it roll into backlog. For deeper investigation, use Vortex Mind to trace the spike to its upstream cause; for natural-language exploration, ask Ask Viq “what drove the conversation spike on Tuesday?”.
Sibling cards merchants should reference together
| Card | Why merchants reach for it |
|---|---|
ic_new_vs_closed | Pairs created volume against closed volume to show whether the team is keeping pace with the inbound you see here. |
ic_backlog_trend | Shows where unworked volume accumulates when intake outruns throughput. |
ic_volume_by_channel | Breaks the same total down by source so you can see which channel is driving a move. |
ic_alert_volume_spike | Fires when a single day breaches the 30-day baseline — the alerting counterpart to this trend. |
ic_top_tags | Tells you what the volume is about when a move needs a reason. |
Reconciling against the vendor’s own dashboard
Where to look in Intercom’s own dashboard: Open Reports → Conversations in Intercom and select the “New conversations” (or “Conversations created”) metric over a custom 30-day range. Intercom’s chart counts by conversationcreated_at, the same anchor Vortex IQ uses, so the daily shape should line up closely. The Inbox views (“All”, “Unassigned”) show current state, not created-per-day, so they will not match this trend — use the Reports surface to reconcile.
Why the Vortex IQ value may legitimately differ:
| Reason | Direction | What to do |
|---|---|---|
| Time zone. Intercom reports in the workspace time zone; Vortex IQ buckets days in your merchant reporting time zone. Conversations near midnight can fall on a different calendar day. | Marginal | Confirm the two time zones match, or compare weekly totals where the boundary effect cancels out. |
| Period boundary. Vortex IQ uses a 30-day rolling window; Intercom Reports defaults to fixed ranges or calendar weeks. | Variable | Set Intercom to the identical custom 30-day range. |
| Bot / automated conversations. Depending on your profile, conversations opened purely by bots or Series may be filtered from the Vortex IQ count. | VortexIQ lower | Match Intercom’s “include bot conversations” toggle to your profile setting. |
Known limitations / merchant FAQs
Q: How often does Conversation Volume Trend update? The card refreshes on the standard data refresh (typically every 30-60 minutes for live integrations). The most recent day’s bar will keep growing until the day closes. Q: Why does my Intercom dashboard show a different number? The most common reasons are time-zone alignment (day boundaries near midnight), period-boundary differences (rolling 30 days vs a fixed range), and whether bot-only conversations are included. Match these settings in Reports → Conversations before assuming a real divergence. Q: Does a reopened conversation count again? No. The trend counts each conversation once, on its originalcreated_at day. Reopens show up in Reopen Rate, not here.
Q: Can I customise the alert threshold?
This card itself has no threshold — it is a read-only trend. Use Volume Spike for the anomaly alert, which is configurable per profile in the Sensitivity tab.