Top Topics (Tags), Intercom

The most-applied conversation tags over 30 days — what your customers are actually contacting you about. How to read it, why it matters, and how to act on it.

Metrics type: Supporting Metrics • Category: Conversation Intelligence

A ranked bar chart of your most-used conversation tags. The voice-of-customer summary your founder reads first.

At a glance

Top Topics (Tags) is a Conversation Intelligence metric tracked from Intercom data. It ranks the conversation tags applied most often over the last 30 days, turning a month of free-form support into a short, ordered list of themes — sizing, returns, shipping-delay, payment-failed, and the like. For the founder at Blitz it is the fastest read on what is driving contacts; for the support lead it is where to point macros, help articles, and product fixes. A tag jumping 50% in a week is an early signal that something upstream broke.

What it counts	The number of conversations tagged with each tag in the trailing 30 days, ranked highest to lowest. Counts come from the Intercom tags model joined to conversations; a conversation with multiple tags contributes to each.
Sample type	Backend API data from Intercom (tags endpoint joined to tagged conversations), refreshed on the standard data refresh.
Why it matters	Tags are the only structured “why” Intercom carries for a conversation. Ranked over a month they show where demand concentrates, so you can fix the top driver instead of replying to it one ticket at a time. A spiking tag is a leading indicator of an incident — often visible here before volume or backlog moves materially.
Reading the value	Read the ranking and the week-over-week deltas. A stable top five is your steady-state demand. The action signal is a tag rising fast: any tag up more than 50% week over week trips the alert and warrants investigation. A long tail of rarely used tags suggests inconsistent tagging discipline, which weakens every topic-based read.
Currency	number
Time window	30D
Alert trigger	any tag +50% WoW
Sentiment key	null
Roles	owner, operations

Calculation

Calculated automatically from your Intercom data. Vortex IQ reads your workspace tags from the tags endpoint and counts, for each tag, how many conversations carried it within the trailing 30 days. Because a conversation can hold several tags, it contributes one count to every tag applied to it — so the tag counts do not sum to the conversation total. Tags are ranked descending for the horizontal bar chart, and each tag’s current 30-day count is compared to the prior week to drive the +50% week-over-week alert. See the worked example below for a typical reading.

Worked example

A representative reading of Top Topics (Tags) for Blitz on Intercom. The steady-state top five reads sizing (310), returns (190), shipping (150), order-status (120), payment (60) — typical for a sports-apparel retailer where fit questions dominate. One week the payment tag jumps from 60 to 140 (+133%), tripping the +50% alert. The support lead drills in and sees the tagged conversations cluster on a single card type. Cross-referencing the Support Spike on Failed Payments cross-platform card confirms an Adyen refusal pattern on that BIN range. The founder escalates to the payments provider; once resolved, the payment tag settles back to baseline the following week. Meanwhile the persistently high sizing tag prompts a separate, slower fix: a size-guide overlay on product pages to deflect the demand entirely. For deeper investigation, use Vortex Mind to trace a spiking tag to its upstream cause; for natural-language exploration, ask Ask Viq “which tags grew fastest this week?”.

Sibling cards merchants should reference together

Card	Why merchants reach for it
ic_volume_trend	Tells you whether a spiking tag also moved total volume or just shifted the mix.
ic_volume_by_channel	Pairs the “what” with the “where” — which channel carries each topic.
ic_negative_ratings	Surfaces which topics are also generating bad CSAT, so you fix the painful ones first.
ic_xc_complaints_on_oos	Joins stock/refund-tagged conversations to inventory — the cross-platform version of a stock topic.
ic_xc_support_spike_failed_payments	Confirms whether a spiking payment tag traces to real Adyen refusals.

Reconciling against the vendor’s own dashboard

Where to look in Intercom’s own dashboard:

Open Reports → Tags (or the “Conversation topics” report) in Intercom and set a 30-day range. Intercom lists each tag with its conversation count; the ranking should match this card’s bars. You can also filter the Inbox by a tag to inspect the underlying conversations, but the Inbox shows live state rather than the 30-day applied count.

Why the Vortex IQ value may legitimately differ:

Reason	Direction	What to do
Tag-applied timestamp vs conversation date. Vortex IQ counts a tag if it was applied to a conversation created in the window; Intercom may count by when the tag itself was applied. A tag added late to an old conversation can fall in a different period.	Variable	Compare weekly totals where the boundary effect washes out.
Multi-tag conversations. Both systems count a conversation under each of its tags, so tag counts exceed conversation counts.	Neutral	Do not expect tag counts to sum to total conversations.
Archived or renamed tags. A renamed tag may appear as a new series; archived tags drop out.	Variable	Reconcile your tag taxonomy before comparing rankings.

Cross-connector reconciliation: when a payment- or stock-related tag spikes, validate it against the Adyen and commerce sibling cards rather than treating the tag as ground truth on its own. For divergence investigations, use Vortex Mind.

Known limitations / merchant FAQs

Q: How often does Top Topics (Tags) update? The card refreshes on the standard data refresh (typically every 30-60 minutes for live integrations).

Q: Why does my Intercom dashboard show a different number? The most common causes are how the tag-applied timestamp is bucketed versus conversation creation date, multi-tag counting, and renamed or archived tags. Reconcile in Reports → Tags over a matching range.

Q: My ranking looks noisy — why? Inconsistent tagging is the usual culprit. If agents do not tag every conversation, or use overlapping tags, the ranking under-represents real demand. Tighten your tagging conventions to make this card trustworthy.

Q: Can I customise the alert threshold? Yes. The “+50% week over week on any tag” trigger is configurable per profile in the Alert Rules tab. Raise it if a few low-volume tags create noisy alerts, or lower it to catch emerging issues sooner.

---

Tracked live in Vortex IQ Nerve Centre

Top Topics (Tags) 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.