Text classification
Related terms
Auto-tagging
Auto-tagging is the automatic application of labels from a defined taxonomy to incoming tickets, calls, chats, or reviews, so volume can be counted, reported on, and routed without an agent choosing tags by hand. It relies on text classification as the underlying technique but is distinct from it: classification assigns a category, while auto-tagging operationalizes that output into the labels a support team actually works with day to day. Example: 4,000 weekly chats tagged by reason code let a support lead see that billing contacts rose week on week, a pattern OnClarity's Voice of Customer surfaces automatically rather than waiting for a monthly manual review. Common failure modes are tag sprawl, where overlapping or duplicate tags accumulate until reporting becomes unreliable, and a stale taxonomy, where categories no longer match current products or language and need periodic review to stay useful.
Topic clustering
Topic clustering groups conversations by theme to size the drivers behind contact volume, without requiring an analyst to define categories in advance. Because clusters emerge from what customers actually said, the method surfaces new issues that a fixed intent taxonomy would file under a generic other.
Ticket routing
Ticket routing is the set of rules and models that decide which queue, team, or agent should receive an incoming request, and where it lands in the priority order. Decisions typically draw on channel, classified intent, language, account tier, and the applicable SLA clock. OnClarity's omnichannel inbox applies automated routing across channels, and its AI agents hand conversations they can't resolve to a human with full context. For example, a payment failure reported by an enterprise account can route directly to a specialist queue with a tightened SLA, rather than sitting in the general inbox behind simpler requests. Routing quality is only as good as the classification feeding it: misread intent sends tickets to the wrong place regardless of how well the rules themselves are built. See also routing and skills-based routing.
