[object Object]

A support team running on shared inboxes and a spreadsheet of escalations spends half its day doing handoff archaeology and the other half answering the same questions. Service Hub is the version where the data, the SLA, and the deflection content all live in one place. The setup is straightforward; the discipline of using it correctly is what separates teams that scale from teams that hire endlessly.

Tickets are the spine

Tickets are Service Hub’s primary object. Each ticket has a pipeline, an owner, a priority, an SLA timer, and a thread of conversations attached. Configure pipelines per support tier or per product line — a billing pipeline rarely shares stages with a technical-support pipeline.

Pipeline: Tier 1 Support
  1. New
  2. Waiting on customer
  3. In progress
  4. Resolved
  5. Closed

Tickets associate to contacts, companies, and deals so the agent sees account context without leaving the record.

Knowledge base that deflects, not decorates

A KB article only earns its keep when it appears at the moment of need. Gate articles by audience, surface them in the help widget by URL match, and report on which articles deflect tickets and which sit unread. Categories should map to your ticket subject taxonomy so reports tie back together.

Customer portal as a self-service surface

The customer portal is the gated home where users see their tickets, submit new ones, and read the KB. It removes “what’s the status?” emails entirely. Brand it with your CSS, restrict by domain, and link it from your in-app help menu so customers find it without thinking.

Portal config:
  domain: support.example.com
  auth: SSO via Google + email magic link
  visible objects: tickets owned by signed-in contact
  KB scope: published articles tagged "external"

SLAs that match the buyer, not the team

Configure SLAs per priority with separate first-response and resolution timers. Business hours calendars decide what counts as elapsed time — a Sydney customer should not start a 4-hour clock on a Boston Friday afternoon.

Priority: P1 critical
  First response: 1 hour, 24x7
  Resolution: 4 hours, 24x7
Priority: P2 standard
  First response: 4 hours, business hours
  Resolution: 1 business day

Escalations on breach should page a real human, not just write to a Slack channel nobody watches.

Conversation intelligence and coaching

Call recording, transcription, and keyword tracking surface coaching moments managers would otherwise miss. Define keyword sets per use case — “cancel,” “refund,” “competitor name” — and review hits weekly. The signal is most valuable for newer reps where pattern correction compounds.

Reporting that drives improvement

Build dashboards for: tickets-per-customer, time-to-first-response by team, KB deflection rate, and reopened ticket rate. Reopened tickets are the truest signal of a poor resolution and the best place to focus quality work.

What to do this week

Audit your ticket pipeline against current customer reality, set both first-response and resolution SLAs per priority, and confirm your top ten KB articles still match the questions actually arriving in the inbox.

[object Object]
Share