Support Analytics That Matter: Resolution Rate, Response Time & Staff Performance

“We get a lot of tickets” is not a metric — it is a feeling. To actually improve support you need numbers that answer specific questions: Are we keeping up? Is the AI pulling its weight? Where is staff time going? Here are the metrics worth watching.

Volume and closure rate

Start with the basics: how many tickets opened in a period, and what share got closed. Volume trends warn you about spikes (a launch, an outage); closure rate tells you whether you are keeping up or accumulating a backlog. Glydebot’s guild analytics break this down by subject with pie and bar charts, so you can see which categories drive your load.

AI vs staff resolution

This is the metric that justifies automation. What percentage of tickets did the AI resolve on first contact versus hand off to staff? A rising AI-resolution share means your knowledge base is doing its job. A subject that escalates almost every time is telling you an article is missing or a prompt needs tuning.

First response time

Members judge support by how fast they hear back. AI first contact makes the first response effectively instant for common questions; for escalations, first response time tells you whether your staffing and routing are healthy. Watch it by subject — a slow-responding category usually means too few people have access or the role mention is misconfigured.

Staff performance — fairly

On higher tiers, staff analytics show per-user responses, claims, closes, first response time, and rankings. Used well, this surfaces who needs help and where the load is uneven — not a leaderboard to shame people with. Pair it with RBAC so the right people own the right tickets in the first place.

AI usage and your daily cap

Automation has a budget. Track tokens, requests, and utilization against your daily cap so you are not surprised mid-month. If you are routinely near the cap, that is a signal to tighten prompts, trim the knowledge base, or move up a tier. Retention of analytics history varies by plan — see the billing docs for specifics.

Turn metrics into a weekly habit

• Which subject generated the most tickets? Should it have a panel of its own?
• Where did AI resolution drop? Write or update the matching article.
• Which subject is slowest to first response? Fix routing or access.

Explore the full breakdown in the analytics documentation, or see which tier includes staff analytics on the pricing page.