Usage Reports on Hub Central

This article applies to:Available in these QorusDocs Editions:

Enterprise+, Proposal Pro, QPilot Essential

This guide covers the reporting available on the Usage page in Hub Central: the Summary, Adoption, and QPilot Reporting tabs. For every KPI, chart, and table it explains what the widget shows and exactly how the number is calculated — including the specific activity and event names QorusDocs records.

In this guide:

1. Accessing the Usage page
2. How the numbers are produced
   • Refresh behaviour
3. Summary tab
   • User Metrics
   • User Usage
   • Monthly Usage Trends
   • Overall Product Usage
   • Monthly QPilot Trends
   • QPilot Usage
   • Monthly Allocated Seats
   • Monthly Active Users
4. Adoption tab
   • Feature usage by Users
   • Pursuit Metrics
   • Usage by Pursuit Type
   • Usage by Pursuit Owner
5. QPilot Reporting tab
   • KPI cards — Total credits · Agent runs · Active users · Tokens
   • Agent Execution Costs
   • Credit Cost per Agent · Cost Distribution
   • Credit Cost by QPilot Area
   • Agent Breakdown
   • Per User Credit Usage
6. Appendix A: Activity → category mapping
   • The ten product categories (Summary tab)
   • Search & QPilot sub-splits
   • The seven per-user feature columns (Adoption tab)
7. Appendix B: QPilot area mapping

Accessing the Usage page

The Usage page lives in Hub Central, under Settings. To open it:

1. In the Hub Central left navigation, expand Settings.
2. Select Usage.
3. The Usage page opens on the Summary tab. Use the tabs at the top of the page to move between Summary, Adoption, and QPilot.

How the numbers are produced

Everything you see in the reports starts life as either an activity log entry (one row per user action, with an activity name such as Search or DownloadFile), a pursuit record, or an AI agent run:

1. Raw events — the activity log (activity name, event name, user, timestamp, pursuit link, client application), pursuit records, and AI agent run records (tokens, credits, agent, area).
2. Reporting refresh — runs on a recurring schedule (typically hourly) and aggregates raw events into daily reporting data; recent days are rebuilt on every run.
3. Report screens — the Usage page queries the daily reporting data. The “Updated …” stamp shows the last successful refresh.

Refresh behaviour

Summary tab

The Summary tab is an executive overview: seats, activity by category, trends, and QPilot usage. 

All activity figures count entries in the activity log, classified by activity name into the ten product categories listed in Appendix A.

User Metrics

Below is a table with definitions for each of the KPI cards above:

CSV export: the roster lists every active user with their display name, email, role, and last activity date (the timestamp of their most recent non-login activity). Users with no recorded activity appear with an empty last-activity date.

User Usage

What it shows: per user, the count of activities in each of the ten product categories (Admin, Design, Search, Preview, Content Engagement, Content Usage, Auto Answer, Assignments, Pursuits, QPilot), plus a total. The chart view shows the top-N users by total; the CSV always contains every user.

How it’s calculated: counts activity-log entries in the selected date range, classified into categories by activity name per Appendix A. The Groups filter restricts to members of the selected user groups (direct members plus members via synced identity groups).

Monthly Usage Trends

What it shows: activity counts per calendar month for each of the ten product categories, over the last 12 months (no date picker).

How it’s calculated: same classification as User Usage, grouped by month (UTC) instead of user.

Overall Product Usage

What it shows: total activity per category for the window. Two bars are stacked from finer splits: Search = Smart Search + Keyword Search; QPilot = Prompting (chat) + AI Smart Fields + Agent Assignments + Other. Percentages are each bar’s share of the grand total.

How it’s calculated: category counts per Appendix A. Smart Search = Search / ExecuteFavoriteSearch activities initiated from AI search (Ai.Search); Keyword Search = the same activities initiated any other way. QPilot named splits per Appendix A; QPilot “Other” = total QPilot category minus the three named splits.

Monthly QPilot Trends

What it shows: monthly counts of QPilot activity in four series: Prompting (chat), AI Smart Fields, Agent Assignments, and Other.

How it’s calculated: Prompting = events Ai.SendPrompt, Ai.StartChat, Ai.StopChat · AI Smart Fields = Ai.SmartFieldGenerated · Agent Assignments = Ai.AssignmentAgentRun · Other = all QPilot-category activity minus the three named series, per month.

QPilot Usage

What it shows: QPilot activity split by sub-category (Prompting, AI Smart Fields, Agent Assignments, Other), further filterable by the client application (“channel”: Hub Central, Add-in, Salesforce, Dynamics 365, External) and by whether the activity happened inside a pursuit.

How it’s calculated: sub-categories use the event names in the previous widget. “In pursuit” means the activity was linked to a pursuit at the time; “out of pursuit” is the remainder. Channel is the client application that logged the activity (unrecognised applications appear as “Unknown”). “Other” = total QPilot minus the named three, computed per channel and per pursuit split.

Monthly Allocated Seats

What it shows: allocated seats as a percentage of total licensed seats, per month.

How it’s calculated: for each month, the last available daily seat snapshot in that month supplies Total Seats and Allocated Seats; the percentage is shown to one decimal place.

Monthly Active Users

What it shows: distinct active users as a percentage of total licensed seats, per month.

How it’s calculated: numerator: distinct users with at least one counted activity in the month (an activity that maps to any reporting category — sign-ins don’t count). Denominator: Total Seats from that month’s last seat snapshot. Months with no snapshot are omitted.

Adoption tab

Feature usage per user, plus pursuit performance.

Feature usage by Users

What it shows: per user: Pursuits Created, Searches Performed, Documents Created, Assignment Activities, Insert Content, Download Content, QPilot, and Total (sum of the seven).

How it’s calculated: counts of activity-log entries in the window, by activity name:

• Pursuits Created — CreatePursuit
• Searches Performed — Search
• Documents Created — CreatePreview, WordCreate, PowerPointCreate, EmailCreate, DesignWordCreate, DesignPowerPointCreate, DesignEmailCreate
• Assignment Activities — CreatedAssignment, EditedAssignment, DeletedAssignment, UpdatedAssignmentStatus
• Insert Content — Insert
• Download Content — DownloadFile
• QPilot — events whose name starts with Ai

The Area facet uses the activity’s pursuit link: Pursuit only counts activities linked to a pursuit; Outside Pursuit is the base count minus the in-pursuit count (Pursuits Created shows 0 in this mode).

Pursuit Metrics

Below is a table with definitions for each of the KPI cards above:

Usage by Pursuit Type

What it shows: per pursuit type: Created, Closed, Win %, Avg Days to Close, Avg Users, Avg Activities, Created Value, Won Value (the CSV adds Closed Value).

How it’s calculated: same cohort rule and Win % rule as Pursuit Metrics, grouped by type. Created Value = sum of pursuit value over the created cohort. Won Value = sum of value over cohort pursuits marked won (not required to close in range). Closed Value (CSV) = sum of value over cohort pursuits closed in range. Avg Users = average distinct users acting on each pursuit (from pursuit-linked activities in range); Avg Activities = average pursuit-linked activity count. The With/Without Agents facet splits on whether the pursuit has any agent-enabled assignment.

Usage by Pursuit Owner

What it shows: the same measures grouped by pursuit owner (no Avg Users column).

How it’s calculated: identical rules to Usage by Pursuit Type, grouped by the pursuit’s owner.

QPilot Reporting tab

Credit and usage reporting for QPilot agents in your hub. All figures are credits and counts. The whole tab is driven by one filter bar (Dates, default last 90 days · Area · Groups); some widgets add their own overrides.

KPI cards — Total credits · Agent runs · Active users · Tokens

Below is a table with definitions for each of the KPI cards above:

The “% vs prev” deltas compare against the immediately preceding window of the same length (e.g. a 90-day range compares to the 90 days before it), rounded to whole percent.

Agent Execution Costs

What it shows: credits per week (Monday-start) or per calendar month, stacked by agent. The chart shows the top 9 agents by credits; everything else rolls into a grey “Other” stack. The CSV mirrors the chart (period × agent, with tokens and credits).

How it’s calculated: credits are bucketed by the UTC day the run completed, then rolled into weeks or months. All of a run’s usage — including sub-agent work — is attributed to the top-level agent that was invoked.

Credit Cost per Agent · Cost Distribution

What it shows: total credits per agent for the window (bars, scaled to the top agent) and each agent’s share of total credits (donut; centre shows the total, legend shows whole-percent shares).

How it’s calculated: sum of credits per top-level agent (sub-agent usage rolled up). Share % = agent ÷ total, rounded.

Credit Cost by QPilot Area

What it shows: credits (and runs) split across four areas: QPilot Chat, QPilot Smartfields, Assignments, Other (MCP). Empty areas are hidden.

How it’s calculated: each run carries the area it was started from (see Appendix B). Credits are summed per area.

Agent Breakdown

What it shows: per agent: Runs, Users, Input tokens, Output tokens, Cost / run (credits per run, rounded), Total credits.

How it’s calculated: per top-level agent over the window: run count, distinct users, token sums, credit sum; Cost / run = credits ÷ runs.

Per User Credit Usage

What it shows: per user: Name, Role, Last AI activity (relative time), Runs, Credits used. Only users with agent usage matching the filters appear.

How it’s calculated: credits and runs summed per user over the window (attributed to the user who initiated the run). Last AI activity is the user’s most recent completed agent action — not limited to the selected window. Role is the user’s licence role name.

Appendix A: Activity → category mapping

The exact activity names recorded in the QorusDocs activity log behind every category used on the Summary and Adoption tabs. Rows shown as Event + name are generic event entries recorded with that specific event name. An activity can belong to more than one category (e.g. DownloadFile is both “Download Content” and part of “Content Usage”) — each report uses a non-overlapping set, so nothing double-counts within a single widget.

The ten product categories (Summary tab)

Search & QPilot sub-splits

The seven per-user feature columns (Adoption tab)

Every column has an “in pursuit” twin (the same activities recorded while linked to a pursuit) — that’s what the Pursuit only / Outside Pursuit facet switches between.

Appendix B: QPilot area mapping

On the QPilot Reporting tab, each agent run is attributed to the area of QorusDocs it was started from. There are four buckets.