Usage Analytics

The Usage Analytics dashboard gives you a complete view of how your team is using AI — which features are consuming the most tokens, which users are most active, how costs break down by product, and whether you are on track to stay within your monthly quota.

Accessing Usage Analytics

Navigate to AI Platform → Usage Analytics in the left navigation.

Dashboard Views

Overview Tab

The Overview tab shows your key metrics at a glance:

Metric	Description
Tokens used	Total tokens consumed this month vs. your tier limit
AI cost this month	Total USD cost from all AI primitives
Requests today	Total API requests from all sources (Playground, Custom GPTs, product features, API)
Active users	Unique users who made at least one AI request this month
Top feature	The AI feature consuming the most tokens this month
Provider breakdown	Percentage of requests served by Azure OpenAI, Azure AI Foundry, and Anthropic

A daily usage chart spans the last 30 days, showing tokens and cost side by side.

By Feature Tab

Breaks down token consumption and cost by feature string:

Column	Description
Feature	The feature identifier (e.g., ticket-suggest, jarvis-assistant)
Requests	Total request count for the period
Tokens	Prompt tokens + completion tokens
Cost (USD)	Calculated from per-token pricing for the model that served the feature
Avg tokens/request	Helps identify features with unexpectedly large prompts
Provider	Primary provider serving this feature

Use this view to identify which products and features are driving your AI spend.

By User Tab

Shows per-user activity for your organization:

Column	Description
User	Name and email
Requests	Total requests initiated
Tokens	Total tokens consumed
Cost	Cost attributed to this user
Top feature	The feature this user invokes most
Last active	Most recent AI request timestamp

ℹ️Admins can see all users. Standard users can only see their own usage.

By Product Tab

Groups usage by The One product source:

• AI Platform (Playground, Custom GPTs, Jarvis)
• PSA (ticket-suggest, RCA generation)
• CRM (enrichment, activity summaries)
• Defend (behavioral-analysis, attack-narrative)
• Voice (call-recap, transcription)
• Books (invoice explanation)
• RMM (device summaries, patch analysis)

Each product shows total requests, tokens, and cost for the selected period.

Quota Tab

Shows your current quota status:

• Tokens used today vs. daily limit
• Tokens used this month vs. monthly limit
• Monthly spend vs. budget cap
• Days remaining in the current billing period
• Projected month-end usage (linear extrapolation)

The projection helps you anticipate whether you will need to upgrade before month-end.

Time Period Selection

All charts and tables support these time periods:

• Last 7 days
• Last 30 days (default)
• Last 3 months
• Last 12 months
• Custom date range

Exporting Data

Click Export (top right of any tab) to download a CSV of the current view. The CSV includes all columns visible in the table. Exports are scoped to your current filters and time period.

Metrics Reference

Token Types

Token type	Counted as
Chat prompt tokens	Input tokens (user message + system prompt)
Chat completion tokens	Output tokens (assistant response)
Embed tokens	Input tokens only (no output for embeddings)
Voice minutes	Separate from token quota — tracked in voice minutes
Phone minutes	Separate from token quota — tracked in phone minutes
Web pages crawled	Separate from token quota — tracked in page count
Emails sent	Separate from token quota — tracked in email count
Governance scans	Separate from token quota — tracked in scan count
Safety moderations	Separate from token quota — tracked in moderation count

Primitives Usage

Beyond token counting, the Primitives section of Usage Analytics tracks:

Primitive	Unit
ai.chat	Tokens used
ai.embed	Embeddings created
ai.image	Images generated
voice.room	Participant-minutes
phone.call	Call minutes
phone.sms	Messages sent
web.crawl	Pages crawled
email.send	Emails sent
governance.scan	Scans run
safety.moderate	Content checks

Alert Configuration

Set up alerts to avoid surprise quota overages:

1. Click Alerts in the Usage Analytics toolbar
2. Click New alert
3. Configure:
   • Metric — Token quota %, monthly spend $, daily request count, etc.
   • Threshold — Alert when metric exceeds this value (e.g., 80%)
   • Notification — Email, PSA ticket, or webhook
4. Click Save

Default alert: a warning at 80% of monthly token quota is pre-configured for all tenants.

API Access

Fetch usage data programmatically:

GET /api/usage/summary?period=30d
Authorization: Bearer oneai_your_key_here

GET /api/usage/by-feature?period=30d
Authorization: Bearer oneai_your_key_here

GET /api/usage/by-user?period=30d
Authorization: Bearer oneai_your_key_here

Responses follow standard pagination with limit and offset query params.