Manage / Analytics & usage

Analytics & telemetry

Gain deep operational visibility into your AI assistant traffic. Track conversation volume, recurring visitor queries, token consumption costs, latency histograms, and monthly plan limits.

WebChat AI Analytics Dashboard showing KPIs, message trends, popular questions, and token usage — WebChat AI Analytics dashboard overview displaying message volume trends, response times, and query metrics.

Core KPI telemetry

Headline metrics tracked across your entire assistant fleet or scoped per assistant.

The top of the Analytics view displays five critical operational health indicators, comparing the selected date range (7d, 30d, or 90d) against the preceding period:

Total Messages

Message Volume

Total inbound questions from visitors and outbound answers generated by your assistants.

Conversations

Unique Sessions

Distinct visitor conversation threads initiated through the embedded widget or test harness.

Response Time

Average Latency

Mean duration in seconds across retrieval, prompt assembly, and token streaming.

Token Cost

Estimated Usage

Calculated consumption of input (prompt) and output (completion) tokens across model providers.

Visitor Feedback

Satisfaction Rating

Percentage of positive ratings (thumbs up) and average 1-5 star user sentiment scores.

Popular & recurring questions

Identify what visitors ask most frequently and detect content gaps.

WebChat AI aggregates incoming visitor questions by normalized question text, ranking top inquiries by ask frequency and satisfaction rating. The Popular Questions table reveals:

Frequency Rank: The exact number of times visitors have asked the question during the selected period.
Question Content: The normalized, trimmed visitor query text.
Average Satisfaction: Visitor feedback scores on answers provided for this inquiry, highlighting areas where knowledge may need expansion.

Optimizing Documentation from Analytics

If a specific question ranks in your top 5 queries with a lower satisfaction score, it typically indicates your source website or documents lack a direct explanation. Uploading an FAQ document or updating that web page immediately resolves the gap.

Quotas & monthly limits

Understanding plan consumption, rate limits, and billing tiers.

Each subscription tier enforces predefined monthly quotas to protect system stability and predictable operating costs:

Plan limits and quotas
Plan Tier	Chatbots	Crawled Pages	Monthly Messages	File Uploads
Free	1	20 URLs	50 / month	10 MB / batch (max 5)
Starter	3	200 URLs	2,000 / month	10 MB / batch (max 5)
Pro	10	1,000 URLs	10,000 / month	10 MB / batch (max 5)
Enterprise	Custom	Custom	Custom	Custom limits

When an assistant reaches its monthly query limit, the widget returns 429 MESSAGE_LIMIT_REACHED with a polite notice to the visitor. Account owners can upgrade anytime in the Billing panel to instantly restore query capacity.