> For the complete documentation index, see [llms.txt](https://docs.satisfilabs.com/resource-center/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.satisfilabs.com/resource-center/ai-agent-engine/agent-performance-console/new-agent-performance-console/glossary.md).

# Glossary

### Weekly Briefing

<table><thead><tr><th width="165.60546875">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Conversation</td><td>A full back-and-forth exchange between a guest and an agent, measured as one unit.</td></tr><tr><td>Conversations Handled</td><td>The total number of conversations your agents managed in the selected week.</td></tr><tr><td>Purchase-Intent Conversations</td><td>Conversations where a guest showed a clear signal they wanted to buy (tickets, merch, parking, etc.).</td></tr><tr><td>Conversational Intelligence High/Low</td><td>Insights that are generated by a prompt that reviews data from all the conversational intelligence reports.</td></tr></tbody></table>

### Campaign Overview

<table><thead><tr><th width="167.55078125">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Campaign</td><td>A structured, AI-driven mission assigned to a single agent. It aligns that agent's behavior with a strategic business outcome, and it is defined by what the guest is trying to accomplish rather than by topic.</td></tr><tr><td>Campaigns At or Above Goal</td><td>A fraction showing how many scored campaigns are hitting their target, with a breakdown of scored vs. not-yet-scored campaigns.</td></tr><tr><td>Campaigns Running</td><td>The total number of campaigns active across your agents.</td></tr><tr><td>KR Score</td><td>A decimal score (against a 0.7 goal) produced in two steps: first, the campaign's task completion rate is measured; second, that rate is scored against the campaign's goal. 0.7 marks performance on par with a strong peer campaign, the completion rate needed to earn it differs by campaign difficulty.</td></tr><tr><td>Volume Pending</td><td>A campaign has 0 conversations attempted yet (not enough activity to score).</td></tr></tbody></table>

<figure><img src="https://167344003-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FsxN9AGLZMIc5C8dx3zU7%2Fuploads%2FucajdFIus5e2xOMoRXdr%2Fimage.png?alt=media&amp;token=5a70b47e-6ad6-4094-9a17-732880d0d758" alt=""><figcaption></figcaption></figure>

### Campaign Page

<table><thead><tr><th width="171.15234375">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Cohort Median</td><td>The median score of peer accounts running the same campaign.</td></tr><tr><td>Completed</td><td>Conversations that achieved a specific goal within a campaign (click-throughs, resolved questions, tickets sold, etc.).</td></tr><tr><td>Conversations Attempted</td><td>Conversations that entered a given campaign in the selected time period.</td></tr><tr><td>Fully Resolved</td><td>The guest got what they came for, at the level they needed, within what the agent is built to handle.</td></tr><tr><td>Partially Resolved</td><td>The need was within the agent's remit and the agent moved the guest specifically toward it, but stopped short of the completeness or specificity the guest wanted.</td></tr></tbody></table>

### Agent Profile

<table><thead><tr><th width="169.10546875">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Objective Score</td><td>An agent's overall performance rating on a 0.0–1.0 scale, rolling up how well it's meeting its objectives across all its campaigns.</td></tr><tr><td>Objectives</td><td>The specific goals an agent is designed to achieve; its campaigns each ladder up to these.</td></tr></tbody></table>

### Performance Overview

<table><thead><tr><th width="170.1796875">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Active Conversations / Week</td><td>Weekly conversation volume over time, showing trend and seasonality.</td></tr><tr><td>Click-Outs</td><td>When a conversation leads the guest to click a call-to-action that takes them out of the chat.</td></tr><tr><td>Conversation Entry Points</td><td>Which chat surfaces guests started from (e.g., Tickets, InApp, Web), showing where demand originates.</td></tr><tr><td>Conversations Analyzed</td><td>The total conversations evaluated in the selected timeframe.</td></tr><tr><td>Partially Resolved</td><td>The need was within the agent's remit and the agent moved the guest specifically toward it, but stopped short of the completeness or specificity the guest wanted.</td></tr><tr><td>Resolved</td><td>The percentage of scored conversations where the agent fully completed the request.</td></tr><tr><td>Returning Guests</td><td>The percentage (and count) of guests who came back for more than one conversation in the selected timeframe.</td></tr><tr><td>Unresolved</td><td>The share of scored conversations the agent couldn't solve.</td></tr><tr><td>Where the Widget Appears</td><td>The specific pages/URLs the chat widget loaded on, showing reach across your properties.</td></tr></tbody></table>

### Revenue

<table><thead><tr><th width="167.17578125">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Acquisition</td><td>Conversations that won a new purchase. One of four Revenue Category classifications.</td></tr><tr><td>Buyers in the Data</td><td>Conversations that showed active purchase intent in the selected timeframe.</td></tr><tr><td>Conversation Mix by Revenue Category</td><td>How purchase-related conversations split across four categories: Acquisition, Protection, Erosion, Engagement.</td></tr><tr><td>Conversation Transitions (First → Last Revenue Category)</td><td>How conversations move between revenue categories from start to finish. Read left to right: of conversations that started in a category, what share ended in each.</td></tr><tr><td>Drift (Purchase → Elsewhere)</td><td>Of conversations that started with purchase intent, the share that ended without buying in chat.</td></tr><tr><td>Engagement</td><td>Purchase-adjacent conversation activity that isn't a direct sale. One of four Revenue Category classifications.</td></tr><tr><td>Erosion</td><td>Conversations trending toward cancellation or refund. One of four Revenue Category classifications.</td></tr><tr><td>In-Chat Ticketing</td><td>Actual ticket transactions confirmed directly through the ticketing vendor's own data feed.</td></tr><tr><td>Keeping Booked Revenue (Protection / Erosion)</td><td>Conversations that protect existing revenue (ticket access, refunds handled well) versus those signaling erosion (cancellation/refund requests).</td></tr><tr><td>Lift (Non-Revenue → Revenue)</td><td>The share of conversations that started as non-purchase but reached purchase intent, the agent creating demand that wasn't there at the start.</td></tr><tr><td>Protection</td><td>Conversations helping guests manage tickets they already own, keeping booked revenue intact. One of four Revenue Category classifications.</td></tr><tr><td>Top Purchase On-Ramps</td><td>Topics that most often appeared in conversations which led to purchase intent.</td></tr></tbody></table>

### Voice of Guest

<table><thead><tr><th width="165.73828125">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Emerging Themes</td><td>Clusters of related guest language rising in the data, shown as a bubble map (sized by volume) and a ranked list.</td></tr><tr><td>Top Topics</td><td>The most common conversation topics, ranked by share of total conversations in the timeframe.</td></tr></tbody></table>

### Resolution Opportunity

<table><thead><tr><th width="165.1796875">Term</th><th>Definition</th></tr></thead><tbody><tr><td>Alternative Recommendations</td><td>Cases where a guest asked for a specific item that wasn't available, and the agent offered a substitute instead of a dead end.</td></tr><tr><td>Content Gaps</td><td>Conversations grouped by topic where the agent's available content didn't fully satisfy what the guest was asking about.</td></tr><tr><td>What Guests Asked About</td><td>Within a content gap category, the specific topics or sub-topics guests used, ranked by frequency.</td></tr></tbody></table>