> ## Knowledge Base Index > Fetch the complete knowledge base index at: https://help.crisp.chat/sitemap.xml > Use this file to discover available pages before exploring further. > Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash). # Getting started with the Crisp Analytics *Use Crisp Analytics to understand your support activity, team performance, customer behavior, knowledge base usage, campaigns, and automation impact.* [Analytics](https://crisp.chat/en/analytics/) gives you prebuilt reports for common support questions and lets you create custom Dashboards when your team needs a more focused view. Open [**Crisp**](https://app.crisp.chat), then go to **Analytics** from the left navigation bar to review your reports. || Crisp Analytics is available from the **Essentials** plan. You can learn more in [Getting started with the Crisp Essentials plan](https://help.crisp.chat/en/article/getting-started-with-the-essentials-plan-1t7ptn1/). __This guide covers:__ * [Getting started](#1-getting-started) → where Analytics lives and how reports are organized * [Default reports](#1-default-reports) → the main report categories available in Crisp * [Customize Analytics reports](#1-customize-analytics-reports) → preferences, actions, configuration, filters, and aggregation methods * [How metrics are calculated](#1-how-metrics-are-calculated) → the most important calculation rules to know * [Frequently Asked Questions](#1-frequently-asked-questions) → common Analytics questions and edge cases --- # ${color}[#0080dd](Getting started) Analytics is split into categories such as **Messaging**, **Visitors**, **Knowledge Base**, **Contacts**, **Status Page**, and **Dashboard**. Each category contains reports that help you monitor a specific part of your workspace activity. ![Introduction to Crisp Analytics](https://storage.crisp.chat/users/helpdesk/website/bacc0702a039b000/analytics-intro_8lkrr5.png =1000xauto) Most reports are already configured with useful default metrics, time periods, and aggregation methods. You can scroll through them directly, export report data when needed, or customize charts with filters, office hours, date splits, and aggregation options. The quickest way to start is to review the default reports first, then build a custom **Dashboard** once you know which metrics your team checks regularly. --- # ${color}[#0080dd](Default reports) Default reports are prebuilt views available in Analytics. From the top bar, you can usually select the **time period** and **timezone** used to generate the data. Many reports can also be exported as CSV from the report action menu. #### ${color}[#445055](Messaging) The **Messaging** category focuses on Inbox activity, response performance, ratings, operators, automation, channels, SLAs, segments, shortcuts, and message volume. ![Messaging Analytics preview](https://storage.crisp.chat/users/helpdesk/website/bacc0702a039b000/analytics-messaging_1ffo6au.png =1000xauto) ###### ${color}[#F08820](Overview) The Overview section gives you quick access to the most common Inbox metrics. ###### ${color}[#F08820](Conversations) Total active conversations created or reopened during the selected period. ###### ${color}[#F08820](Visitors) Unique visitors detected on pages where the chatbox is installed. ###### ${color}[#F08820](Conversations over Time) Evolution of conversation volume across the selected period. ###### ${color}[#F08820](Conversations per Period) Heatmap of conversation volume by time period. ###### ${color}[#F08820](Conversations per Operator) Treemap of conversations handled by each operator. ###### ${color}[#F08820](Conversations per Response Origin) First response origin, such as human operator, automation, AI, or auto-responder. || Deleted conversations may still be counted in some reports, while conversations marked as spam are not included. ###### ${color}[#F08820](Operator Messages) The **Messaging → Overview** section also includes reports for tracking operator message volume. __Available views include:__ * **Operator Messages** → total operator messages across the selected period * **Messages per Operator** → compare the total number of messages sent by each operator * **Messages per Operator over Time** → compare operator message activity by hour, day, week, or another selected date split You can change the chart type or date split to view the same message data differently. This lets you compare operator message activity or follow how message volume changes over the selected period. In compatible custom charts, the **Message** metric can also be split by **Message Type** to compare different types of messages. || Message volume measures activity. Use it alongside metrics such as conversations handled, response time, resolution time, and ratings when reviewing operator performance. ###### ${color}[#F08820](Responsiveness) Responsiveness reports help you measure how quickly conversations are handled. ###### ${color}[#F08820](Response Time) Average delay between the first user message and the first response, including automations and human agents. ###### ${color}[#F08820](Human First Response Time) Average first response time for conversations where the first response comes from a human operator rather than an automation. ###### ${color}[#F08820](AI First Response Time) Average first response time for conversations where the first response comes from AI. ###### ${color}[#F08820](Time to Handle) Time between an operator’s first and last message before resolution. ###### ${color}[#F08820](Response Time per Period) Heatmap showing response time patterns across days or hours. **Time to Handle — During Office Hours** calculates handle time with office hours applied. || Resolving a conversation without any reply does not create a response time value, because no response was sent to the user. #### ${color}[#445055](Ratings) Ratings reports help you monitor customer feedback. ###### ${color}[#F08820](Overall Rating) Average rating across all rated conversations. ###### ${color}[#F08820](Automation Rating) Average rating for conversations handled only by automations or AI. ###### ${color}[#F08820](Ratings Evolution) Rating trend over time. ###### ${color}[#F08820](Ratings) Detailed report with the visitor, note, comment, assigned agent, and conversation link. If a user leaves a new rating for the same conversation, the previous one is updated instead of counted as a separate rating. ###### ${color}[#F08820](Operators) The **Operators** report shows individual performance for each operator, including conversations handled, average response time, average resolution time, and average rating on conversations assigned to that operator. ###### ${color}[#F08820](Automation) Automation reports help you measure how much work is handled before humans step in. ###### ${color}[#F08820](Automated Conversations) Conversations first replied to by an automation or AI. ###### ${color}[#F08820](Fully Automated Conversations) Conversations resolved without human intervention. **Automated Conversations over Time** shows the evolution of conversations first handled by automation. Use these reports to understand whether Hugo, workflows, and other automations are reducing repetitive work. ###### ${color}[#F08820](Channels) Channels reports show where conversations come from. ###### ${color}[#F08820](Chat Conversations) Conversations initiated from the chatbox. ###### ${color}[#F08820](Email Conversations) Conversations initiated by email, including outbound emails created from the Inbox. ###### ${color}[#F08820](Others) Conversations from other connected channels such as WhatsApp, forms, or social messaging. ###### ${color}[#F08820](Most Used Messaging Channels) Comparison of conversation volume by channel. ###### ${color}[#F08820](SLAs) SLA reports help you monitor conversations that exceed a response-time threshold. ###### ${color}[#F08820](Conversations SLA Above 1 Hour) This default report focuses on conversations with a response time above **1 hour**. The filter can be adjusted to track the threshold that matters to your team. ###### ${color}[#F08820](Conversations SLA Above 1 Hour During Office Hours) This report applies **Office Hours** to exclude periods when your team is not expected to reply. ###### ${color}[#F08820](Segments) Segments reports show how often conversation segments are used. ###### ${color}[#F08820](Most Used Segments) Number of times each segment was added to conversations. ###### ${color}[#F08820](Conversations Per Segment) Number of conversations associated with each segment. ###### ${color}[#F08820](Segments Used Over Time) Segment usage trends across the selected period. Segments are counted when they are added to a conversation, not simply when a conversation already contains them. In compatible custom charts, the **Conversation** metric can also be split by **Segments** to compare conversation volume between groups. ###### ${color}[#F08820](Shortcuts) Shortcuts reports show how agents use saved replies. ###### ${color}[#F08820](Most Used Shortcuts) Number of times each shortcut was selected by agents. ###### ${color}[#F08820](Shortcuts Used Over Time) Shortcut usage trends across the selected period. || A shortcut is counted when it is selected, even if the agent edits the reply or does not send it. It can also be counted several times in the same conversation if used several times. #### ${color}[#445055](Visitors) The **Visitors** category helps you monitor website visits and trigger activity on pages where the Crisp chatbox is installed. ![Visitor Analytics preview](https://storage.crisp.chat/users/helpdesk/website/bacc0702a039b000/analytics-visitors_uvi3tp.png =1000xauto) #### ${color}[#445055](Website Visits) __Website visit reports include:__ * **Visits Yesterday** → visits to your website on the previous day * **Visits Today** → visits to your website on the current day ###### ${color}[#F08820](Visitors per Country) Geographical heatmap based on detected visitor location. ###### ${color}[#F08820](Daily Visits) Evolution chart of daily website visits. ||| Crisp does not directly track user location. Country data relies on public IP registries published by internet service providers. #### ${color}[#445055](Triggers) __Trigger reports include:__ * **Triggers Yesterday** → triggers fired on the previous day * **Triggers Today** → triggers fired on the current day ###### ${color}[#F08820](Daily Triggers) Evolution chart of trigger usage over time. #### ${color}[#445055](Knowledge Base) The **Knowledge Base** category helps you understand how visitors use your articles, which locales are viewed most, what users search for, and which articles receive feedback. ![Knowledge Base Analytics preview](https://storage.crisp.chat/users/helpdesk/website/bacc0702a039b000/analytics-helpdesk_12ck0el.png =1000xauto) #### ${color}[#445055](Visits) The **Visits** summary shows the total number of visits to your Knowledge Base during the selected period. #### ${color}[#445055](Knowledge Base Visits) __Knowledge Base visit reports include:__ * **Visits Yesterday** → visits to your Knowledge Base on the previous day * **Visits Today** → visits to your Knowledge Base on the current day ###### ${color}[#F08820](Knowledge Base Visits) Visit evolution across the selected period. ###### ${color}[#F08820](Visits per Knowledge Base Locale) Visits grouped by language or locale. #### ${color}[#445055](Search) __Search reports include:__ * **Searches Yesterday** → search queries performed on the previous day * **Searches Today** → search queries performed on the current day ###### ${color}[#F08820](Popular Searches) Most searched terms in your Knowledge Base search input. Use these reports to find content gaps, unclear article titles, and topics that users expect to solve through self-service. #### ${color}[#445055](Articles) The **Articles** report shows per-article data such as visits, reactions, usefulness, rating details, comments, and visit trends. Select the relevant Knowledge Base locale from the report controls before analyzing article performance. #### ${color}[#445055](Contacts) The **Contacts** category shows data about contact creation and campaign activity. ![Contacts Analytics preview](https://storage.crisp.chat/users/helpdesk/website/bacc0702a039b000/analytics-contacts_n2k5tk.png =1000xauto) #### ${color}[#445055](Contacts Saved) __Contact reports include:__ * **Contacts Saved Yesterday** → contacts created on the previous day * **Contacts Saved Today** → contacts created on the current day ###### ${color}[#F08820](Contacts Saved) Contact creation trend across the selected period. #### ${color}[#445055](Campaigns Reached) **Campaigns Reach** shows the number of recipients reached by automated or one-shot email campaigns during the selected period. #### ${color}[#445055](Campaigns Activity) The **Campaigns** report shows global campaign performance, including delivered, opened, bounced, clicked, and unsubscribed counts. || Email opens are indicative. They can be affected by email clients, privacy features, and tracking limitations. #### ${color}[#445055](Status Page) The **Status Page** category helps you monitor downtime for the nodes you track in Crisp. ###### ${color}[#F08820](Total Downtimes) Total duration during which monitored nodes were detected as down. ###### ${color}[#F08820](Downtimes) Individual downtime duration for each monitored node. #### ${color}[#445055](Dashboard) The **Dashboard** category lets you build custom reports from scratch. Dashboards are useful when you want a focused view that combines the metrics your team checks most often. Once a dashboard is selected, click **Add chart** to add a report. You can choose chart types such as **Summary**, **Chart**, **Articles**, **Map**, **Heat Map**, **Operators**, and **Rating**. ![Custom report imported in an Analytics Dashboard](https://i.imgur.com/O7ndoO3.gif =1000xauto) The **Message** metric is available under the **Messaging** category when creating a custom chart. This lets you build your own message volume reports and combine the metric with compatible filters, operator splits, date splits, and chart types. For example, you can create a chart showing: * Total operator messages over time * Message volume split by operator * Operator messages grouped hourly, daily, or weekly Dashboards can be configured with the same types of preferences, filters, date splits, and aggregation methods as default Analytics reports. || Dashboard preferences and some display settings are stored locally in your browser. Clearing browser storage or switching browser/device may reset local display preferences. Useful resources: [build a custom dashboard in Crisp Analytics](https://help.crisp.chat/en/article/how-to-build-a-custom-dashboard-in-crisp-analytics-77heh0/) and [import or export a Crisp Analytics dashboard](https://help.crisp.chat/en/article/how-to-import-export-a-dashboard-in-analytics-30678w/). --- # ${color}[#0080dd](Customize Analytics reports) Most reports can be adjusted from the Analytics interface. This lets you focus on a specific period, compare channels, isolate segments, or change how data is grouped. ![Navigating the Analytics interface](https://storage.crisp.chat/users/helpdesk/website/bacc0702a039b000/analytics-how-to_778sqj.png =1000xauto) #### ${color}[#445055](Preferences) Preferences apply across Analytics as you browse reports. Use **Time Period** to select the period you want to analyze, and **Timezone** to align reports with your local time. Some reports may not be available for every period depending on their purpose. #### ${color}[#445055](Actions) The report action menu lets you perform contextual actions such as **Add to Dashboard** or **Export Data**. CSV exports are useful when you want to run additional analysis outside Crisp. #### ${color}[#445055](Configuration) Configuration controls how an individual report is displayed and calculated. __Common configuration options include:__ * **Chart Type** → line, bar, treemap, or other supported visual formats depending on the metric * **Split By** → break a metric into separate values using a compatible dimension, such as operator, segment, conversation type, or message type * **Office Hours** → include only data generated during your configured business hours * **Date Split** → group data hourly, daily, weekly, monthly, yearly, or across all selected time * **Aggregation Type** → calculate values using average, moving average, median, minimum, maximum, or sum Not every option is compatible with every report. When a setting does not apply, it may be hidden or disabled. On compatible line charts, you can use the **eye icon** to show or hide values directly on the chart. Hovering over a point also displays the date range represented by the selected date split. #### ${color}[#445055](Aggregation example) For the dataset `50, 60, 70, 80, 90, 200, 100, 110, 120, 130`, the aggregation changes the way the report is summarized. __Here is how common aggregation types would read that dataset:__ * **Average** → total divided by the number of values: **101** * **Moving average** → rolling average across a chosen number of periods, useful to smooth short-term spikes * **Median** → middle value after ordering the dataset: **95** * **Minimum** → smallest value: **50** * **Maximum** → largest value: **200** * **Sum** → total value: **1010** #### ${color}[#445055](Filters) Filters isolate a subset of data before the report is generated. You can include or exclude conversations based on segments, country, rating, visitor origin, first or last operator, response time, handle time, message type, and other compatible attributes. Adding multiple values inside the same filter condition usually works as an **OR** condition. To apply an **AND** condition, add another filter condition. || Some filter types are not compatible with every metric or chart. --- # ${color}[#0080dd](How metrics are calculated) Metrics use different calculation rules depending on what they measure. Understanding these rules helps you interpret reports more accurately, especially when conversations are reopened, resolved, automated, or filtered. #### ${color}[#445055](Conversation) A conversation is counted when an actual message, file, image, or voice message is exchanged. Private notes, reminders, and system messages are not counted as conversation messages. ###### ${color}[#F08820](Total) **Total conversations can be counted again when a resolved conversation is reopened.** A reopened conversation can be counted again during the selected period, with a maximum of once per day. ###### ${color}[#F08820](Unique) **Unique conversations are counted once for the selected period.** If the same conversation is reopened later, it is not counted again in that unique total. ###### ${color}[#F08820](Response Time) **Response time is the delay between a user message and the next response sent to the user.** The response can come from a human agent, automation, AI, or another operator origin. If a conversation is resolved without a response, no response time is calculated. ###### ${color}[#F08820](Resolution Time) **Resolution time is the delay between the first user message and the moment the conversation is resolved.** If a conversation is reopened and resolved several times, Crisp can calculate resolution time again for that conversation. ###### ${color}[#F08820](Handle Time) **Handle time is the delay between the first operator message and the last operator message before resolution.** If only one operator message is sent before resolution, the first and last message are the same, so handle time is **0**. ###### ${color}[#F08820](Messages) **Messages count user and operator messages, including files and automation messages.** Private notes and system messages are excluded. ###### ${color}[#F08820](Visitor Messages) **Visitor Messages count messages sent by users.** This includes regular messages and files sent from the user side. ###### ${color}[#F08820](Operator Messages) **Operator Messages count messages sent by operators or operator origins.** You can use this metric to measure total message volume, split results by operator, or analyze operator messages over time using hourly, daily, weekly, or other compatible date splits. Filters can also be used to focus on specific operator origins, for example to isolate human messages or exclude automations. ###### ${color}[#F08820](Rating) **Rating is the note left by users on conversations.** Ratings are between **1** and **5** stars. Learn more in [how to monitor customer ratings in Crisp](https://help.crisp.chat/en/article/how-to-monitor-user-ratings-thanks-to-crisp-1jj36o3/). #### ${color}[#445055](Conversation Assigned) **Conversation Assigned counts assignments or reassignments to agents.** Assignments can happen manually or through routing rules, and each assignment event can be counted. #### ${color}[#445055](Conversation Segment) **Conversation Segment counts when a segment is added to a conversation.** It does not simply count conversations that already contain a segment. A segment can be counted again for the same conversation only if it is removed and added again. #### ${color}[#445055](Conversation Shortcut) **Conversation Shortcut counts when an agent selects a shortcut.** It can be counted even if the agent edits the text or does not send the message, and several shortcuts can be counted inside the same conversation. #### ${color}[#445055](Campaign Activity) **Campaign Activity counts individual email campaign messages sent.** It covers automated and one-shot campaigns. #### ${color}[#445055](Campaign Sent) **Campaign Sent shows overall email campaign performance.** It can include delivered, opened, bounced, clicked, and unsubscribed metrics. #### ${color}[#445055](Knowledge Base Read) **Knowledge Base Read counts article reads.** A visitor opening the same article multiple times during the same session is counted once. #### ${color}[#445055](Knowledge Base Search) **Knowledge Base Search shows frequently used search keywords.** Generic terms such as `the` or `how` may be excluded from the report. #### ${color}[#445055](People Created) **People Created counts contacts created in Crisp during the selected period.** Contacts may be created when an email address is set in a conversation, imported from CSV, synchronized from an integration, or pushed through an API. #### ${color}[#445055](Status Downtime) **Status Downtime measures how long monitored nodes were detected as down.** Learn more in [how to configure a Crisp Status Page](https://help.crisp.chat/en/article/how-to-use-status-page-yxwhd9/). #### ${color}[#445055](Visitor Trigger) **Visitor Trigger counts how many times triggers fired for visitors during the selected period.** A trigger fires once per visitor, which helps you review trigger usage and adjust your setup. Learn more in [how to configure Crisp Triggers](https://help.crisp.chat/en/article/how-to-use-triggers-pq0znd/). #### ${color}[#445055](Visitor Visit) **Visitor Visit counts unique visitors on pages where the chatbox is installed.** Visitors are identified through sessions, so opening multiple tabs or browsing multiple pages during the same session does not count the same visitor again. --- # ${color}[#0080dd](Frequently Asked Questions) *Still have questions which were not covered in this article? Here is a collection of the most frequently asked questions on this topic.* ###### ${color}[#F08820](How can I track which URLs users are contacting me from?) **Use segments or custom data to group conversations by page URL.** You can add a segment automatically when a visitor contacts you from a specific page, or use the [Crisp JavaScript SDK](https://docs.crisp.chat/guides/chatbox-sdks/web-sdk/dollar-crisp/) to set custom data based on the current page. Once the segment or custom data exists, you can monitor it in Analytics through compatible filters or use [custom Crisp Inbox filters](https://help.crisp.chat/en/article/how-can-i-create-custom-filters-for-my-conversations-19yv8eo/) to review matching conversations. ###### ${color}[#F08820](Can I track specific Crisp events in Google Analytics?) **Yes, but Crisp Analytics and Google Analytics are used for different things.** Crisp Analytics helps you review Inbox, conversation, operator, knowledge base, visitor, and campaign data inside Crisp. If you want to track specific events in Google Analytics, follow the guide to [monitor Crisp events in Google Analytics 4 with Google Tag Manager](https://help.crisp.chat/en/article/how-can-i-use-google-tag-manager-gtm-to-monitor-crisp-event-in-google-analytics-4-ga4-bre55i/). ###### ${color}[#F08820](Why did the Dashboard I created disappear?) **Some Dashboard display preferences are stored locally in your browser.** Clearing browser storage, switching browsers, or using another device can reset local Dashboard preferences or display settings. When you need to reuse a Dashboard reliably, export it as a **.json** file and import it again later. See [how to import or export a Crisp Analytics dashboard](https://help.crisp.chat/en/article/how-to-import-export-a-dashboard-in-analytics-30678w/). ###### ${color}[#F08820](How can I compare New and Existing conversations?) **Use the Conversation Type split or filter.** When configuring a report, use **Split by → Conversation Type** to compare **New** and **Existing** conversations in the same chart, or use **Conversation Type** as a filter to focus on only one of them. This is useful when you want to separate first-contact volume from reopened or recurring conversations. ###### ${color}[#F08820](Can Analytics export all conversation messages?) **No, Analytics is not a full conversation export tool.** Analytics can export report data and may include conversation IDs depending on the report, but it does not export full conversation messages and metadata. To export conversation content or metadata programmatically, use the [Crisp REST API](https://docs.crisp.chat/references/rest-api/v1/). || Want to go further? Read [Set up Crisp Analytics for Customer Support: 4 dashboard templates](https://help.crisp.chat/en/article/set-up-crisp-analytics-for-customer-support-4-dashboards-templates-1f02z4o/), [how to build a custom dashboard in Crisp Analytics](https://help.crisp.chat/en/article/how-to-build-a-custom-dashboard-in-crisp-analytics-77heh0/), or [how to import and export Crisp Analytics dashboards](https://help.crisp.chat/en/article/how-to-import-export-a-dashboard-in-analytics-30678w/).