Integrate
...
Analytics and data
Fullstory connector tools
14 min
The Fullstory connector tools in Mark (Optimizely Agent Platform) let you interact with your Fullstory account to retrieve user session links, behavioral summaries, and work with Fullstory segments. After an administrator configures the Fullstory connector on the Connectors page, the tools become available in Mark chat, agents, and workflows.
Complete the following steps to enable Fullstory tools.
Configure the Fullstory connector
Generate a Fullstory API key
See the Managing API Keys documentation in the Fullstory help center for steps to generate API keys in Fullstory. Copy the API key as you need it in the following section.
Note
If you want to use the segment export tools, you need an "Admin" level API Key. See the Key Permissions section in the Fullstory help center.
Configure the connector on the Connectors page
When you click Configure, Mark opens the Configure Fullstory dialog. If the connector is not installed yet, an Install modal displays first. The dialog then has two steps: Settings, where you enter the details the connector needs to authenticate, and Tools, where you choose which tools to enable.
As a Mark administrator, complete the following steps:
- Go to Connectors.
- (Optional) Enter Fullstory in the Enter name field, or use the Provider, Active Status, and Provider Type filters to find the connector.
- Click Configure on the Fullstory tile.
Note The Install modal only displays if the connector is not installed yet. Installing adds the connector to your organization and makes its tools available. You install a connector once, for everyone.
- On the Settings step, complete the following, and then click Next. The app stores these settings and applies them to everyone in your organization.
- Enter the Fullstory API key that you generated in the Generate a Fullstory API key section.
- (Optional) Toggle Enable AI-powered session analysis on if your Fullstory plan includes this capability and you want to use it in Mark.
- On the Tools step, select the tools that users can use by default, and then click Save. Each tool displays its type, such as Read, Create, Update, or Delete. Use the search field to find a tool, or click Deselect all to clear the selection. Tools that you do not select still display on the connector's Tools tab, but remain disabled until you select them.
Note
You may not see all of the Fullstory connector tools in Mark if you do not have access to AI-powered session analysis.
After you configure the connector, the Fullstory tile displays Connect for you and for every Mark user in your organization. Click Connect to connect it.
Considerations
Fullstory session analysis may take some time, depending on the session length and associated metadata.
Fullstory connector tools
After you add and register the Fullstory connector tools, you can call the following tools in Mark. Click on a tool's name to expand it and learn more about when to use the tool, required and optional parameters, and example prompts on how to call the tool. If you do not provide a required parameter, Mark prompts you for it.
Session exports and segment operations
You can use the fullstory_create_segment_export tool in the following example use cases:
Export all user data
- When to use
- When you need a complete list of all users in your FullStory account.
- To get basic user profile information (like email, display name, first or last seen) for CRM updates or general user analysis.
- To understand the overall size and composition of your user base over a specific period.
- Parameters
- type – TYPE_INDIVIDUAL. To get user information.
- timeRange – A start and end date for the export. For example, {"start": "2024-01-01T00:00:00Z", "end": "2024-01-31T23:59:59Z"})
- format – FORMAT_CSV. For spreadsheet-friendly data.
- segmentId – "everyone". To include all users.
- Example prompts
- I need to export all user profiles from FullStory for the last month in CSV format.
- Can you give me a CSV export of all users, including their email and display name, for the past 90 days?
- Export all individual user data for the current quarter as a CSV file.
Export specific events for a custom segment
- When to use
- When you want to analyze specific user actions (events) performed by a particular group of users (segment).
- To understand how a defined user segment interacts with a specific feature or completes a critical flow.
- For deeper behavioral analysis of a target audience, such as "High-Value Customers" or "Churn Risks."
- Parameters
- type – TYPE_EVENT. To export event data.
- timeRange – The start and end date for the export.
- format – FORMAT_JSON. For structured data that can be easily parsed by other systems or scripts.
- segmentId – The ID of your custom segment. For example, "high-value-customers".
- (Optional) – fields – A comma-separated list of specific event fields you want to include. For example, "event_name,user_id,product_id,event_properties.price".
- Example prompts
- Export all 'Product Added to Cart' events for my 'Frequent Shoppers' segment from the last 60 days. I need it in JSON format and please include the product name and price.
- Can you get me a JSON export of 'Login Success' events for the 'Admin Users' segment from last week, showing the user ID and login method?
- I need a JSON export of 'Form Submission' events for the 'Trial Users' segment from the last 30 days, including the form name and submission status.
Export all session data with time zone adjustment
- When to use
- When you need detailed session-level information for all users, especially for large datasets that might be processed in a data lake or advanced analytics platform.
- To ensure that all timestamps in your export are aligned with a specific geographic time zone for accurate reporting or analysis across different regions.
- For comprehensive session replay analysis or to build custom session-level metrics.
- Parameters
- type – TYPE_SESSIONS. To export session data.
- timeRange – The start and end date for the export.
- format – FORMAT_NDJSON. Newline-delimited JSON, ideal for streaming and processing very large datasets efficiently.
- segmentId – "everyone". To include all sessions from all users.
- (Optional) timezone – A valid IANA time zone string. For example, "America/New_York" or "Europe/London" to adjust timestamps in the export.
- Example prompts
- Export all session data for everyone from the last week in NDJSON format, and please adjust the timestamps to 'America/Los_Angeles' time zone.
- I need an NDJSON export of all sessions from the last quarter, adjusted to 'Asia/Tokyo' time zone.
- Can you export all user sessions from the past 30 days as NDJSON, with timestamps in 'UTC'?
Export page-level events for a specific segment
- When to use
- To understand all events that occurred on specific pages within a segment's sessions.
- To analyze user journeys and interactions on critical pages (for example, product pages or checkout pages) for a targeted segment.
- For optimizing page performance or identifying friction points on key conversion paths.
- Parameters
- type – TYPE_PAGES. To export page-level event data.
- timeRange – The start and end date for the export.
- format – FORMAT_CSV. For easy review of page URLs and associated events.
- segmentId – The ID of your custom segment. For example, "blog-readers".
- (Optional) fields – A comma-separated list of specific page-related fields you want to include. For example, "url,referrer,page_title".
- Example prompts
- I want to get all page-related events for users in my 'Blog Readers' segment from the last 30 days. Please provide this as a CSV file, including the page URL and referrer.
- Export page event data for the 'Checkout Abandoners' segment from last month in CSV, showing the page URL and any error messages.
- Can you give me a CSV export of page events for the 'Product Page Viewers' segment from the last 7 days, including the page title and URL?
- When to use
- Immediately after you have initiated a segment export using the fullstory_create_segment_export tool and received an operationId.
- To periodically check if a long-running export job has finished processing.
- When you need to retrieve the temporary download URL for a completed export.
- To verify if an export operation encountered any issues or failed.
- Parameters
- operationId – The unique identifier for the export operation that was returned when you created the export.
- Example prompts
- What is the status of my FullStory export with operation ID U0VBUkmIX1VYUE9SVFplcmRteUlPV3YyV3RnRVZCdxdSeV?
- Is the export job U0VBUkmIX1VYUE9SVFplcmRteUlPV3YyV3RnRVZCdxdSeV complete yet?
- Check the progress of operation U0VBUkmIX1VYUE9SVFplcmRteUlPV3YyV3RnRVZCdxdSeV and tell me if it is ready to download.
- When to use
- After a search export operation. For example, from a FullStory search API call, (not create_segment_export) has completed and you have received its unique id.
- When you need to access the actual exported data file from a FullStory search.
- To obtain the temporary URL to download the results of a previously initiated search export.
- Parameters
- id – The unique identifier of the search export. This ID is typically returned as a result of a SEARCH_EXPORT operation within FullStory.
- Example prompts
- Can I get the download link for search export ID U0VBUkmIX1VYUE9SVFplcmRteUlPV3YyV3?
- Please provide the results for my search export with ID U0VBUkmIX1VYU.
- I have a completed search export with ID U0VBUkmIX1VYUE9SVFplcmRteUlPV3YyV3RnRV. Give me the download URL.
- When to use
- When you have a segment ID (a custom one or a built-in one like "everyone"), and you need to retrieve more information about that specific segment.
- To verify the existence of a segment before using it in other operations. For instance, before exporting data from it.
- To understand the properties and metadata of a segment, such as its creation date or the user who created it.
- When a user asks for details about a segment by its name or ID.
- Parameters
- id – The unique identifier of the segment you want to retrieve details for. This can be a custom segment ID (for example, "12345abcdefg") or a built-in segment ID like "everyone".
- Example prompts
- Tell me about the 'everyone' segment.
- What are the details for segment ID 12345abcdefg?
- Can you provide information on the segment named 'High-Value Customers'?
- Show me the creation date and creator of segment my-new-segment-id.
- When to use
- When you need to discover all segments that exist in your FullStory account.
- To find the id of a segment that you want to use in other tools. For instance, the fullstory_create_segment_export tool or the fullstory_get_segment tool.
- To get an overview of your organization's segmentation structure.
- If you want to browse segments created by a specific team member.
- Parameters
- (Optional) limit – The maximum number of segments to return, between 1 and 100, defaults to 20. Useful for controlling the size of the response.
- (Optional) paginationToken – A token obtained from a previous fullstory_list_segments call to retrieve the next page of results.
- (Optional) creator – An email address to filter segments by the user who created them.
- Example prompts
- List all FullStory segments.
- Show me the segments created by [email protected].
- What are the top 10 segments in FullStory?
- Can you list all segments and paginate the results?
- I need to see all segments, but only 50 at a time.
Session summaries and behavioral analysis
- When to use
- For rapid triage or initial assessment of a user session to quickly understand its main points.
- During daily reviews of multiple sessions to get a concise overview.
- When you need a brief summary (two to three sentences) for reporting or sharing with stakeholders.
- To quickly identify if a session was successful, problematic, or inconclusive.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- Give me a quick summary of FullStory session 1234567890:0987654321.
- Summarize what happened in session 1234567890:0987654321.
- What is a brief overview of session 1234567890:0987654321?
- Can you provide a quick AI summary for session 1234567890:0987654321?
- When to use
- When you need a deep understanding of a user's behavior patterns, beyond a quick summary.
- To identify potential user personas (for example, "power user," "first-timer," or "explorer") within a session.
- For UX researchers and product teams looking to analyze feature usage, navigation styles, and engagement levels.
- To understand the likely goals and motivations behind a user's actions in a session.
- When investigating specific user journeys for optimization opportunities.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- Analyze the user behavior in FullStory session 1234567890:0987654321.
- Give me a detailed behavioral summary of session 1234567890:0987654321.
- What kind of user behavior patterns can you identify in FullStory session 1234567890:0987654321?
- Provide a behavioral analysis for session 1234567890:0987654321.
- I need to understand the user's intent and engagement in session 1234567890:0987654321. Can you provide a behavioral summary?
- When to use
- When investigating a specific customer support ticket to understand the user's experience and issues.
- To troubleshoot user problems by identifying technical errors, rage clicks, or repeated failed actions.
- When you need to understand complex customer issues in detail, including user intentions, frustrations, and the final outcome of the session.
- To get actionable recommendations for support teams on how to resolve the user's problem or follow up effectively.
- For analyzing sessions where users reported a problem or abandoned a critical flow.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- Give me a support summary for FullStory session 1234567890:0987654321.
- What were the user's issues in session 1234567890:0987654321 from a support perspective?
- Provide a detailed support analysis of session 1234567890:0987654321.
- I need to understand the problems encountered in session 1234567890:0987654321 for a support ticket.
- Can you summarize session 1234567890:0987654321 with a focus on support recommendations for our support team can take?
- When to use
- When debugging user-reported issues that might have a technical root cause.
- To identify technical barriers or performance problems encountered by a user during a session.
- For developers and QA engineers to pinpoint JavaScript errors, network failures, or page loading issues.
- When investigating a support ticket where the user experienced a bug or unexpected system behavior.
- To understand the technical health of a user's interaction with your application.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- Give me a technical summary of FullStory session 1234567890:0987654321.
- What technical issues occurred in session 1234567890:0987654321?
- Provide a detailed technical analysis of session 1234567890:0987654321.
- I need to see any console errors or network failures from session 1234567890:0987654321.
- Can you summarize the technical problems in session 1234567890:0987654321?
- When to use
- When you need to analyze a user's journey through a specific conversion funnel. For example, signup, checkout, and booking.
- To identify where users are dropping off or showing hesitation within a critical flow.
- For ecommerce optimization and conversion rate analysis to understand what prevents users from completing a goal.
- To pinpoint conversion blockers and friction points in a user's path.
- When investigating why a user did not complete a desired action.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- Summarize the conversion funnel for FullStory session 1234567890:0987654321.
- Can you analyze the user's conversion path in session 1234567890:0987654321?
- What were the drop-off points in the funnel for session 1234567890:0987654321?
- Provide a conversion funnel summary for session 1234567890:0987654321.
- I need to understand the user's journey through the checkout process in session 1234567890:0987654321.
- When to use
- To understand what a user was trying to achieve early in their session.
- For UX research to classify user motivations. For example, exploring, researching, buying, or seeking support.
- To inform personalization strategies by identifying user intent and tailoring experiences.
- To understand user motivations and goals from their initial actions and behavioral signals.
- When you need to categorize sessions based on user intent for further analysis or segmentation.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- What was the user's intent in FullStory session 1234567890:0987654321?
- Can you provide a user intent summary for session 1234567890:0987654321?
- Analyze session 1234567890:0987654321 to determine the user's goal.
- What was the likely intent of the user in FullStory session 1234567890:0987654321 based on their early actions?
- Give me an AI-generated summary of the user's intent for session 1234567890:0987654321.
- When to use
- When the standard summary types (quick, behavioral, support, technical, conversion, or intent) do not provide the specific insights you need.
- For specialized research questions that require a unique analytical focus or a custom output format.
- When you want to explicitly define the AI's role (for instance, "You are a product analyst") and the exact information it should extract.
- To perform deep dives into specific aspects of a user session that are not covered by pre-defined summaries.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- pre_prompt – The initial prompt that sets the analytical context and role for the AI.
- post_prompt – The output formatting prompt that defines what insights to extract and how to structure them.
- (Optional) – temperature – Controls the creativity and randomness of the AI analysis (a number between 0.0 and 1.0. The default is 0.4).
- Use lower values (0.1-0.3) for consistent, factual analysis
- Use higher values (0.5-0.8) for more creative insights.
- Example prompts
- I need a custom analysis of session 1234567890:0987654321. Act as a product manager. Focus on feature discoverability and usage. Provide a list of features interacted with and suggestions for improvement.
- For session 1234567890:0987654321, I want to understand how the user navigated the site. My pre-prompt is 'You are a navigation expert.' My post-prompt is 'List all pages visited, the entry and exit points, and any observed navigation patterns (for example, linear, exploratory, search-driven).'
- Please analyze session 1234567890:0987654321 with a temperature of 0.2. My pre-prompt is 'You are a technical support engineer looking for performance bottlenecks.' My post-prompt is 'Identify any slow loading elements, network requests with high latency, and JavaScript errors that impacted page performance. Provide timestamps for each issue.'
User sessions and replay links
- When to use
- When you need to find session replays for a specific user to troubleshoot an issue they reported.
- To get a list of recent user activity for customer support or behavioral analysis.
- When a user asks to see all or a certain number of sessions for a particular user, identified by their email or user ID.
- To access session replays directly to investigate user journeys or pain points.
- Parameters
Note
You must provide an uid or email. If both are provided, the tool returns a union of results for both parameters.
- uid – The user's ID as set through FS.identify in your application. For example, "user123".
- email – The user's email address as set through FS.identify or FS.setUserVars. For example, "[email protected]".
- (Optional) limit – Maximum number of sessions to return. Must be between 1 and 100, defaults to 20 if not specified.
- Example prompts
- List the last 10 sessions for user [email protected].
- Show me the sessions for user ID 12345.
- Can you find recent sessions for [email protected]?
- I need to see the last 5 sessions for user user_abc.
- What are all the sessions for [email protected]?
- When to use
- When you need a holistic view of a user's experience across several sessions, rather than just a single session.
- For customer support teams preparing for a call or investigating a user's history of issues and interactions.
- For UX researchers to identify recurring behavior patterns, common frustrations, and overall user journey trends for a specific user.
- To proactively identify struggling users or those with persistent technical problems.
- When you need a summary that integrates insights from multiple sessions, including user goals, frustration indicators, technical errors, and actionable recommendations.
- Parameters
Note
You must provide either an uid or email. You cannot provide both.
- uid – Your application's user ID for the user to analyze. For example, "user123".
- email – The email address of the user to analyze. For example, "[email protected]".
- (Optional) limit – Maximum number of recent sessions to analyze. Must be between 1 and 20, defaults to 5. Analyzing more sessions provides richer insights but takes longer.
- (Optional) llm – An object to configure the AI analysis with the following custom prompts:
- pre_prompt (string) – Context prompt to guide the analysis perspective. For example, "You are a customer success specialist analyzing user behavior.".
- post_prompt (string): Output format and focus guidance. For example, "Focus on identifying opportunities for user engagement and success.".
- temperature (number) – Analysis creativity level (0.0-1.0, default 0.2). Lower values for factual analysis, higher for creative insights.
- Example prompts
- Analyze the FullStory profile for [email protected].
- Give me a comprehensive UX insight report for user ID user123 based on their last 10 sessions.
- I need a support intelligence summary for [email protected]. Focus on identifying any recurring issues or pain points they've experienced across their sessions.
- Can you visit the profile of [email protected] and provide an analysis of their recent activity, specifically looking for patterns related to feature adoption and engagement?
- Provide a detailed analysis of user_id_456's recent activity. Act as a technical lead and identify any persistent technical issues or performance bottlenecks across their sessions, and suggest potential solutions.
Error and CMP detection
- When to use
- When a user reports issues related to consent banners, privacy preferences, or any unexpected behavior with your Consent Management Platform.
- For QA testing of your CMP implementation to ensure it's functioning correctly and not causing errors.
- To troubleshoot general error messages, visual failure indicators (like red banners or toast messages), console errors, or network failures within a specific user session.
- When performing systematic error monitoring, especially for privacy-related features or critical user flows where errors could impact compliance or user trust.
- To quickly identify the root cause of a reported error by examining a session for all types of technical and visual errors.
- Parameters
- sessionId – The unique identifier for the session to analyze. This ID consists of two parts, userId and sessionNumber, separated by a colon. For instance, "1234567890:0987654321".
- Example prompts
- Check session 1234567890:0987654321 for any CMP errors or general error messages.
- Were there any consent management platform issues detected in session 1234567890:0987654321?
- Analyze session 1234567890:0987654321 for error banners, console errors, or network failures.
- I need to see if there were any visual errors or technical problems in session 1234567890:0987654321.
- Provide an error detection report for session 1234567890:0987654321.
Note
If you use Opti ID, administrators can turn off generative AI in the Opti ID Admin Center. See Turn generative AI off across Optimizely applications.