Data Export - botBrains Docs

Your conversation data is valuable for understanding customer needs, meeting compliance requirements, and performing custom analysis. botBrains provides flexible export options to download conversations, messages, and user data in multiple formats for use outside the platform.

Why Data Export Matters

Exporting your data unlocks capabilities beyond the built-in analytics dashboard:

Compliance and auditing - Meet GDPR, CCPA, and internal data governance requirements
Custom analytics - Analyze data with your preferred tools (Python, R, Excel, Tableau)
Backup and archival - Maintain historical records for regulatory or business purposes
Cross-platform insights - Combine conversation data with CRM, product analytics, or business intelligence tools
AI-powered analysis - Use tools like ChatGPT or Claude to ask custom questions about your data
Report generation - Create executive reports or team dashboards with exported data

All exports respect your current permissions. You can only export data from projects you have access to.

Export Types

botBrains offers two complementary approaches to data export, each optimized for different use cases.

Full Dataset Export

The Full Dataset Export provides a complete snapshot of all your project data in a single asynchronous operation. When to use:

Complete data backups
Comprehensive analysis requiring all historical data
AI-powered analysis with tools like ChatGPT or Claude
Data migration to other systems
Compliance audits requiring full records

How it works:

Request export with optional date range filter
botBrains processes your entire dataset in the background
Receive email notification when export is ready (typically 5-15 minutes)
Download JSONL file via link in email
File remains available for 7 days

Format: JSONL (JSON Lines) - Each line contains a complete user object with nested conversations and messages. Limitations:

Limited to 2 exports per day per project (across all team members)
Processing time varies with dataset size (typically 5-15 minutes)
Downloads available for 7 days after generation

Use Full Export when you need complete data relationships preserved. The nested JSONL format maintains the natural hierarchy: Users → Conversations → Messages.

Entity-Specific Export

Entity exports provide immediate downloads of specific data types with customizable fields and formats. When to use:

Quick exports of filtered data
Spreadsheet analysis in Excel or Google Sheets
Targeted reports (e.g., “all escalated conversations this month”)
Specific field subsets to reduce file size
Sharing sanitized data with external stakeholders

How it works:

Navigate to entity list page (Conversations, Messages, or Users)
Apply filters and search criteria
Click Export button
Select format and fields
Download immediately

Available entities:

Conversations - Metadata, status, ratings, and nested messages
Messages - Individual messages with text content, tags, and analytics
Users - Contact information and engagement metrics (no nested conversations)

Formats:

CSV - Best for Excel, Google Sheets, or simple analysis
JSONL - Ideal for programmatic processing or data pipelines
Excel (CSV) - CSV optimized for Excel compatibility

Limitations:

Maximum 10,000 rows per export
Exports respect active filters from the list page

For datasets larger than 10,000 rows, use multiple filtered exports or the Full Dataset Export.

Choosing the Right Export Method

Scenario	Recommended Method	Format
Complete backup of all data	Full Export	JSONL
Analysis in Excel or Google Sheets	Entity Export	CSV or Excel
Filtered conversation subset	Entity Export	CSV or JSONL
AI-powered custom analysis	Full Export	JSONL
GDPR data request	Full Export	JSONL
Weekly metrics report	Entity Export	CSV
Import to analytics platform	Full Export or Entity Export	JSONL
Share with non-technical stakeholders	Entity Export	Excel

Performing a Full Dataset Export

Prerequisites

Before requesting a full export:

Ensure you have sufficient project permissions (Admin or Owner role)
Check that you haven’t already requested 2 exports today
Verify your email address is correct to receive the download link

Step-by-Step Instructions

Step 1: Navigate to Data Export Go to Settings → Data Export in your project. Step 2: Configure Export Options In the Full Dataset Export section, click Full Export. Optional: Set Date Range

Leave empty to export all historical data
Set “From” date to export data created after a specific date
Set “To” date to export data created before a specific date
Use date filters to create incremental backups

Example date range scenarios:

All data: Leave both fields empty
Last 30 days: From = 30 days ago, To = today
Q1 2024: From = Jan 1 2024, To = Mar 31 2024
Since last backup: From = previous backup date, To = empty

Step 3: Start Export Job Click Start Export Job. You’ll see a confirmation with your task ID. Step 4: Wait for Email Processing typically takes 5-15 minutes depending on data volume. You’ll receive an email titled “Your botBrains Data Export is Ready” with:

Task ID for reference
Download link (valid for 7 days)
File size information
Expiration date

Step 5: Download File Click the download link in the email. The JSONL file will download to your computer.

Download links expire after 7 days. If you need the data later, request a new export. Downloads cannot be extended or regenerated after expiration.

Rate Limits and Scheduling

Full exports are limited to 2 per day per project across all team members. This prevents system overload while allowing regular backups. Best practices for scheduled exports:

Create a dedicated export schedule (e.g., weekly on Sundays)
Coordinate with team members to avoid conflicts
Use date ranges for incremental exports instead of full exports
Archive downloads systematically with clear naming: botbrains_export_2024-03-15.jsonl

If you hit the rate limit: You’ll receive an error message indicating when you can retry (typically the next calendar day in your timezone). The error includes the exact retry time.

Performing Entity-Specific Exports

Exporting Conversations

Step 1: Navigate to Conversations Go to Analyze → Conversations. Step 2: Apply Filters Use the filter panel to narrow results:

Date range - Conversations created within specific dates
Status - Resolved, Unresolved, Escalated
Rating - Filter by CSAT score (1-5)
Topic - Conversations about specific topics
Channel - Web, Zendesk, Salesforce, Slack, etc.
Labels - Custom conversation labels
Search - Text search across conversation content

Example filter combinations:

Poorly-rated escalations: Status = Escalated, Rating = 1-2
This month's resolved chats: Date = Last 30 days, Status = Resolved, Channel = Web
Premium customer feedback: Label = Premium, Rating = 1-5

Step 3: Open Export Dialog Click the Export button in the top-right corner of the conversation list. Step 4: Choose Format Select your preferred export format:

CSV - Opens in Excel, Google Sheets, or text editors
JSONL - One JSON object per line, easy to process programmatically
Excel (CSV) - CSV with Excel-specific formatting

Step 5: Select Fields (Optional) By default, all available fields are exported. To reduce file size or focus on specific data:

Click Clear to deselect all fields
Check only the fields you need

Common conversation fields:

id - Unique conversation identifier
created_at - When conversation started
updated_at - Last message timestamp
status - Resolved, Unresolved, Escalated
user_id - Associated user identifier
rating - CSAT score (0-5)
channel - Communication channel
messages - Nested array of all messages (CSV exports messages as separate rows)

Leave all fields selected for your first export. Once you understand the data structure, create focused exports with specific fields to optimize file size.

Step 6: Export Data Click Export. The file downloads immediately to your browser’s download folder.

Exporting Messages

Messages export individual messages as separate rows, useful for text analysis, sentiment tracking, or message-level reporting. Step 1: Navigate to Messages Go to Analyze → Message Search (or use the Messages tab). Step 2: Apply Filters Filter messages by:

Sender type - User, Assistant, Operator
Content - Text search
Date range - Message creation dates
Sentiment - Positive, Neutral, Negative
Topics - Associated conversation topics
Tool usage - Messages that used specific tools

Step 3: Export Follow the same export process as conversations (format selection, field selection, download). Common message fields:

id - Unique message identifier
text - Message content
role - user, assistant, operator
conversation_id - Parent conversation
created_at - Message timestamp
sentiment - Detected emotional tone
topics - Detected or assigned topics

Exporting Users

User exports provide contact information and engagement metrics without nested conversations. Step 1: Navigate to Users Go to Analyze → Users (or Users tab). Step 2: Apply Filters Filter users by:

Activity date - Last interaction date
Conversation count - Users with X or more conversations
Email status - Has email, no email
Custom attributes - Any external attributes you’ve set via SDK

Step 3: Export Follow the same export process. Common user fields:

id - Unique user identifier
email - User email address
name - User display name
created_at - First seen date
last_seen_at - Most recent activity
conversation_count - Total conversations
message_count - Total messages sent
external_attributes - Custom data from your application

User exports do NOT include nested conversations or messages. Use Conversation export or Full Export to get complete user interaction history.

Understanding Export Formats

CSV (Comma-Separated Values)

Best for: Excel, Google Sheets, quick manual analysis Structure:

id,created_at,status,rating,user_id,channel
123,2024-03-15T10:30:00Z,resolved,5,456,web
124,2024-03-15T11:00:00Z,escalated,2,789,zendesk

Considerations:

Nested data (like messages in conversations) creates multiple rows
Text fields are quoted if they contain commas
Easy to open and read
Limited to flat data structures
Special characters may need escaping

When to use:

Creating charts or pivot tables in spreadsheet software
Sharing with non-technical team members
Quick visual inspection of data
Simple filtering and sorting

JSONL (JSON Lines)

Best for: Programming, data pipelines, AI analysis Structure:

{"id": 123, "status": "resolved", "rating": 5, "messages": [{"id": 1, "text": "Hello"}, {"id": 2, "text": "Hi there!"}]}
{"id": 124, "status": "escalated", "rating": 2, "messages": [{"id": 3, "text": "Help needed"}]}

Considerations:

Each line is a complete, valid JSON object
Preserves nested data structures
Easily processed line-by-line in any language
More compact than pretty-printed JSON
Human-readable but not as intuitive as CSV

When to use:

Importing to data warehouses or analytics platforms
Processing with Python, JavaScript, or other languages
Uploading to AI tools (ChatGPT, Claude) for analysis
Maintaining data relationships (users → conversations → messages)

Excel (CSV)

Best for: Microsoft Excel users Structure: Same as CSV, but with Excel-specific formatting optimizations:

UTF-8 BOM for proper character encoding
Optimized delimiter handling
Better date format compatibility

When to use:

You exclusively use Microsoft Excel
Working with international characters or special symbols
Need Excel to auto-detect data types correctly

Working with Media Files

Exports do not contain actual media files (images, videos, documents). Instead, they include time-bound download links for each media attachment.

How Media Links Work

In your exported data, media fields contain URLs like:

{
  "id": "msg_123",
  "text": "Here's the screenshot",
  "attachments": [
    {
      "id": "file_456",
      "type": "image/png",
      "filename": "screenshot.png",
      "url": "https://storage.botbrains.io/files/file_456?expires=1234567890&signature=abc123",
      "size": 245678
    }
  ]
}

URL characteristics:

Pre-signed with temporary access credentials
Valid for a limited time (typically 7 days)
Direct download - no authentication required
Original file quality and format preserved

Archiving Media Files

If you need permanent copies of media files: Option 1: Manual Download

Export your data
Extract media URLs from export file
Download each file before URLs expire
Store in your preferred storage system

Option 2: Automated Download Use a script to process exports and download media:

import json
import requests
from pathlib import Path

def download_media_from_export(export_file, output_dir):
    """Download all media files from a JSONL export"""
    output_path = Path(output_dir)
    output_path.mkdir(exist_ok=True)

    with open(export_file) as f:
        for line in f:
            user = json.loads(line)
            for conversation in user.get('conversations', []):
                for message in conversation.get('messages', []):
                    for attachment in message.get('attachments', []):
                        url = attachment['url']
                        filename = attachment['filename']

                        # Download file
                        response = requests.get(url)
                        if response.status_code == 200:
                            filepath = output_path / filename
                            filepath.write_bytes(response.content)
                            print(f"Downloaded: {filename}")
                        else:
                            print(f"Failed: {filename}")

# Usage
download_media_from_export('export.jsonl', 'media_archive/')

Media URLs expire after a limited time. Download media files promptly after exporting if you need permanent archives.

Custom Analysis with AI Tools

One of the most powerful uses of exported data is custom analysis with AI tools like ChatGPT, Claude, or Gemini. These tools can answer questions that go beyond built-in metrics.

Prerequisites for AI Analysis

Data Privacy Compliance Before uploading data to external AI tools:

Review your organization’s data policies - Ensure AI tool usage complies with GDPR, CCPA, HIPAA, or other regulations
Check AI provider policies - Verify data handling, retention, and training policies
Consider data sanitization - Remove PII (emails, names, phone numbers) if required
Use business subscriptions - Most AI providers offer business plans with stricter data protections
Document your decision - Maintain records of compliance verification

You are responsible for ensuring appropriate use of exported data. botBrains is not liable for misuse or policy violations when sharing data with external services. Most organizations allow ChatGPT Business with training disabled, but verify your specific policies around sensitive data.

Choose the Right Format

Use JSONL or CSV format for AI analysis
Full Export works best for complex cross-conversation questions
Entity exports work for focused analysis on specific data types

Base Prompt Template

Upload your exported file to your AI tool, then use this prompt template before asking your question:

You are a data analyst, using your Code Interpreter / Python Interpreter tools to write code and answer my questions. You've been given a JSON Lines file, of the conversations had by the ai-assistant, operators (human agents) and customers. Each line is a user profile, with all their conversations and all messages within those conversations, including metadata. For legacy reasons, not all conversations have a user, so a mock user with ID 00000000-0000-0000-0000-000000000000 is assigned all other conversations.

Follow this workflow:
1. Familiarize yourself with the data model by looking at the first 5 lines.
2. Think about my question and the appropriate chart / way to answer my question with the given values.
3. Work to write the code and find appropriate outputs format (chart-type, tables, singular values...) for the data.

Question: [Your question here]

Prompting Strategy: For best results, ask the AI to first filter data by your criteria (poor ratings, specific attributes), then analyze larger message content chunks to identify and summarize patterns.

Example Analysis Questions

Example 1: Identify Most Active Users

Export needed: Full Export Question:

Which users are the most active, as measured by messages? Count both user and assistant messages per user. Output a table sorted by message count descending, showing user ID, email, and total message count.

Use case: Identify power users, potential advocates, or users who may need extra support Who benefits: Product managers, customer success teams

Example 2: Analyze Premium Customer Issues

Export needed: Full Export Question:

What is the common issue users with external_attributes.plan = PREMIUM describe? First filter users by this attribute, then read their conversation messages. Analyze the content to identify recurring themes, issues, or feature requests. Output a summary with top 5 themes and example conversation IDs.

Prerequisites: You must set external attributes via Web SDK or API to segment users by plan Use case: Prioritize improvements for high-value customers, identify churn risks Who benefits: Product managers, account managers, product teams

Example 3: Understand Customer Demographics

Export needed: User Export Question:

How many users have an email associated? What is the most common email domain? Extract domains from email addresses, group by domain, and show a table with domain counts sorted descending. Also show the percentage of users with vs without emails.

Use case: Understand customer composition, identify B2B vs B2C mix Who benefits: Marketing teams, sales teams

Example 4: Track Feature Interest

Export needed: Conversations Export (includes messages) Question:

How many users ask about our Product/Feature "AcmeProduct"/"AcmeFeature"? Lookup the keywords in the messages content and output matching conversations. Read those and summarize common questions or concerns.

Use case: Validate product roadmap, measure feature demand, identify gaps in documentation Who benefits: Product managers, marketing teams

Example 5: Improve Low-Rated Conversations

Export needed: Conversations Export (includes messages) Question:

Search for conversations with poor rating (1-2). Read the message content in these conversations to identify product or feature feedback, pain points, and complaints. Categorize the issues and output the top 5 categories with example conversation IDs and representative quotes.

Use case: Systematically improve AI responses, identify product issues, reduce DSAT Who benefits: Chatbot improvement teams, product teams, support operations

Advanced Analysis Patterns

Sentiment Trend Analysis

Plot sentiment scores over time, grouped by week. Identify weeks with significant negative sentiment spikes and list the conversation topics during those periods.

Conversation Flow Analysis

For conversations with ratings of 1-2, analyze the message sequence. How many messages does it take before sentiment turns negative? What triggers the sentiment change? Provide examples.

Topic Distribution by User Segment

Group users by external_attributes.company_size (small, medium, large). For each segment, show the distribution of conversation topics. Identify which topics are unique to each segment.

Tool Effectiveness Analysis

Identify conversations where the AI used specific tools (search_web, handoff, etc.). Compare the average rating of conversations with tool usage vs without. Determine if certain tools correlate with higher or lower satisfaction.

Data Structure Reference

Understanding the data structure helps you select the right fields and interpret exports correctly.

Full Export JSONL Structure

{
  "id": "user_123",
  "email": "user@example.com",
  "name": "John Doe",
  "created_at": "2024-01-15T10:00:00Z",
  "last_seen_at": "2024-03-15T14:30:00Z",
  "conversation_count": 5,
  "message_count": 23,
  "external_attributes": {
    "plan": "premium",
    "company": "Acme Corp"
  },
  "conversations": [
    {
      "id": "conv_456",
      "created_at": "2024-03-15T10:00:00Z",
      "updated_at": "2024-03-15T10:15:00Z",
      "status": "resolved",
      "rating": 5,
      "channel": "web",
      "topics": ["billing", "pricing"],
      "messages": [
        {
          "id": "msg_789",
          "text": "How do I upgrade my plan?",
          "role": "user",
          "created_at": "2024-03-15T10:00:00Z",
          "sentiment": "neutral"
        },
        {
          "id": "msg_790",
          "text": "I'd be happy to help you upgrade! You can...",
          "role": "assistant",
          "created_at": "2024-03-15T10:00:05Z",
          "tool_calls": []
        }
      ]
    }
  ]
}

Key relationships:

Each line = one user
Users contain array of conversations
Conversations contain array of messages
Maintains full hierarchy and relationships

Conversation Export Structure

JSONL format:

{
  "id": "conv_456",
  "user_id": "user_123",
  "created_at": "2024-03-15T10:00:00Z",
  "status": "resolved",
  "rating": 5,
  "messages": [...]
}

CSV format:

id,user_id,created_at,status,rating,message_id,message_text,message_role
conv_456,user_123,2024-03-15T10:00:00Z,resolved,5,msg_789,"How do I upgrade?",user
conv_456,user_123,2024-03-15T10:00:00Z,resolved,5,msg_790,"I'd be happy to help",assistant

Note: CSV format duplicates conversation fields for each message row.

Message Export Structure

Messages as individual records, useful for message-level analysis:

{
  "id": "msg_789",
  "conversation_id": "conv_456",
  "text": "How do I upgrade my plan?",
  "role": "user",
  "created_at": "2024-03-15T10:00:00Z",
  "sentiment": "neutral",
  "topics": ["billing"]
}

User Export Structure

Users without nested conversations:

{
  "id": "user_123",
  "email": "user@example.com",
  "name": "John Doe",
  "created_at": "2024-01-15T10:00:00Z",
  "conversation_count": 5,
  "message_count": 23
}

Common Use Cases

When a customer requests their personal data:

Navigate to Analyze → Users
Search for user by email or ID
Note the user ID
Go to Settings → Data Export
Click Full Export
Download export file
Extract the user’s line from JSONL (contains all conversations and messages)
Provide to data subject

For frequent GDPR requests, create a script to extract specific user data from full exports. Store exports monthly for faster response times.

Weekly Executive Reports

Create a recurring export process:

Every Monday morning, export previous week’s conversations
Filter: Date range = Last 7 days
Export format: CSV
Open in Excel or Google Sheets
Create pivot table summarizing:
- Total conversations
- Average CSAT
- Resolution rate
- Top topics
Share with stakeholders

Integration with Analytics Platforms

Send data to Tableau, Looker, or similar tools:

Schedule weekly Full Exports (manually or via API if available)
Download JSONL files
Use ETL process to load into data warehouse
Create dashboards combining conversation data with product analytics
Track correlations between product usage and support conversations

Quality Assurance Sampling

Sample conversations for QA review:

Navigate to Analyze → Conversations
Filter: Random sampling (export all, randomly select subset)
Export with fields: id, rating, status, messages
Review sampled conversations for quality
Create improvement tasks based on findings

Compliance Audits

Maintain audit trail for regulatory requirements:

Monthly full export with date range = previous month
Archive JSONL files with naming convention: audit_2024-03_export.jsonl
Store in compliant storage (encrypted, access-controlled)
Maintain index of export dates and file locations
Produce when requested for audits

Troubleshooting

Export Failed or Empty

Symptoms: Export completes but file is empty or contains no data Possible causes:

Date range filters exclude all data
No data exists for selected entity
Active filters on entity page are too restrictive

Solutions:

Remove all filters and try again
Verify data exists for selected time range
Check date range settings (ensure “From” is before “To”)

CSV Not Opening Correctly in Excel

Symptoms: Special characters display incorrectly, columns not separated properly Possible causes:

Character encoding mismatch
Excel locale settings

Solutions:

Use “Excel (CSV)” export format instead of standard CSV
Import CSV using Excel’s “Get Data” feature with UTF-8 encoding
Open in Google Sheets first, then download as Excel format

Rate Limit Exceeded

Symptoms: Error message “Export limit reached, try again at [time]” Possible causes:

Already requested 2 full exports today
Another team member requested exports

Solutions:

Wait until specified retry time
Coordinate with team on export schedule
Use entity exports instead (no rate limit)
Use date range filters for incremental exports instead of full exports

Media Links Not Working

Symptoms: 403 Forbidden or 404 Not Found when accessing media URLs Possible causes:

URLs have expired (typically after 7 days)
URLs were copied incorrectly

Solutions:

Request new export to get fresh URLs
Download media files immediately after export
Set up automated download process

Large File Handling

Symptoms: Export file is too large to open in Excel (more than 1 million rows) Possible causes:

Dataset exceeds spreadsheet limits
Conversation export with many nested messages creates excessive rows in CSV

Solutions:

Use JSONL format instead of CSV
Use date range filters to create smaller exports
Process data programmatically (Python, R) instead of spreadsheet tools
Use entity exports with specific date ranges

JSONL File Format Issues

Symptoms: File appears as single long line, difficult to read This is normal: JSONL files are designed for programmatic processing, not human reading Solutions:

Use a code editor with JSONL support (VS Code, Sublime Text)
Process with Python or JavaScript instead of opening directly

Use command-line tools like jq for inspection:

# View first user prettily
head -n 1 export.jsonl | jq '.'

# Count total users
wc -l export.jsonl

# Extract specific field
jq -r '.email' export.jsonl

Best Practices

Regular Backup Schedule

Establish a consistent backup routine: Weekly backups:

Every Sunday night, export previous week
Date range: Last 7 days
Naming: weekly_backup_2024-03-15.jsonl
Store in cloud storage with versioning

Monthly full backups:

First of each month, export previous month
Date range: Previous calendar month
Naming: monthly_backup_2024-03.jsonl
Archive for long-term retention

Before major changes:

Export all data before deploying new AI guidance
Enables rollback analysis if issues arise

Field Selection Strategy

For backups: Export all fields to preserve complete records For analysis: Select only needed fields to reduce file size and processing time For sharing: Remove PII fields (email, name) when sharing with external parties

File Naming Conventions

Use descriptive, sortable names:

Good:
- botbrains_full_2024-03-15.jsonl
- conversations_escalated_2024-03-01_to_2024-03-31.csv
- users_premium_2024-Q1.csv

Avoid:
- export.jsonl
- data.csv
- file1.jsonl

Data Retention Policy

Define how long you keep exports: Regulatory requirements:

GDPR: Retain only as long as needed for business purposes
HIPAA: Minimum 6 years for medical data
CCPA: Enable deletion on request

Business needs:

Active analysis: 90 days
Historical trends: 1-2 years
Compliance archives: 7+ years

Storage optimization:

Compress old exports (gzip reduces size by 80 percent or more)
Move to cold storage after 90 days
Delete non-compliant exports regularly

Data Sanitization

When sharing exports with external parties or AI tools: Remove PII:

import json

def sanitize_export(input_file, output_file):
    """Remove PII from export file"""
    with open(input_file) as f_in, open(output_file, 'w') as f_out:
        for line in f_in:
            user = json.loads(line)
            # Remove identifiable information
            user['email'] = 'redacted@example.com'
            user['name'] = f"User {user['id']}"
            user.pop('external_attributes', None)

            f_out.write(json.dumps(user) + '\n')

sanitize_export('export.jsonl', 'export_sanitized.jsonl')

Aggregate before sharing:

Share summary statistics instead of raw data
Provide anonymized samples rather than full datasets

Next Steps

Now that you understand data export:

Review Conversations - Filter and analyze conversations before exporting
Track Metrics - Use built-in analytics before resorting to custom exports
Message Search - Find specific messages to export
Project Settings - Configure project-level settings
Security & Permissions - Understand who can export data

Start with small, focused exports to understand the data structure before attempting large-scale exports or complex analysis workflows.

Getting Started

Core Concepts

Train

Deploy

Analyze

Guides

Security

Other

More

​Why Data Export Matters

​Export Types

​Full Dataset Export

​Entity-Specific Export

​Choosing the Right Export Method

​Performing a Full Dataset Export

​Prerequisites

​Step-by-Step Instructions

​Rate Limits and Scheduling

​Performing Entity-Specific Exports

​Exporting Conversations

​Exporting Messages

​Exporting Users

​Understanding Export Formats

​CSV (Comma-Separated Values)

​JSONL (JSON Lines)

​Excel (CSV)

​Working with Media Files

​How Media Links Work

​Archiving Media Files

​Custom Analysis with AI Tools

​Prerequisites for AI Analysis

​Base Prompt Template

​Example Analysis Questions

​Example 1: Identify Most Active Users

​Example 2: Analyze Premium Customer Issues

​Example 3: Understand Customer Demographics

​Example 4: Track Feature Interest

​Example 5: Improve Low-Rated Conversations

​Advanced Analysis Patterns

​Data Structure Reference

​Full Export JSONL Structure

​Conversation Export Structure

​Message Export Structure

​User Export Structure

​Common Use Cases

​GDPR Data Subject Access Requests

​Weekly Executive Reports

​Integration with Analytics Platforms

​Quality Assurance Sampling

​Compliance Audits

​Troubleshooting

​Export Failed or Empty

​CSV Not Opening Correctly in Excel

​Rate Limit Exceeded

​Media Links Not Working

​Large File Handling

​JSONL File Format Issues

​Best Practices

​Regular Backup Schedule

​Field Selection Strategy

​File Naming Conventions

​Data Retention Policy

​Data Sanitization

​Next Steps

Why Data Export Matters

Export Types

Full Dataset Export

Entity-Specific Export

Choosing the Right Export Method

Performing a Full Dataset Export

Prerequisites

Step-by-Step Instructions

Rate Limits and Scheduling

Performing Entity-Specific Exports

Exporting Conversations

Exporting Messages

Exporting Users

Understanding Export Formats

CSV (Comma-Separated Values)

JSONL (JSON Lines)

Excel (CSV)

Working with Media Files

How Media Links Work

Archiving Media Files

Custom Analysis with AI Tools

Prerequisites for AI Analysis

Base Prompt Template

Example Analysis Questions

Example 1: Identify Most Active Users

Example 2: Analyze Premium Customer Issues

Example 3: Understand Customer Demographics

Example 4: Track Feature Interest

Example 5: Improve Low-Rated Conversations

Advanced Analysis Patterns

Data Structure Reference

Full Export JSONL Structure

Conversation Export Structure

Message Export Structure

User Export Structure

Common Use Cases

GDPR Data Subject Access Requests

Weekly Executive Reports

Integration with Analytics Platforms

Quality Assurance Sampling

Compliance Audits

Troubleshooting

Export Failed or Empty

CSV Not Opening Correctly in Excel

Rate Limit Exceeded

Media Links Not Working

Large File Handling

JSONL File Format Issues

Best Practices

Regular Backup Schedule

Field Selection Strategy

File Naming Conventions

Data Retention Policy

Data Sanitization

Next Steps