Skip to main content
Filtering lets you narrow a dataset to only the records relevant to your question. Most analysis starts here.

​
Apply Filters

You can filter data using natural language conditions:
  • Filter by value: “Show transactions from California”
  • Filter by number: “Show orders over 1000”
  • Filter by date: “Show customers who joined in 2024”
Multiple conditions can be combined naturally: “Show enterprise customers from California with orders over 1000.”

​
Filter Types

Filter TypeDescriptionExamples
Comparisons and RangesFilter by numeric comparisons and value ranges• “Orders between 100 and 500”
• “Customers with more than 5 purchases”
• Greater than / less than
• Between two values
• At least / at most
Text PatternsFilter text fields by patterns• Contains
• Starts with
• Ends with
Time-BasedFilter by dates and time ranges, expressed naturally or explicitly• “Last month”
• “This year”
• “Between January and March 2024”
• “Last 30 days”

Zark automatically interprets relative time ranges.
Missing DataFilter based on missing or empty values. Useful for validating data quality or identifying gaps.• “Show records where email is missing”
• “Exclude rows with empty regions”
• “Count customers with and without phone numbers”
Exclude/NegationFilter out specific values. Helps identify exceptions, gaps, and outliers.• “Orders not from California”
• “Customers who haven’t purchased this year”
• “Tickets not resolved”