Outlier Mgmt
Upload a dataset to begin
Supported formats: .csv,.xlsx,.xls,.xlsm,.xlsb,.tsv
Sign in for free for unlimited files, no payment required.
About Outlier Mgmt
Cap, remove, or flag anomalies in your dataset.
This is the treatment step that follows outlier detection: once you know which values are extreme, this tool lets you decide what happens to them.
Choose to winsorize (cap values at a percentile boundary so they stay in the dataset but stop skewing averages), delete the offending rows entirely, or simply tag them in a new column for manual review later.
The right choice depends on context, capping preserves row count for modeling, while removal is often better when you're confident the value is a data-entry error.
Category: AI & Machine LearningCommon Use Cases
- Capping extreme salary values before calculating a department average
- Removing sensor glitches before running a regression
- Flagging suspicious order amounts for a finance team to review manually
Key Features
- Winsorization (Percentile Capping)
- Row Removal
- Flag-Only Mode for Manual Review
- Threshold-Based Rules
Detect anomalies with Outliers first, then bring the same column here to cap, remove, or flag them based on what you found, checking Stats afterward to confirm the mean and variance settled down.
Frequently Asked Questions
You can cap them to a threshold, remove them entirely, or just flag them for review, depending on what fits your analysis.
No, it's the action step, pair it with the Outliers tool's detection if you want to inspect anomalies before deciding what to do with them.
Built for analysts and data engineers who need to decide, not just discover, what happens to extreme values before modeling or reporting.
Read the related guide →