Null Value Analyzer
See exactly where data is missing. The analyzer counts empty cells and placeholders (NULL, N/A, NaN, “-”, plus your own such as “unknown”) per column, shows how they were written, how many rows are complete and which combinations of missing columns occur together. Export a CSV with all placeholders made empty, without incomplete rows, or without columns that are more than half empty.
- Runs in your browser
- No sign-up
- Free to use
Processed entirely in your browser – your data is not uploaded. Files up to 50 MB.
How to use Null Value Analyzer
- Paste data or open a file.
- Add extra placeholders (optional).
- Review columns and patterns.
- Export a cleaned CSV if needed.
Null Value Analyzer features
Per-column counts
With share and bar.
Placeholders
Shows how missing values were written.
Patterns
Which columns are missing together.
Clean exports
Blank, drop rows or drop columns.
JSON keys
Absent keys count as missing.
Private
Runs in your browser; nothing is uploaded.
When to use Null Value Analyzer
- Preparing data for analysis or machine learning.
- Checking CRM exports.
- Survey data cleaning.
- Data migration QA.
Null Value Analyzer FAQ
Should I delete rows with missing values?
Only if few rows are affected; otherwise you may bias the data.
Why do patterns matter?
Values missing together often share a cause, such as an optional form section.
Is “0” missing?
No, unless you add it as a placeholder.
Should I replace missing numbers with zero?
Only when zero is really the meaning. Otherwise averages and totals become misleading; leaving the value missing is often more honest.
What placeholders are recognised?
Empty cells, NULL, N/A, NA, NaN, “-” and similar, plus any custom placeholders you add, such as “unknown” or “tbd”.
Can it fill missing values?
No. It shows where and how values are missing so you can decide how to handle them in the cleaner or your analysis tool.
Missing is information
How data is missing tells you about the process that produced it.
Missing data is written in many ways: empty cells, NULL, N/A, NaN, “-” or words such as “unknown”. The analyzer counts all of them per column, including your own placeholders, and shows how they were written – a sign of data coming from different systems.
Patterns matter as much as counts. If phone and address are always missing together, the records probably came from a source that does not collect them. Complete-row counts show how much data is left if you drop incomplete records.
Decide per column: drop the column, drop rows, fill a sensible default or keep the gap and handle it in analysis. The analysis runs in your browser; nothing is uploaded.
Missing data is not always random. If values are missing more often for one region, product or time period, analyses based on complete rows can be biased. Look at where gaps cluster before deciding to drop incomplete records.
Record how you handled missing values so that reports built on the data can explain their assumptions.