Find and report empty cells and missing values
Missing values can silently skew analysis results or break downstream processes. This analyzer scans every column for empty cells and common placeholders, then reports how complete each column is, classifies the overall pattern of missingness, and can fill the gaps for you using a strategy you pick per column.
Initializing in your browser…
A dataset feels incomplete and you need to quantify exactly where the gaps are before cleaning.
Input
survey.csv (5,000 rows, 20 columns)
Report
income: 31% missing region: 0% 42 rows entirely blank after column 6 Columns with any nulls: 7 of 20
A per-column completeness breakdown plus row-level gap patterns tells you whether to impute, drop, or go back to the source, and reveals structural issues like a block of rows truncated mid-file. Guessing at missingness is how biased analyses happen.
Missing values can silently skew analysis results or break downstream processes. This analyzer scans every column for empty cells and common placeholders, then reports how complete each column is, classifies the overall pattern of missingness, and can fill the gaps for you using a strategy you pick per column.
The tool treats a cell as missing when it is empty or one of the placeholders "null", "n/a", "na", or "-" (case-insensitive). It calculates a missing count and percentage for every column and an overall completeness figure, then classifies the pattern of missingness as random, systematic, clustered, or none (for example, flagging a column that is more than half empty as systematic). Each column with gaps is shown as a card with a color-coded severity bar and percentage badge (green to red as the missing rate rises), and consecutive-gap runs are flagged, so you can spot problem areas at a glance across many columns.
Beyond analysis, the tool can repair the data. For each column with gaps you choose a fill strategy, forward fill, backward fill, a constant value, the column mean, median, or mode, or linear interpolation, and it suggests a sensible default (for example, mean for numeric columns, and no auto-fill for high-uniqueness columns like IDs or emails). An optional preview highlights the cells that were filled, and you can then download the filled CSV. A separate plain-text report captures the per-column missing counts, the overall percentage, and the specific missing-cell locations.
Understand how complete your dataset is before running statistical models that assume no missing values.
Verify that data deliveries from third parties meet agreed completeness thresholds.
Identify columns with too many gaps to migrate into a new system with NOT NULL constraints.
Decide which columns need imputation and which strategy (mean, median, forward-fill) to apply based on gap patterns.
Empty cells and the placeholders "null", "n/a", "na", and "-" (matched case-insensitively).
Yes. It classifies missingness as random, systematic, or clustered, for example flagging a column that becomes mostly empty as systematic, which can point to schema changes or data-collection issues.
Yes, the analyzer processes files with hundreds of thousands of rows efficiently in your browser.
Rows and columns are parsed and transformed in memory in your browser. No record ever reaches a server.