Choose the next step for your CSV export

What are you trying to do?

Check one export before working with it

Read supported CSV or TSV structure, then optionally analyze an exact key. See empty and whitespace-only keys, duplicate groups and every member. No chosen key means key checks are not run. A parse failure reports the first error; it does not collect every error or certify the file.

Compare an earlier and later export

Choose a shared exact key to see added, removed, changed and unchanged records. Row reordering alone is not a change. Columns align by exact header name; an absent column differs from an empty cell. Blank and ambiguous duplicated keys are excluded visibly rather than matched by guesswork.

Add fields from a reference export

Choose the left and lookup IDs, then append selected reference fields in an explicit order with explicit names. Every left record stays in source order, including repeated IDs. All duplicate nonblank lookup IDs block merging. Blank lookup rows are ignored and reported; blank or missing left matches remain unmatched. Metadata distinguishes matched empty values from unmatched empty values.

Review repeated identities before retaining records

Select one or more identity columns and inspect every member. Deliberately retain the first or last source occurrence, then export retained and removed records separately. There is no default winner. If any selected identity field is empty or whitespace-only, that record is incomplete, ungrouped and always retained.

Prepare the required fields and headers

Select a subset of columns, arrange their order and give exact nonblank unique output names. All records stay in original order, including repetitions, and raw selected strings stay unchanged. Header renaming does not convert values or prove compatibility with a destination.

A useful sequence, when you need it

Check structure first. Compare a recurring export or add reference fields when that is your job. Review duplicate identities only when you intend to choose a retention policy. Prepare columns for the final file. Each tool works independently; pages do not carry your data into the next tool. Merged, retained-duplicate and prepared-column CSVs can be inspected and explicitly loaded into another tool. Comparator CSV reports and health JSON diagnostics are not replacement source exports.

Exact values and protective limits

Inputs require UTF-8 CSV or TSV with a header and an explicitly chosen comma or tab separator. Each input is limited to 2 MiB, 20,000 data records, 100 columns and 200,000 total cells including the header. These caps are not a phone performance promise. Supported quoted separators, quotes and multiline cells preserve their decoded strings. Case, nonblank spaces and leading zeros matter.

Health checking exports JSON diagnostics. The other tools offer raw JSON and explicit CSV handling modes. Literal CSV preserves its exported values but spreadsheet software may interpret formulas. Spreadsheet-prefix mode adds apostrophes to formula/control-leading fields, changes those strings and is not a universal spreadsheet safety guarantee. Raw JSON stays unprefixed. Read each page's output budgets and report format; a comparator report is not a replacement source export. No tool repairs data, evaluates business correctness or certifies an import.

Data handling in this collection

Selected file contents are processed in your browser without input uploads, remote APIs, analytics, advertising or browser storage. The hosting service receives requests for the pages and their static assets. Downloaded CSV and JSON can contain sensitive original values; store and share them deliberately. Clear removes displayed inputs and results, but does not delete saved downloads. Browser extensions and other software on your device are outside these pages' control.