Turbo Excel Cleaner

Blank rows left by an export, duplicated records from two systems merged into one sheet, dates written four different ways, trailing spaces that break every lookup formula, a header row repeated eleven times because someone concatenated eleven CSV files. Fixing that by hand means Find & Replace, Go To Special and sort-then-delete, one workbook at a time.

Turbo Excel Cleaner turns the job into a checklist. Load one file or a queue of exports, tick the operations you want from a list of 43, and the app applies them to every sheet in every file. Originals are never modified; each result is a separate cleaned copy.

The 43 features sit in four categories matching the four kinds of mess a spreadsheet accumulates. Rows & Columns handles structural junk. Text Cleaning handles what is inside the cells. Data Standardization makes values consistent. Structure & Formatting deals with the workbook itself — formulas, comments, hyperlinks, formatting rules, images, sheets.

Every feature starts unchecked, so nothing runs until you opt in. One Recommended button applies a curated set of 19 safe, non-destructive features, chosen so they are safe on data you have not reviewed yet. Everything runs locally, so workbooks of financial or customer records never leave the machine.

The engine works on rows, columns and cells rather than assuming a source, so the same feature list applies whether the file came from an accounting package, a CRM, a survey platform or an internal template.

Features that only make sense for a real Excel file — hidden rows, merged cells, comments, hyperlinks, formatting rules, images, validation, gridlines, freeze panes, auto-fit — are skipped automatically for CSV, TSV and TXT. Numbers written to delimited formats come back as 200, not 200.0.

These attack the shape of the sheet rather than the contents of the cells. Each is an independent checkbox.

Where the invisible problems get fixed — the ones that make VLOOKUP return #N/A on two strings that look identical on screen.

All three case options leave email addresses, phone numbers and already-standardized YYYY-MM-DD dates untouched, whichever mode you pick.

These take values that mean the same thing but are written differently and make them match. Each acts only on values it can confidently recognize.

This category works on the workbook rather than its values: the objects, rules and decorations that make a file heavy, fragile or awkward to hand on.

The last two operate across the whole workbook by definition, so they run only in All Sheets mode. Target a single sheet and they are skipped automatically.

Forty-three checkboxes is a lot of decisions about a file you have not opened. Recommended applies a fixed set of 19 features chosen to be broadly useful and non-destructive.

• Remove hidden rows, remove merged cells
• Remove repeated header rows, remove separator rows
• Remove blank rows, remove duplicate rows
• Fix encoding, trim spaces, collapse whitespace, remove non-printable characters
• Clear null placeholders, trim column headers
• Standardize dates, preserve leading zeros, fix scientific notation, convert text numbers to numbers
• Freeze header row, auto-fit column widths, remove empty sheets

What it leaves out matters as much: no case rewriting, no outlier removal, no boolean normalization — nothing opinionated about what your data should look like. It always produces the same 19 regardless of what is ticked, so it doubles as a reset. Select All and Deselect All sit beside it; Select All turns on one case mode rather than all three, since only one can apply per run.

A compact fixed window in three areas: a top bar, a sidebar for input and output, and a main panel holding the feature checklist beside a live activity log.

Top bar

License status on the left — the email the copy is licensed to, or "Unlicensed" — and three buttons on the right: Help for a step-by-step guide, About for version and publisher, and Check for Updates, which opens the product page carrying the latest build.

Sidebar

Add Files and Clear, a scrollable queue with a remove control on each entry, a running count, then Sheet Selection and Choose Output Folder. A green Run Cleaning button and progress bar are pinned to the bottom. Add Files rejects unsupported types up front, naming which files were skipped, and never queues the same file twice.

Feature checklist and log

The 43 features sit two per row inside four bordered cards, one per category, with Select All, Deselect All and Recommended in the header. The two features carrying an extra setting render it inside their own card — the z-score box sits directly under outlier removal, so you cannot change a threshold without seeing the feature it belongs to.

A dark panel down the right streams progress: which file is opening, which sheet is being cleaned, and a line per feature reporting how many cells, rows or sheets it actually changed. Features that changed nothing stay quiet, so the log reads as a summary rather than a wall of noise.

• Exactly one file loaded: the dropdown fills with that workbook's real sheet names plus All Sheets. Pick one and only that sheet is cleaned; every other sheet is written back untouched.
• Zero or several files loaded: it locks to All Sheets, since per-sheet targeting stops being meaningful across workbooks with different sheet names. Every sheet in every file is cleaned.
• CSV, TSV and TXT: always report a single sheet, since plain text has no concept of multiple sheets.

The dropdown refreshes whenever files are added or cleared, and re-locks the instant a second file joins. If a targeted sheet name does not exist in a given file, that file is skipped with a warning rather than failing the batch.

Cleaning operations interfere with each other. Run them in the wrong sequence and you get a file that is technically cleaned and practically wrong, so the pipeline applies a fixed order no matter which boxes you tick.

1. Workbook-level sheet removal runs first, so no work is wasted on a sheet about to disappear.
2. Merged cells are flattened before any deletion, because merge ranges do not re-anchor reliably when rows shift.
3. Repeated headers and separator rows are detected before text cleaning. Both match whole rows against raw content; trim the data rows first and a trimmed duplicate header no longer matches the untrimmed original.
4. Text cleaning runs next — encoding, trimming, whitespace, null placeholders, control and special characters, quotes, case, headers.
5. Phone and email standardization run before number conversion, or a digit-only phone number is silently turned into a plain number.
6. Leading-zero preservation is respected by number conversion, so ZIP and ID columns keep their zeros.
7. Row and column reduction runs after text cleanup, so blank and duplicate detection compare cleaned values.
8. Formula flattening and formatting cleanup run last, once all content work is done.

Case is three plain checkboxes inside the Text Cleaning card rather than an enable toggle paired with a mode dropdown elsewhere. That pairing is easy to get wrong — pick a mode without ticking the second control and you get a cleaned file identical to the original. Here, ticking a mode is the enable action, and ticking one clears the other two.

Title casing is word-aware rather than naive. A naive routine treats an apostrophe as a word boundary and produces O'Brien'S Data; this one treats each letter run including internal apostrophes as one word.

Originals are never modified. Every run produces new files, written beside the source with a _cleaned suffix or into an output folder you choose.

Native Excel formats are saved back as they came in. Legacy binary formats — .xls, .xlsb, .ods — are always converted to .xlsx. Delimited formats keep their own: a CSV returns a CSV, a TSV a TSV.

Two things block a save regularly, neither a bug: a previous cleaned copy is still open in Excel, or Windows Controlled Folder Access is protecting Downloads, Documents and Desktop, which it does by default. Rather than a raw error, the app retries with alternate filenames — _cleaned_2.xlsx, _cleaned_3.xlsx — up to four attempts, logging which it used. Only if all four are blocked does it stop, with one clear message naming the likely causes.

Cleaning happens entirely on your machine. Nothing is uploaded, there is no cloud step and no account to create, and once installed the app works with the network disconnected. The only traffic it generates is license verification, which sends your purchase email and key and nothing else. That matters when the files you need to clean are payroll runs or client ledgers.

Minimum

Recommended

Cleaning holds a workbook in memory, so peak memory use scales with the largest single file rather than with how many files are queued.

1. Purchase and download the Windows build. No subscription, no recurring fee.
2. Launch it. On first run, an activation window asks for the email you purchased with and your license key.
3. Activate once. The pair is verified, then cached locally so you never re-enter it.
4. Later launches are silent. The cached license is re-verified before the main window opens; the activation window only reappears if that check fails.
5. Check your status any time in the top-left corner of the main window.

To switch licenses, delete the cached license file from the app's folder in your local application data directory and relaunch.

Ready to clean your spreadsheets automatically, in bulk, fully offline?