What it reads, and how well.

Supported is not one thing. A CSV is the data. A table inside a PDF has to be reconstructed. A photograph of a stocktake sheet is read by eye. All three are accepted here, and each says which it is.

Formats with a page: 15
Fidelity tiers: 3
Anything refused: before upload

Read directly

The file is the data. Nothing is inferred and nothing is lost.

Data extracted

The numbers live inside something built for another purpose. They come out intact, and every decision made on the way out is listed.

Read by eye, then checked

Figures are read from an image. Each one is re-derived against the totals in the same picture, and anything that cannot be read confidently is reported rather than guessed.

Spreadsheets

The common case. Every sheet is read, including the ones that are notes.

.xlsxExcel workbookEvery sheet read, including the ones that are notes rather than data.
.csvCSVDelimiter, quoting and encoding detected rather than assumed.
.odsOpenDocument spreadsheetLibreOffice and OpenOffice workbooks, read like Excel.
.tsvTab-separatedRead as delimited text.
.xlsExcel 97–2003Read the same as a modern workbook.
.xlsmExcel, macro-enabledValues and formulas are read. Macros are never executed.
.numbersApple NumbersTables are pulled out of the Numbers package sheet by sheet.

Documents & decks

Reports written for people, with tables buried inside them.

.pdfPDFChecked for a text layer first. A PDF written by software keeps its table structure; a scan is read as an image and told apart from the first kind.
.docxWord documentTables are lifted out of the document; the prose around them is kept as context.
.pptxPowerPoint deckTables and the data behind native charts are read slide by slide.
.rtfRich textTables extracted, formatting discarded.

Database & app exports

Whatever your warehouse, API or admin panel hands over.

.parquetParquetRead columnwise, with the file's own types kept.
.jsonJSONNested objects are flattened into tables, keeping the path as the column name.
.sqliteSQLite databaseEvery table read, with the foreign keys used to work out how they relate.
.xmlXMLRepeated elements become rows; attributes become columns.
.ndjsonNewline-delimited JSONStreamed a record at a time, so an export of any size is fine.
.dbSQLite databaseRead as SQLite when the file is one, and reported plainly when it is not.

Dashboards

The data behind a report someone else built — not a picture of it.

.pbixPower BI reportThe tables inside the report are read out of its model, along with the measures that were defined on top of them, so your own definitions are used rather than re-invented.
.twbxTableau packaged workbookThe packaged extract is read as data, and the workbook beside it supplies the calculated fields.
.pbitPower BI templateA template carries the model and the measures without the data, and is read for both.Definitions only
.twbTableau workbookCalculated fields and the shape of each sheet are read.Definitions only
.hyperTableau extractRead straight through, table by table.

Bank & accounting

What a bank will actually give you, with the reference codes intact.

.ofxOFX statementTransactions come with their own type codes and references, which makes matching far better than a CSV.
.qfxQuicken statementRead as OFX, which is what it is underneath.
.qboQuickBooks statementRead as OFX, with the QuickBooks account mapping kept.
.staMT940 statementThe SWIFT tags are parsed into transactions, balances and references.

Survey & research

Files that carry their own codebook, so the labels survive the trip.

.savSPSS dataVariable labels, value labels and missing-value codes are all read, so a 5 stays “Strongly agree” and a 99 stays missing.
.dtaStata dataVariable and value labels are kept, so the codebook does not have to be re-typed.
.zsavSPSS data, compressedRead the same as .sav.
.porSPSS portableRead the same as .sav.
.sas7bdatSAS dataRead with its formats, including dates.
.xptSAS transportThe interchange format SAS itself exports. Read like a .sas7bdat.

Screenshots & scans

A photograph of a printed sheet is a data source like any other.

.jpgJPEG photoA photograph of a printed report or a handwritten sheet is read into a table.
.pngPNG imageScreenshots of a dashboard or a table are read, and every figure is checked against the totals in the same image.
.jpegJPEG photoThe same as .jpg.
.heiciPhone photoAccepted as it comes off the phone. Safari usually hands it over as a JPEG.
.webpWebP imageRead like any other screenshot.
.tiffTIFF scanMulti-page scans are read page by page.

Notes & bundles

Context for the analysis, and a way to send a whole folder at once.

.zipZip archiveUnpacked, and each file inside is treated as though you had uploaded it yourself.
.txtPlain textRead as notes, or as data when the contents turn out to be a table.
.mdMarkdownMarkdown tables become tables; the rest becomes context for the analysis.

Every format above is accepted today. The ones that link through to a page are the ones where there is something specific worth saying about analysing them, which is a smaller set than the ones that upload.