Blog

2026 in Review: New Formats, Connectors and Data Prep Tools

Business file formats you can now open as spreadsheets

Much of 2026 went into formats that businesses receive but cannot open in Excel. Each one arrives as XML or JSON built for machines, and each is now a source format for conversions, viewing, and Data Prep. Upload the file, pick the dataset you want in File Settings, and download it as Excel, CSV, JSON, or any other target.

  • OFX, QFX, and QBO statements (September): bank and brokerage downloads in the Open Financial Exchange family, including the Quicken and QuickBooks variants, so the transactions in a statement download can become a spreadsheet.
  • ACES and PIES (November): the Auto Care Association standards for automotive fitment and product data. ACES applications, assets, and digital file information, and PIES items, descriptions, prices, attributes, packages, kits, interchanges, and digital assets, are read one dataset at a time.
  • BMEcat and ETIM xChange (November): the XML and JSON catalog standards that manufacturers and wholesalers exchange, most often with ETIM classification. BMEcat gives you the header, products, features, prices, media, and references; ETIM xChange gives you the catalog, products, trade items, features, prices, packaging, and attachments.
  • Peppol UBL e-invoices (November): Peppol BIS Billing 3.0 invoices and credit notes, split into the invoice, its lines, the VAT breakdown, allowances and charges, payment details, and attachments. Columns are named after the EN 16931 business terms.
  • cXML procurement documents (November): the XML that networks such as SAP Ariba and Coupa exchange with suppliers. Purchase orders, punch-out carts, and invoices are split into header, line, accounting, tax, and extrinsic datasets.

New ways to get data in and out

Connectors came first. Signed-in users can save a connection to Amazon S3 (October), browse the bucket to load files, and send results back to it. Google Drive followed at the end of October: pick files in your Drive as the input and save converted or transformed files into a Drive folder, with only the files you select, and the files DataConverter.io creates, shared with it. Google Sheets works both ways too: load a spreadsheet, every tab included, and send the result back as a new sheet in the Drive folder you choose.

Output formats grew as well. Microsoft Word (.docx) arrived in October with control over page size and orientation, and LaTeX (.tex) in June, for tables headed into a paper or a report. JSON Lines (NDJSON) became both a source and a target in July, and WebP images joined JPG and PNG as inputs for image reading in September.

Three smaller changes save a step each time. The CSV delimiter (comma, tab, pipe, semicolon, and more) is detected and pre-filled when you upload, paste, or load from a URL. Workbooks with more than one sheet show which sheet is being read and list the others in a dropdown (Excel to CSV). Text files saved in older encodings such as Windows-1251 or Shift_JIS are detected more reliably, so accented and non-Latin characters come through intact instead of as garbled symbols (CSV to Excel).

Data Prep grew into a workbench

The Data Prep tool launched in May as one place to view, clean, transform, and convert data without switching between tools. Nested values show as an interactive tree or as formatted JSON, and the steps you add stack on top of each other until the output looks right.

Two combine steps arrived in July. Join Datasets matches rows with a second uploaded dataset on the columns you choose, with inner, left, right, and full joins to control which unmatched rows are kept. Merge Datasets stacks datasets underneath each other, with control over which column from each dataset fills each output column, which suits exports from different periods, regions, or systems.

November added three ways to keep only the rows you want: Filter by Value for text, number ranges, dates, true/false, and regular expressions; Remove Blank Rows; and Filter by Expression for and/or logic, comparisons between columns, and calculated conditions.

Nested JSON and XML got two improvements in July. A smarter structure detection engine groups similar repeating structures together, producing cleaner columns with less manual configuration. The new Cascade option repeats a field's last-seen value on records that do not have their own, which is handy for parent values such as an order number or a customer name (JSON to Excel).

Batch jobs

Converting one file at a time does not scale to a folder of statements. Since November, paid monthly plans can upload up to 100 files, or a zip of them, to Batch, convert them all to one format, and download the results together. PDFs and JPG, PNG, and WebP images are included, so a month of invoices, receipts, or scanned tables can be extracted in one job. Batch processing is listed on the pricing cards, and each job can be followed from your account while it runs.

Document extraction from PDFs and images

Document extraction reads tables from PDFs and from JPG, PNG, and now WebP images, and returns the data as a dataset you can convert, view, or clean like any other source. In 2026 it gained WebP input and batch jobs, and it anchors the new use-case guides for bank statements, invoices, receipts, and shipping documents. Page limits and daily allowances depend on your plan; the pricing page lists them.

For developers

The REST API converts files from scripts and pipelines, and since August it also runs your saved templates: apply a template to an uploaded file, a URL, or a saved dataset in one request, with a ready-to-copy cURL command and code samples on each template's page. The Swagger UI and the Postman collection make the endpoints easy to try before you write any code. Saved datasets and templates have their own pages in the account menu, so the same cleanup runs the same way every month.

What is next

Several formats are planned and already have their own pages, such as CAMT.053 bank statements; each page takes sign-ups so we can tell you when the converter is ready. If your team receives a file type that is not here yet, ask us and tell us what you do with it today.

This post opens the DataConverter.io blog. Upcoming articles explain the formats above one at a time, with sample files and the fields you get, and walk through common cleanup jobs in Data Prep. Follow along through the RSS feed, or check the What's New page for the dated list of changes.