Turn ITR and audited PDFs into a checked Excel, right where they already are.

It reads ITR-3, ITR-5 or ITR-6 PDFs and audited balance sheet and profit and loss statements, then fills an Excel template with the figures. It converts units to Lakhs, ties profit and the balance sheet back to the printed totals, and lists every exception by sheet and cell.

PDF files with a text layer (ITR acknowledgement, Part A-BS, Part A-P&L, audited statements, tax audit reports), scanned PDFs and photos (read with OCR), and Excel or CSV files. You can add several files at once, for example one set per year.

No. Every step runs inside your browser, and no file or figure is sent anywhere. There is no AI service and no account involved.

Pass 1 maps each statement line to a template row. Pass 2 re-finds every number in its source line, rebuilds profit after tax from the extracted lines, and compares mapped assets and liabilities with the printed totals. Differences appear as exceptions. Scanned documents and unusual layouts need more review, so check the figures against the source before relying on them.

No. Only input cells receive values. Formula cells are skipped, and after the file is built the tool compares every formula with the original and shows how many are identical. Calculated values are refreshed so the file opens with numbers, and Excel recalculates on open.

It is left exactly as it is and reported in the exception summary with the sheet and cell, for example a formula that refers to a deleted cell, a total that adds the same row twice, or a row marked Auto Populate that has no formula.

The sample template says Financial Year, so that is the default (March 2025 year-end is 2024-25). If your team labels columns by assessment year, switch the option and the headers will read 2025-26 for the same data.

The tool looks for statements such as in Lakhs or Rs. in Crores on each page and converts every figure to Lakhs. ITR forms are in absolute Rupees. If a document does not state its unit, choose it manually in the Units list.

Yes. Pages without a text layer are read with an OCR engine that downloads on first use, so an internet connection is needed for scans. OCR can misread digits, so the balance and profit checks matter most for these files.

Those already in your template: debtors days, average days in inventory, current ratio, debt-equity on net worth and with quasi equity, DSCR placeholder and growth in PAT. The tool fills the inputs and your formulas do the rest.

Some rows, such as payments to related parties or loans and advances taken, rarely appear in the primary statements or are ambiguous. They are left blank and listed so you can complete them from the notes to accounts.

Yes. Share capital also reads proprietor's capital and partners' capital accounts, and the interest and remuneration rows pick up partner and director lines.

Yes. Version 2.0 matches lines by meaning, not exact wording, and copes with side-by-side T-format statements and Indian number formats. Anything it cannot name is listed under 'Lines we could not name'. Choose where a line belongs once and the tool remembers it in your browser for every future file.

Yes. Click the coloured dot beside any figure to see the document, page and the exact source lines. You can also download a source-trace CSV listing every figure with its source line.

It is a quick signal, not a guarantee. It rises when profit and the balance sheet tie to the printed totals and falls when lines are derived, need confirmation, come from scans or fail the checks. Always review the exception summary.

After your first visit, the tool is cached by your browser and opens without a connection. Scanned pages also work offline once the OCR engine has been downloaded one time.

Yes. It is free to use with no sign-up, and it needs no paid services because everything runs in your browser.

No. It speeds up data entry and flags inconsistencies. A qualified person should review the figures and the template logic before any lending or filing decision.