Convert bank statement PDF to Excel on Mac (2026 guide)

You opened the folder again. Eight months of checking-account PDFs from one bank, four months of a business account at another, and a credit-union savings statement that you keep forgetting belongs in the same picture. The budget spreadsheet in Excel for Mac has been waiting since March. Numbers tells you what’s happened; Excel is where the pivot tables, the SUMIFS formulas, and the cash-flow chart actually live.

The Mac is the right tool here. The keyboard is fast enough for three hundred rows of spot-checks. The display shows the source PDF and the extracted spreadsheet side by side. And on macOS 14.4+ on Apple Silicon, the on-device AI model reads twelve statements in a minute without sending one transaction line to a cloud service.

This guide is the workflow that gets you from “a folder of bank-statement PDFs accumulated across the year” to “one Excel workbook that drives the cash-flow pivot” without re-typing a transaction, without uploading personal banking data to a free-tier web converter, and without giving up the bank-by-bank, account-by-account structure that makes the pivot table actually useful.

Why Excel (not CSV, not Numbers) for the Mac batch

For bank statements destined for QuickBooks or Xero, CSV is the right output. For a personal sheet you live inside Numbers, .numbers works. For a Mac user whose financial analysis lives in Excel — pivot tables, conditional formatting, SUMIFS by category and month, a cash-flow chart with a 12-month trailing axis — XLSX is what the workflow actually wants. Three concrete reasons:

Multi-sheet structure matches the data. A year of statements has line-level transactions and account-level metadata (opening balance, closing balance, statement period, account number masked to last four). CSV is single-table. XLSX with a transactions sheet and an accounts sheet joined on account_last_four plus statement_period_start is how the data is actually shaped, and it’s what the pivot table needs.
Pivot tables survive the handoff. “How did month-over-month cash flow change in the operating account?” is a pivot grouped by account_last_four × month × transaction_type. “Which recurring vendors hit which account, and on what day of the month?” is a pivot on merchant_normalized × day_of_month. CSV is just rows. XLSX preserves the pivots, the formulas, and the conditional formatting between sessions.
Formula columns matter. Once you have amount, you want a running_balance formula, a month derived column, a vendor_category lookup against a small reference sheet. Excel for Mac handles all of those natively with full M1/M2/M3 acceleration; rebuilding them in Numbers after a CSV round-trip is a tax you don’t need to pay.

If your destination is QuickBooks Online, switch to the CSV-on-Mac path — the importer there prefers CSV and the column mapping fights are smaller. For everything else that ends in Excel for Mac, XLSX is shorter.

Why the Mac is the right device for this

The iPhone bank-statement Excel walkthrough is for the one statement that arrives in Mail while you’re not at the desk. The iPad CSV walkthrough is for the couch-side monthly. The Mac is for the batch — the half-year you fell behind on, the year-end consolidation, the cash-flow analysis that spans three accounts and twelve months.

Three reasons the Mac wins for that case:

Folder semantics. A year of statements arrives in Mail attachments, downloads from each bank’s website, and the credit union’s “secure messages” PDF download. macOS lets you stage them in one folder, organized by bank and month, and drag the whole thing into ignitai at once. iPad multi-select tops out before forty files.
Big screen for the spot-check. A 27-inch display shows the source PDF on the left, the extracted rows in the middle, and the pivot table on the right. Three hundred transaction rows is twenty minutes of scroll-and-verify on Mac.
macOS 14.4+ on Apple Silicon runs the on-device model fast. A typical 12-page statement extracts in 3–6 seconds on an M2 or newer; a year-long batch of twelve statements per account across three accounts finishes in under three minutes. The fans don’t spin up — the Neural Engine handles it cool.

If your Mac has a T2 chip or earlier (Intel-era), the on-device path isn’t available; ignitai falls back to a hosted pipeline with documented zero retention. For Apple Silicon Macs (M1 onward) on macOS 14.4+, the entire batch runs locally.

Why a bank-statement batch isn’t just any PDF batch

A generic “PDF to Excel” tool treats every PDF as a table-grid extraction job. Bank statements have four extraction problems that a generic tool will get wrong:

Header noise. Page one has a logo, an address block, an account summary, marketing copy. The transaction table doesn’t start until midway down, sometimes not until page two. A naive extractor grabs everything and the workbook has ten rows of garbage at the top of every imported statement.
Running balance reconciliation. Most bank statements include a running balance column. The numbers in that column carry information the rest of the table doesn’t — they’re the proof that no transaction was missed, duplicated, or misread. A tool that drops the balance column makes the most useful reconciliation check impossible.
Sign convention by row type. Money out is negative. Money in is positive. Some statements use parentheses for negatives, some use a leading minus, some put debits and credits in separate columns. A multi-bank batch needs the output to be one signed amount column, normalized across all source layouts.
Multi-line transactions and continuation pages. A wire transfer can wrap across two physical lines; a long merchant description plus a foreign-currency reference becomes two lines visually but one transaction logically. A continuation page can restart with a header row. Mashing these incorrectly creates ghost rows that throw off every sum downstream.

Method 1: ignitai folder batch on macOS (the main path)

ignitai handles a bank-statement batch the same way it handles an invoice batch on Mac — drag the folder, write the prompt once, get one consolidated workbook. The Mac flow:

Stage the folder. Create ~/Documents/Bank-statements/2026/. Inside, one subfolder per account: Operating-checking/, Personal-checking/, Savings/. Drop each month’s PDF into the corresponding subfolder. Filename convention: 2026-01-operating.pdf, 2026-02-operating.pdf. If your filenames are still the bank’s defaults (StatementDoc_4892_20260131.pdf), the extraction still works, but renaming makes the audit trail in the spreadsheet readable later.
Open ignitai. Drag the parent folder onto the window. macOS 14.4+ on Apple Silicon (M1 onward) runs the on-device model — no statements leave the Mac. ignitai walks the directory recursively, so the subfolder structure is preserved as metadata on each row.
Describe what to extract, in plain English. For a multi-account annual consolidation:

“For each statement PDF, return one row per transaction with: transaction_date (ISO 8601, from the transaction line), post_date (ISO 8601, if different from transaction_date), description (the original merchant or payee string), merchant_normalized (the canonical name — collapse ‘AMZN MKTP’, ‘AMAZON MARKETPLACE’, ‘AMZN’ to ‘Amazon’; collapse all variants of one utility company to one name), category (one of: payroll, transfer, payment, deposit, ach_debit, ach_credit, check, wire, fee, interest, atm, card_purchase, other), amount (positive for credits and deposits, negative for debits and withdrawals), running_balance (the balance after this transaction, as printed on the statement), check_number (if it’s a check), account_last_four, bank_name, statement_period_start (ISO 8601), statement_period_end (ISO 8601), source_file (the original filename). Skip account summary rows, marketing inserts, and the running-balance restart rows on continuation pages. In a separate sheet, return one row per statement with: bank_name, account_last_four, statement_period_start, statement_period_end, opening_balance, closing_balance, total_deposits, total_withdrawals, total_fees, average_balance (if shown).”*

Save the prompt as a preset called “Annual bank consolidation — Excel.” Next year it’s one click.
Pick XLSX as the output format. Two-sheet workbook: transactions for the line-level rows, accounts for the statement-level metadata. The two sheets join on account_last_four plus statement_period_start. Currency formatting auto-applied to amount columns, ISO dates parsed as Excel date cells, category as a text column with a data-validation dropdown matching the prompt’s category list.
Hit Extract. Three accounts × twelve months = thirty-six statements. On an M2 Mac, the whole batch processes in about two and a half minutes. Each statement runs through the on-device model in 4–7 seconds. The progress bar shows file-by-file completion.
Open the output in Excel for Mac. Around four hundred transaction rows depending on activity; thirty-six account-statement rows. Run the four reconciliation checks below before you trust the totals.

End to end on a year’s worth of statements across three accounts: about forty-five minutes of human time, most of it the spot-check phase.

Method 2: pdfplumber + pandas (the DIY path)

If you maintain Python locally and your statements all come from one bank with a stable layout:

brew install python
pip install pdfplumber pandas openpyxl

Then a script that walks the folder, extracts each PDF’s table, normalizes the columns, and writes a multi-sheet XLSX:

import pdfplumber, pandas as pd
from pathlib import Path

transactions, accounts = [], []
for pdf_path in Path("./statements").rglob("*.pdf"):
    with pdfplumber.open(pdf_path) as pdf:
        for i, page in enumerate(pdf.pages):
            for table in page.extract_tables():
                for row in table:
                    transactions.append([*row, pdf_path.name, i + 1])

pd.DataFrame(transactions).to_excel(
    "bank-2026.xlsx", sheet_name="transactions", index=False
)

This is fine for one bank, one stable layout, all text-based PDFs. It breaks the moment the bank redesigns its template (column order shifts silently), one statement is a scan (no text returned), a transaction wraps two lines (two rows in your output), or the bank uses separate Debit and Credit columns (needs a per-bank post-processing step). For one-bank stability, the script is a one-time cost. For real bookkeeping across multiple banks, the maintenance burden eats the savings.

Method 3: Excel for Mac “Get Data from PDF”

Excel for Mac’s Data → Get Data → From PDF uses Power Query, which uploads the PDF to Microsoft’s servers for the table-extraction pass — same privacy posture as a generic web converter, just from a more reputable vendor. It also treats each statement as a generic tables document: you get a description column and nothing else. No categorization, no normalization, fees and interest as separate detected tables you have to manually merge. For personal banking where the data should stay on the Mac, the on-device path is shorter.

Method 4: bank-native CSV exports (always check first)

Before extracting from PDFs at all, check whether your bank offers a CSV export. Most US and UK retail banks do, buried under different names: Chase, Bank of America, Wells Fargo, Capital One, and the big UK banks all offer 12–18 months of CSV history under Activity → Download. Where bank-native CSV exists, use it and skip extraction.

The PDF-extraction workflow is for the cases the bank doesn’t cover: small-business accounts that only ship PDF, brokerage cash sweeps, foreign banks without an English CSV option, historical statements past the bank’s online retention window, or PDFs your accountant sent you from a system you don’t have access to.

The four reconciliation checks for the Mac batch

Once the XLSX is open in Excel for Mac, four quick checks separate “I have a file” from “I have a workbook the pivot table can trust”:

Running-balance integrity, per statement. Filter transactions to one account_last_four + statement_period_start. Sort by transaction_date. Add a formula column: =previous_running_balance + amount. It should equal the extracted running_balance column row by row. Any divergence means a transaction was missed, duplicated, or misread on extraction. Cross-check against the source PDF for the affected row. This single check catches almost every extraction error.
Statement totals reconciliation. For each row of the accounts sheet, sum amount from transactions filtered to the same account_last_four and statement_period_start, split by sign. The positive sum should equal total_deposits on the accounts sheet; the absolute value of the negative sum should equal total_withdrawals. A divergence on one statement means a transaction got missed during extraction.
Sign-convention sweep. Sort transactions ascending by amount. Every negative number should be a debit (purchase, withdrawal, ACH out, fee, check). Every positive number should be a deposit, refund, interest credit, or ACH in. A positive amount on a row that’s clearly a withdrawal will inflate the deposit total at year-end. If you see one, the prompt didn’t enforce the convention — fix the row, then update the saved prompt for next year.
Merchant normalization spot-check. Pivot transactions by merchant_normalized, sort descending by row count. The top forty rows are your most-frequent payees. Scan for variants that didn’t merge — Trader Joes vs Trader Joe's, an electric utility appearing under three slightly different names, Venmo vs VENMO PAYMENT. Edit the column in place, then update the saved prompt. Three iterations and the prompt is reusable across years.

Skip these four and the year-end cash-flow picture is wrong in subtle ways — a category total under-counts because two payee variants stayed split, a deposit total over-counts because a wire-out landed positive, a quarterly cash-flow trend shifts because a continuation-page header was read as a real row.

Building the cash-flow pivot in Excel for Mac

Once the four checks pass, the pivot is the payoff. The structure that answers the most useful questions in one place:

Rows: month (a =TEXT(transaction_date,"YYYY-MM") derived column) × account_last_four.
Columns: category from the prompt’s enumerated list.
Values: SUM(amount).
Filter: exclude transfer rows so inter-account movement doesn’t double-count cash flow.

That pivot answers, in one view: month-by-month net cash flow per account, which categories swing the most, which months were unusually heavy on a category, and when the operating account started running thin relative to its three-month moving average. Add a sparkline column on the right for visual scan; conditional formatting on the values cell for traffic-light reading.

For a small-business operator, this single pivot replaces what a personal-finance SaaS subscription provides at $14.99/month — without sending the underlying transactions anywhere off the Mac.

When the Mac batch isn’t the right tool

Honest edge cases:

Monthly mid-cycle review. Don’t fire up the batch for one statement. Use the iPhone Excel path for a single PDF that landed in Mail, or the iPad CSV walkthrough for couch-side monthly review.
QuickBooks or Xero as the destination. Switch to the CSV-on-Mac path. Accounting tools prefer CSV; the column-mapping fights are smaller.
Brokerage statements. Positions, lots, dividends, and corporate actions don’t fit the transaction-extraction prompt. Use the brokerage’s tax-document export (often a 1099 in CSV form) and fall back to PDF extraction only for the cash transactions on the side. For the 1099 itself, the 1099 PDF to Excel guide on Mac covers the form-level extraction.
Crypto-exchange “statements.” Most exchanges have a real CSV export under account history. Use that. Extracting from a PDF screenshot of a CSV is a workflow built backwards.
Statements older than the online retention window, scanned from paper. The extraction still works because ignitai routes scans through the OCR-then-extract pipeline; spot-check more carefully because OCR adds a small accuracy hit.

Bottom line

For a Mac user staring down a folder of bank-statement PDFs that needs to become one Excel workbook for the cash-flow pivot or the year-end review: stage the PDFs by account in subfolders, drag the parent into ignitai, apply the saved preset, pick XLSX, run the four reconciliation checks, build the pivot. The whole batch lives on the Mac. A year of transactions across three accounts never leaves the device.

The same app handles the iPhone single-statement extraction, the iPad CSV path, the CSV-on-Mac flow into QuickBooks, and the broader Mac batch across mixed document types. Presets sync via iCloud — write the annual prompt once on Mac and the monthly iPhone reviews next year pick up the same normalization rules automatically.

Get ignitai on the App Store — free download, $19.99/mo unlocks unlimited batch extractions after the 3-day trial.