Bank Statement API

Three calls from PDF to spreadsheet

No start call, no webhook to host and no file size to compute. Polling the statement after the upload is what begins extraction.

1. 1POST/v1/bank-statementsCreate the statement with its filename. The response has the upload_url.
2. 2PUTupload_urlSend the file straight to the presigned URL. Nothing passes through the API gateway.
3. 3GET/v1/bank-statements/{id}Poll until status is completed or failed. The first poll after the upload starts extraction, and a completed statement carries a download url for each format.

Download the file from the link in the completed statement. Links last one hour; GET the statement again for fresh ones. Concurrency: 2 documents at once; 10 with a contract or 10,000+ pages bought.

Code: statement PDF to Excel

Each example reads your key from FASTOCR_API_KEY and writes statement.xlsx. Change outputs.xlsx to any format below. Endpoint details are in the Bank Statement API reference.

export FASTOCR_API_KEY="your_api_key"

# 1. Create a statement
RESPONSE=$(curl -sS --fail-with-body -X POST https://api.fastocr.org/v1/bank-statements \
  -H "Authorization: Bearer $FASTOCR_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{"filename":"statement.pdf"}')
STATEMENT_ID=$(echo "$RESPONSE" | jq -r .id)
UPLOAD_URL=$(echo "$RESPONSE" | jq -r .upload_url)

# 2. Upload the file to the presigned URL
curl -sS --fail-with-body -X PUT "$UPLOAD_URL" \
  -H "Content-Type: application/pdf" \
  --data-binary @statement.pdf

# 3. Poll until the status is completed or failed. The first poll after the upload starts extraction.
while true; do
  BODY=$(curl -sS --fail-with-body https://api.fastocr.org/v1/bank-statements/$STATEMENT_ID \
    -H "Authorization: Bearer $FASTOCR_API_KEY") || exit 1
  STATUS=$(echo "$BODY" | jq -r .status)
  case "$STATUS" in completed|failed) break;; esac
  sleep 3
done
[ "$STATUS" = "completed" ] || { echo "$BODY" | jq .error; exit 1; }

# Download the Excel file from the link in the completed statement
curl -sS --fail-with-body "$(echo "$BODY" | jq -r .outputs.xlsx.url)" -o statement.xlsx

// Node.js 18+, no dependencies. Save as statement.mjs and run: node statement.mjs
import { readFile, writeFile } from 'node:fs/promises';
import { randomUUID } from 'node:crypto';

const BASE = 'https://api.fastocr.org';
const auth = { Authorization: `Bearer ${process.env.FASTOCR_API_KEY}` };

async function api(path, init = {}) {
  const response = await fetch(BASE + path, { ...init, headers: { ...auth, ...init.headers } });
  const body = await response.json();
  if (!response.ok) throw new Error(`${response.status} ${body.error?.code ?? body.message}`);
  return body;
}

const pdf = await readFile('statement.pdf');

// 1. Create a statement
const created = await api('/v1/bank-statements', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json', 'Idempotency-Key': randomUUID() },
  body: JSON.stringify({ filename: 'statement.pdf' }),
});

// 2. Upload the file to the presigned URL
const upload = await fetch(created.upload_url, {
  method: 'PUT',
  headers: { 'Content-Type': 'application/pdf' },
  body: pdf,
});
if (!upload.ok) throw new Error(`Upload failed: ${upload.status}`);

// 3. Poll until the status is completed or failed. The first poll after the upload starts extraction.
let statement;
do {
  statement = await api(`/v1/bank-statements/${created.id}`);
  if (!['completed', 'failed'].includes(statement.status)) await new Promise((resolve) => setTimeout(resolve, 3000));
} while (!['completed', 'failed'].includes(statement.status));
if (statement.status === 'failed') throw new Error(statement.error.code);

// Download the Excel file from the link in the completed statement
await writeFile('statement.xlsx', Buffer.from(await (await fetch(statement.outputs.xlsx.url)).arrayBuffer()));
console.log(statement.status, statement.pages_billed, statement.needs_review);

from fastocr_sdk import FastOCR

# Reads the key from the FASTOCR_API_KEY environment variable.
with FastOCR() as client:
    job = client.extract_statement("statement.pdf")
    client.bank_statements.save(job.id, "statement.xlsx", "xlsx")
    print(job.status, job.pages_billed, job.needs_review)

What a completed statement looks like

needs_review is false when the balance and source checks found no issue. When it is true, compare the result with the statement before you rely on it. Accounting formats list one link per account under files.

{
  "id": "stmt_a1b2c3d4e5f6",
  "status": "completed",
  "pages_billed": 3,
  "needs_review": false,
  "outputs": {
    "json": { "available": true, "url": "https://...", "expires_at": "2026-11-05T13:00:00Z" },
    "csv": { "available": true, "url": "https://...", "expires_at": "2026-11-05T13:00:00Z" },
    "xlsx": { "available": true, "url": "https://...", "expires_at": "2026-11-05T13:00:00Z" },
    "qbo": {
      "available": true,
      "accounts": [1],
      "files": [{ "account": 1, "url": "https://...", "expires_at": "2026-11-05T13:00:00Z" }]
    }
  }
}

Output formats

One balance for the website and the API

The first 50 statement pages are free, whether you start on the website or with an API key. After that you buy a one-time Bank Statement Pack, from $4.99 for 100 pages. There is no subscription. Check what is left with GET /v1/balance. See API pricing.

Frequently asked questions