Preserves statement metadata and balances with exact values, including bank_name "Standard Chartered", statement_date "16 Jul 2019", currency "INR", opening_balance 114453.65, and closing_balance 116149.46.
What was measured
Extraction Accuracy
Are field values correct, complete, and free of OCR or parsing errors, including numerical precision on financial fields?
decisive for this rankingtransformation
Correct field values are the core of the job; wrong or incomplete extraction means the tool failed to retrieve the structured data from the document. (3 of 3 judges)
What was given, what came back
Test input: Bank Statement PDF · pdf · group: business-document-extraction
Input — what we sent
Research media bank statement 2 jul.png
Bank Statement PDF
A four-page bank statement PDF with dense transaction tables, balance-forward bridges, account metadata, rewards data, and disclaimer text. It was used to stress schema-driven extraction, multi-page continuity, row completeness, and financial numerical accuracy.
Why this input is hard
- · Table extraction across 50+ transaction rows
- · Multi-page continuity with balance-forward bridges
- · Parsing structured account metadata alongside unstructured transaction descriptions
- · Numerical accuracy for deposits, withdrawals, running balances, and summaries
- · Extraction of nested rewards and disclaimer sections
Output — unretouched
Also checked on this input — same tool, 5 other criteria
Schema Adherence✓ WorkedReconstructs the supplied bank-statement hierarchy instead of flat OCR, with nested statement.metadata, account_holder.address, account, branch, statement_period, balances, transactions, summary, rewards, and disclaimers objects present in the extracted JSON flow.Semantic Field Enrichment✓ WorkedClassifies extracted bank rows with semantic transaction_type labels such as Withdrawal and Deposit, showing derived type tagging beyond raw transaction text.Semantic Field Enrichment✗ FailedDoes not preserve source transaction identifiers, instead assigning sequential transaction_id values ("1", "2", "3", ...) to the rows.Structural Clean Output✓ WorkedDelivers JSON that is directly usable downstream, with the demo moving from upload to extraction results and the report noting downloadable output with no extra transformation step.Table & Record Completeness✓ WorkedReturns the full transaction table as separate records; the report says all 51 statement entries were extracted without collapsing rows or dropping records.
Provenance
- Observation
- 5efe531d-6c34-4551-88f5-d36447fed696
- Evidence run
- a061b9e7-a9c5-443d-a171-b296aaf51b8c
- Study
- Extract and query structured data from documents using natural language
- Research task
- 86b9y25e5
- Tested at
- not recorded
- Source
- first-party
- Evidence state
- verified
- Proof shown
- input + output shown
- Cost / latency
- not captured
- Repeat run
- not captured
- Tester
- not captured
The last three rows are honest blanks, not placeholders — our capture has no field for them yet.
Query this
get_evidence({
tool: "landing-ai",
scenario: "business-document-extraction"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 8 other tools
measured on Extraction Accuracy
Datalab⚠ StruggledMisplaces a bank-statement transaction across dates: a row that belongs to 18 Jun is attached to the 19 Jun record after merging, so date association is unreliable.Docsumo◐ MixedIn the rewards scheme table, zero-value cells render as blank rather than as 0, so the extractor preserves the row structure but loses explicit zero values.Extend AI✗ FailedThe summary aggregation is incorrect: `total_transactions` is 49 in the output, while the report says the expected count after exclusions is 40.LlamaParse⚠ StruggledThe tool leaves value_date blank on some bank transactions even though the source statement contains value dates, so transaction metadata is only partially accurate.Nanonets✓ WorkedCorrectly extracts high-level statement values, including account number 42710540422 and total deposits 70986.83, rather than corrupting the header fields.Reducto✓ WorkedCorrectly extracts the account-holder/customer name from the bank statement and attaches a citation bounding box to the source text.Retab✓ WorkedDocsumo extracts the bank statement into structured customer, branch, and summary fields, including the account holder name, address, totals, and closing balance, and also supports QA over the document.Unstract✗ FailedUnderstates the derived bank-summary count: summary.total_transactions is reported as 43 even though the PDF contains 51 transactions, a 16% undercount.
This evidence is published in
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com