How well it reads
Measured on every release, on the same test sets, for every category.
- Precision
- 96.7%
- Coverage
- 72.5%
- F1
- 81.3%
- Payee named
- 44.0%
On 963 transactions it never saw, measured 27 Sept 2026. Each category is held to 95% precision; one below it on a test set is marked.
Test sets
| Test set | Precision | Coverage | Recall | F1 | F1 since first release |
|---|---|---|---|---|---|
| Transactions it never saw | 96.7% | 72.5% | 70.1% | 81.3% | +13.4 pts |
963 labelled transactions from public datasets, held out while the rules were written. The labels are the datasets’ own. Payee named on 44.0%. Measured 27 Sept 2026. | |||||
| Real payee names | 99.6% | 48.5% | 48.4% | 65.1% | +32.7 pts |
1,102 Australian businesses from public registers and map data, each with a right answer. Payee named on 34.8%. Measured 27 Sept 2026. | |||||
| The bank’s own words not independent | 100.0% | 79.0% | 79.0% | 88.3% | +34.3 pts |
681 lines from 14 banks. The rules were written from them, so this checks nothing was lost; it does not measure accuracy. 46 lines that should not be read were left unread. Measured 27 Sept 2026. | |||||
| BPAY billers not independent | 99.5% | 64.0% | 63.7% | 77.6% | +71.1 pts |
6,217 billers’ short names as a bank prints them, against their full registered names. Agreement, not correctness. Measured 27 Sept 2026. | |||||
- Transactions it never saw 81.3%
- Real payee names 65.1%
- The bank’s own words 88.3%
- BPAY billers 77.6%
By category
F1 and the number of lines in each test set. Open a category for precision, coverage and recall, and for what the wrong answers were labelled.
| Category | Transactions it never saw F1 · lines | Real payee names F1 · lines | The bank’s own words F1 · lines | ||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Salary or wages | 100.0% · 6 | — | 92.6% · 29 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Government benefit | — | 82.9% · 24 | 100.0% · 9 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Other income | 71.4% · 8 | — | 63.9% · 66 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Rent | 100.0% · 4 | 47.1% · 26 | 92.9% · 15 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Mortgage repayment | 80.0% · 3 | 97.6% · 21 | 66.7% · 2 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Strata and body corporate | — | — | — | ||||||||||||||||||||||||||||||
Not measured: Owners corporations are in no public register, so there is nothing independent to test against. The words were checked against 715 strata billers’ names, the same names they were written from: a check, not a measurement. | |||||||||||||||||||||||||||||||||
| Utilities below 95% | 72.0% · 15 | 88.7% · 84 | 96.3% · 14 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Groceries | 84.6% · 183 | 85.2% · 31 | 95.8% · 25 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Transport and fuel | 95.0% · 136 | 37.0% · 225 | 97.2% · 37 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Insurance | 100.0% · 8 | 77.1% · 66 | 95.2% · 22 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Health | 83.3% · 42 | 63.3% · 41 | 84.9% · 19 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Education and childcare below 95% | 80.0% · 3 | 75.9% · 17 | 80.0% · 12 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Phone and internet | 100.0% · 4 | 66.7% · 6 | 100.0% · 20 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Loan repayment | — | 69.4% · 113 | 88.9% · 10 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Credit card payment | — | 100.0% · 4 | 85.7% · 16 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Buy now pay later | 100.0% · 3 | said 2 | 100.0% · 9 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Eating out and takeaway below 95% | 74.4% · 221 | 68.5% · 190 | 83.0% · 31 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Shopping | 59.0% · 120 | 62.1% · 193 | 75.0% · 55 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Entertainment and subscriptions below 95% | 69.9% · 77 | 87.5% · 9 | 96.2% · 27 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Travel | 16.7% · 11 | 66.7% · 8 | 76.2% · 13 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Gambling | — | 100.0% · 1 | 100.0% · 3 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Cash withdrawal | 98.5% · 33 | 37.7% · 43 | 90.3% · 34 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Transfer | 96.1% · 80 | — | 88.0% · 140 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Paid to or from a person | — | — | said 22 | ||||||||||||||||||||||||||||||
Not measured: People are in no public register, by design. Read from the bank’s own wording and a given name, and checked on the bank-wording lines and on transfers between common Australian names. | |||||||||||||||||||||||||||||||||
| Bank fees and interest | 90.9% · 6 | — | 95.7% · 60 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
| Tax, rates and government charges | said 3 | — | 100.0% · 13 | ||||||||||||||||||||||||||||||
Precision, coverage and recall
| |||||||||||||||||||||||||||||||||
Releases
Three categories and BPAY 27 Sept 2026 · a72888b
Tax and government, paid to a person, and strata. BPAY billers by code and short name. Car finance.
Transactions it never saw F1 81.3% +2.8 ptsReal payee names F1 65.1% +3.5 pts
Kinds of business 27 Sept 2026 · 9c4b485
Kinds of business mined from real venue names. A transfer under the holder’s own name is their own money.
Transactions it never saw F1 78.5% +5.8 ptsReal payee names F1 61.7% +3.6 pts
The bank’s own words 27 Sept 2026 · 6c6f733
Fees, cash, transfers, pay and benefits, read from how 14 banks write them. Licensed bookmakers and debt collectors.
Transactions it never saw F1 72.7% +3.2 ptsReal payee names F1 58.1% +3.5 pts
Australian brands 27 Sept 2026 · 8e4b79f
Brands from a national index of shop names, one name for each merchant.
Transactions it never saw F1 69.5% +0.7 ptsReal payee names F1 54.5% +10.4 pts
Enrich 20 Sept 2026 · 67c74e1
Each reading names the payee and says how sure it is. Energy retailers and lenders named from public registers.
Transactions it never saw F1 68.7% +0.9 ptsReal payee names F1 44.2% +11.7 pts
First release 19 Sept 2026 · 413245b
A bank export read into income, commitments and spending.
Transactions it never saw F1 67.8%Real payee names F1 32.4%
What the figures mean
- Precision
- Of the lines Statement put in a category, the share that were right.
- Coverage
- The share of lines Statement put in any category. The rest it declines.
- Recall
- The share of lines put in the right category.
- F1
- Precision and recall together, as their harmonic mean.
- Payee named
- The share of lines where the business was identified by name.
AUC is not reported: it measures how well a score ranks, and Statement answers or declines rather than ranking. The same figures, for a program: /v1/statement/accuracy.