A display rule is written like a validation rule and applied to a cell: amber when a supplier's late rate passes 5%, a purple bar when the value came from a steward, a dotted underline under a low-confidence value, a mask for a role that may see only part of a number.
| supplier | legal name | late rate | payment terms | bank account | open POs | confidence |
|---|---|---|---|---|---|---|
| L-00731 | Acme Components GmbH | 2.1% | Net 45 | •••• 4471 | 62 · €1.2M | 0.97 |
| L-00912 | Delta Fasteners Ltd | 7.4% ▲ | Net 30 (steward) | •••• 0088 | €462K over cap | 0.91 |
| L-01120 | Brightline Metals SA | 4.8% | Net 60 | hidden | 18 · €210K | 0.68 · review |
The left border of a cell shows its provenance; the background shows a rule result; the bank column shows the mask this role gets; a low-confidence cell is italic with a note that a review is open.
| Rule | Effect | Used on |
|---|---|---|
| late_rate > 5% | amber background, warning icon, note with the count | supplier grids and pivots |
| source = steward edit | purple left bar; hover shows who and when | every golden view |
| confidence < 0.7 | italic, dotted underline, 'review open' note | recently merged records |
| expires within 90 days | gold icon and the date | certificates, contracts, rates |
| value differs from source | blue dot; hover shows the source value and the survivorship rule | coexistence domains |
| role lacks read | masked or hidden, with the policy name on hover | bank, credit, personal data |
Personal display rules are allowed on personal views; shared views carry the domain's rules, versioned and approved like any configuration.
They are evaluated in the engine with the query, so the cost is one expression per cell, and windows stay small.
The export carries the rule as metadata and a colour field; the dashboard binds to it.
Bring two systems that hold the same customers, suppliers or materials. In one working day we connect them, run the match on live rows, review the queue with your steward and publish the first golden view with its API.
