Auto-Grade: A Button, Not a Memo

Shared grading rules keep two testers from turning the same laptop into two different letters. A machine that passed every required check should not become “F2” for one tester and “F3” for another because they remembered the rule differently at four in the afternoon. The auto-grade engine doesn’t replace the tester. It removes the part of the job that depends on remembering which threshold lives where.

What it suggests

The engine suggests two of the four sub-grades: the functional grade, read from the test checklist, and the battery grade, read from the reported battery health. Cosmetic and data grades, and the overall letter, are the tester’s call — the engine does not compute an overall grade, does not weigh dimensions against each other, and does not apply a floor. Higher numbers are better on every scale: F3 is a perfect functional result, F0 does not function.

How the functional rule works

The rules run top-down and the first match wins. A failed test that is marked auto-fail gives F0. All required tests passed and nothing failed gives F3. All required tests passed but an optional test failed gives F2. A partial pass — some passed, the checklist not complete — gives F1. Anything else falls through to F0. The checklist itself comes from the testing template for the category, so “required” means what your organization said it means for laptops, not what it means for monitors.

How the battery rule works

Battery health at or above 80 percent suggests B3, 60 to 79 suggests B2, 30 to 59 suggests B1, and below 30 — or no battery reading at all — suggests B0. A missing battery and a dead battery are the same commercial fact, so they get the same grade.

The tester’s side of the deal

On the Grading tab of the testing detail, the suggestion appears next to the functional and battery fields with an Apply suggestion button for each. One click writes it into the grade; the tester can still type something else. The tester also picks the cosmetic grade from the defects they recorded, sets the data grade, chooses the overall letter and the disposition, and the platform blocks completion until those fields are there. The typing-from-memory step is gone; the decision is not. On a queue of 200 laptops, that’s not a small saving.

Where the criteria live

The rules sit in tenant settings under Core > Grading rules — not in code, not in a Notion doc. That page holds the grade bands for each dimension and the auto-grade rules themselves, and your organization can override the defaults. Edit a rule once, and every tester from that point forward gets the new suggestion. The rule that used to live in the head of the most senior tester now lives where the newest one can read it.