Standard review confirms the math. Totals foot. Schedule R ties. It doesn't confirm that you can walk the build from components without rebuilding it.
Most firms validate that the factor ties. Very few validate that the build can be explained without rebuilding it.
You don't lose time on returns that fail review. You lose time on returns that pass — but don't explain cleanly when revisited. During extension prep. During a client meeting when the CFO asks why CA doesn't tie to internal numbers. When a state examiner asks you to walk the factor from source to return and the workpaper can't answer independently. The factor is correct. The build isn't always provable.
Emmanuel Kyumba — Five years in state and local tax technology, working at the point where apportionment data moves from tax software into workpapers. The work was validating multi-state exports before they reached the review layer — confirming that entity names were consistent, state codes resolved correctly, factors footed from their components, and nothing was being silently excluded from the rollup. The gap this diagnostic catches surfaced repeatedly in that work: a payroll factor that tied perfectly at the Schedule R level, but when you looked at the entity-state rows individually, a subsidiary's CA factor was written with the wrong sign. The total was unchanged because the offsetting value cancelled it out. Standard review passed. The workpaper showed the factor tied. It just couldn't be supported from the exported components without going back to the software. That's the check that didn't exist. Based in Los Angeles.
LinkedIn →
In this export, SubCo I LLC shows −0.3420 for CA payroll in Column AP — while the parent entity shows +0.3420 for the same entity and period. The numerator and denominator are identical for both rows. Divide either one: $3,420,000 ÷ $10,000,000 = +0.3420. The subsidiary's factor should be +0.3420. The export wrote −0.3420. The values cancel in the column total. Schedule R ties. Standard review passes. The sign inversion is only visible when you divide the exported numerator by the denominator and compare the result to the reported factor.
| # | Entity_ID | Entity_Name | State | Entity_Type | Numerator ($) | Denominator ($) | Col AP (Factor) |
|---|---|---|---|---|---|---|---|
| R11 | TC-001 | TechCorp LLC | CA | Parent S-Corp | $3,420,000 | $10,000,000 | +0.3420 |
| R12 | TC-001 | TechCorp LLC | NY | Parent S-Corp | $2,180,000 | $10,000,000 | +0.2180 |
| R13 | TC-001 | SubCo I LLC | CA | QSSS Sub | $3,420,000 | $10,000,000 | −0.3420 |
| R14 | TC-001 | SubCo I LLC | NY | QSSS Sub | $2,180,000 | $10,000,000 | +0.2180 |
| R52 | TC-001 | SubCo II LLC | CA | QSSS Sub | $1,870,000 | $10,000,000 | −0.1870 |
| Total | All entities · CA column | +0.2490 | |||||
$3,420,000 · denominator $10,000,000. Re-derive: 3,420,000 ÷ 10,000,000 = +0.3420. Reported factor: −0.3420. The arithmetic produces the correct magnitude — but the sign is inverted. The factor can be re-derived in magnitude, but not in direction. This is the specific signature of a QSSS consolidation method mismatch: the software wrote the subsidiary's factor with the opposite sign.CF-15 (Factor Arithmetic Integrity) is the anchor check. It re-derives the apportionment factor directly from the export's own numerator and denominator columns and compares the result to the reported factor. If they don't match within a defined tolerance band, it flags the variance and documents the most likely cause.
CF-01 through CF-14 document the structural conditions that cause a factor gap. They are organized into two tiers: Critical (conditions that prevent re-derivation or produce materially wrong factors) and Flagged (conditions that require documentation before the build path can be confirmed clean).
All checks are deterministic or pattern-based with documented detection logic. No inferences about tax positions. No filing recommendations. Structural conditions only.
The diagnostic runs entirely in your browser via WebAssembly. The file is processed in-memory and never transmitted to our servers. This is not a policy promise — it is a technical architecture fact your firm's IT team can verify in under 60 seconds.
Run the diagnostic on synthetic data first. See CF-15 fire. See the four-step reveal. Then execute on your own exports and see what it finds on real files.
If your extension filings are in progress and you can't say with certainty that every entity-level factor can be proven from the exported components — this is a 90-second way to document the answer before the partner signs. Access is immediate. Run it on an extension workpaper today.