No data leaves your machine — runs locally in your browser
No IT setup required · No vendor onboarding · Partner discretionary authority
15-point diagnostic protocol · Export Exception Summary on every diagnostic
Synthetic data · nothing uploaded Run on your files →
This passed review. Try to walk the factor.
The totals tie. The return is correct.
Now re-derive it from the export.
Does it close — or do you have to explain it?
Start here

Tab 1 shows the problem. Tabs 2–5 show why it happens.

Most firms validate that the factor ties. Few validate that it can be walked without rebuilding it. These scenarios show what that gap looks like — and what it costs when it surfaces after sign-off.

CCH Axcess · Multistate Apportionment Export · Factor Integrity Check

The factor tied.
The build didn't foot.

The factor ties to the return. But numerator ÷ denominator doesn't match the reported factor. The difference is 8.3 basis points — above the 5 bps threshold (configurable; separates rounding noise from structural gaps). No override or rounding explanation is visible in the export. The build can't be walked from the export alone.

This isn't an error.
It's a documentation condition.

The number is right. But the workpaper doesn't close on its own. If a state examiner asks you to walk this factor from source to return — or someone tries to re-derive a prior year position — this file requires explanation that isn't in the export.

Where this surfaces after sign-off
The factor foots to Schedule R. Standard review passes. The condition only surfaces when someone tries to re-derive the factor from its components — in extension prep, a client meeting, or a state exam. That's when a 5-minute question turns into hours of tracing on a file that already cleared review.
apportionment_Q4_2025.xlsx 742 rows · multistate apportionment
Columns: Entity · State · Numerator · Denominator · Factor
① Looks clean ② Re-derive the factor ③ Gap appears ④ Why it happens Step 1 of 4
#EntityStateNumerator ($)Denominator ($)Factor (Col AP)Ties to return?
R11TechCorp LLCCA$3,421,000$18,750,00018.3286%
R12TechCorp LLCNY$2,100,000$18,750,00011.2000%
R13TechCorp LLCTX$1,875,000$18,750,00010.0000%
✓ Factor ties to Schedule R · Cross-footing passes · Standard review complete
Step 1 of 4
◆ Verify this yourself — 60 seconds No tool required
If your export has separate numerator, denominator, and factor columns:
  1. In Excel: =Numerator / Denominator for any entity-state row.
  2. Does the result match the reported factor — or only after explanation?
  3. If not: the build isn't fully traceable from the export. That's what this documents before sign-off.
CF-15 runs this check across every entity-state pair in the file in under 90 seconds and documents the findings in a workpaper-ready output. It applies only to exports where numerator, denominator, and factor are present as separate columns. If those columns aren't in your export, CF-15 is skipped and the other 14 checks still run.
Apportionment Export · Synthetic data · 742 rows
The number is right. The workpaper doesn't close on its own. See what the diagnostic finds — and what it produces for the workpaper.
Running structural diagnostic…
Analyzing synthetic export data
0% analyzed
Parsing file structure
Detecting vendor format fingerprint
Running sign consistency analysis
Running entity collision detection
Running jurisdiction code validation
Computing Data Consistency Score
Generating Export Exception Summary
Exceptions
Scan complete — exceptions require review
This synthetic scenario demonstrates how the diagnostic surfaces structural errors that pass every review layer.
Scan ID  DIAG-7F3A1B22
SHA-256  9b81c2f7…f2d1
Rows  742
Source  CCH Axcess
74
/ 100
C+
Data Consistency Score · 4 dimensions
⚠ EXCEPTIONS IDENTIFIED — PARTNER REVIEW REQUIRED
1 critical · 2 flagged · Material exceptions require partner review before sign-off.
Structural61/100
Calculation68/100
Jurisdiction72/100
Consolidation82/100
Rows scanned742
Findings3
Critical1
Flagged2
Avg. confidence96%
Est. fix time13–17 hrs
Resolution pathTrace to source system
View: All findings · full detail · resolution status per item
Partner View — Consequence Summary
Data Consistency Score: 74/100 — Below Threshold  ·  Partner sign-off requires review of all critical items  ·  Attach Export Exception Summary to workpaper file
1 critical · 2 flagged
What this means on a real engagement file
If this were your file, not this synthetic scenario:
The return would still file correctly
The factor would still tie to Schedule R
Standard review would still sign off
The factor cannot be re-derived from the export alone
No explanation exists in the workpaper
Walking it back requires going to the source system
That's the gap this diagnostic documents
Whether the findings are material or not, you now have a documented Export Exception Summary showing that structural validation was performed before sign-off.

A clean result isn't nothing. It's workpaper evidence that the check ran — the audit trail that currently doesn't exist anywhere in your review process.
Signs Apportionment factor sign inversions identified ⚑ Flagged — verification required at source
Entities Entity name variants identified — canonical form applied ✓ Normalized in cleaned export
Structure State labels normalized to USPS codes ✓ Normalized in cleaned export
Codes Unmapped account codes identified ⚑ Flagged — manual jurisdiction assignment required
Values No numeric values modified. All original values traceable to source. ✓ Untouched — provenance preserved
Validation Workpaper Note — Copy / Paste Ready
Export integrity check performed: The provided file was run through a structural reliability scanner (Diagnostic ID: DIAG-7F3A1B22, SHA-256: 9b81c2f7…f2d1). No PII was detected or retained. RESOLVED ITEMS: CF-13 — Same Entity, Same State, Opposite Signs — RESOLVED in cleaned export. Canonical form applied per frequency analysis. Original variants preserved. CF-03 — State Code Outside USPS Reference List — RESOLVED in cleaned export. Standardized to USPS 2-letter codes per 50-state + DC lookup table. Original labels preserved. FLAGGED ITEMS (require human review): CF-01 — Sign Inconsistency Among Peer Values — FLAGGED, NOT CORRECTED. Original values preserved. Requires verification of sign convention at source system before reliance. CF-04 — Text-Format Currency Excluded from SUM — FLAGGED, NOT CORRECTED. Currency conversion required before reliance. Output auto-deleted within 48 hours. Entity labels are harmonized for comparison only; elections, attributes, and numeric values are never inferred or altered. Structural screening only — no tax advice provided.
Attach Export Exception Summary PDF for full workpaper documentation.
Preparer Attestation — Workpaper Sign-Off
Reviewed By
 
Date
 
Title / Role
 
Engagement
 
Structural screening only — not a tax opinion. This report identifies data format and consistency conditions; it does not evaluate the correctness of reported amounts or tax positions. All determinations regarding tax positions, entity classification, and sign convention correctness remain with the reviewing professional.
In real scans · this is what your team sees

Want to show this to a colleague? Forward an anonymized version.

When running on real client files, you can generate an anonymized version of the Export Exception Summary — entity names blanked, dollar values scaled — and forward it from your own email with the templates below. Zero client-identifying data, safe to share firm-wide. The diagnostic itself stays in the partner's hands; what spreads is the conversation about whether export integrity should be a standard pre-sign-off step.

↓ Anonymized Export Exception Summary
Generate a peer-shareable PDF.
Identical structure to the full Export Exception Summary, with all entity-identifying data redacted. Numeric values scaled to preserve the shape of findings without revealing actuals. Safe to forward firm-wide.
What's redacted on real scans: entity names → "Entity 1, Entity 2, …" · row-level dollar values → scaled to nearest $10K · client identifiers removed · CF code findings preserved · score and grade preserved.
📧 Email forward — peer review framing
Subject: Take a look at this — pre-sign-off diagnostic on one of our SALT files [Colleague], I ran one of our recent SALT exports through a structural diagnostic this week. It's a 90-second pre-sign-off check — caught a few conditions that survived our standard review (sign inconsistency, schema-type violations, jurisdiction code issues). Worth a look — I've attached an anonymized version of the report so you can see the structure without any client data. The diagnostic itself runs in-browser, no upload, no IT approval needed: financeocr.com/verify [Your name]
Best for: Senior managers, fellow SALT partners, the tax director who would approve adoption firm-wide.
💬 Slack / Teams message — quick share
Ran one of our SALT exports through a 90-second structural diagnostic. Caught a few conditions that survived our normal review — factor gaps at the component level, sign inconsistency, jurisdiction code issues. Anonymized version of the report attached. Browser-only, no upload: financeocr.com/verify
Best for: Internal Slack/Teams channels, quick FYI to a colleague or two.
◈  In-Browser Processing
Your data never leaves your machine.
Diagnostic executes in-browser via WebAssembly. Zero transmission to external servers. Verifiable in your browser's Network tab.
🔐  SHA-256 Provenance
Cryptographic integrity on every file.
A SHA-256 hash of your file is computed locally for workpaper integrity. The hash proves what the file contained at scan time. No file content is retained anywhere.
🔍  Structure Only
We read structure — not substance.
Entity labels are harmonized for comparison only. Numeric values are never inferred or altered. No tax advice. Professional judgment remains with your firm.
Verify the security claim yourself — 60 seconds Right-click → Inspect → Network tab.  Run a scan. Observe zero outbound POST requests. The file never leaves your browser.
No trust required · Architecture, not policy
◆  The real question

Seen on synthetic data.
What does your own export show?

The number in this scenario is right. The workpaper doesn't close on its own. The only answer that matters is whether the same condition exists on your own exports. The scan runs the same way on real files — in-browser, zero transmission, 90 seconds each.

How the real test works
01
Upload 3–5 recent exports
Apportionment schedules, provision exports, multi-entity rollups. Any mix of CCH, GoSystem, ONESOURCE, or Corptax.
02
Each file processes in 90 seconds
In your browser, via WebAssembly. Nothing transmitted. Your IT team can verify zero outbound traffic in the Network tab.
03
Receive an Export Exception Summary per file
Workpaper-ready PDF with Data Consistency Score, findings, and preparer attestation section. Attach directly to the engagement file.
04
Decide based on your own data
Forward the summaries to your SALT director. If findings are material, engage the pod license. If files are clean, you have documented workpaper evidence that structural validation was performed — and you walk away knowing.
Discovery Phase — No Cost
No calls. No sales process. This is a file-level decision.
Test it on your own exports.
Three to five files. Ninety seconds each. You'll have concrete evidence for or against in under ten minutes — and an Export Exception Summary for every file you can forward internally.
Next step: run on 3–5 of your recent files →
If your files are clean: you receive documented Export Exception Summaries showing structural validation was performed. That documentation has value — it's the audit trail that didn't exist before. If our output isn't something your team would attach to a workpaper, you don't pay.
No IT ticket
No procurement
Zero retention
InfoSec review required first?  Download the architecture memo.
Before you run your own files

Three questions worth answering first.

01 Does this integrate with my tax software directly?
No — and that's deliberate. The diagnostic runs on exported files, not on live connections to CCH Axcess, GoSystem RS, ONESOURCE, or Corptax. No API keys, no IT coordination, no firewall exceptions, no vendor integration certification. Your team exports as they already do for review, runs the diagnostic in-browser in 90 seconds, and generates a workpaper artifact before sign-off. A direct integration would require months of enterprise procurement. Export-based review starts today.
02 What actually happens to the file when I upload it?
Nothing is uploaded. The diagnostic runs entirely in your browser via WebAssembly — the file is processed in-memory and never transmitted to our servers. This isn't a policy promise — your IT team can verify it in 60 seconds. A SHA-256 hash is computed locally for workpaper provenance, but no file content leaves your machine.
03 What happens after I run my own files?
You receive a scored Export Exception Summary for each file — workpaper-ready, formatted for risk committee review, plus a Data Consistency Scorecard and a Corrected Workpaper Reference (Excel). If findings are material, forward the summaries to your SALT director and review the engagement page for next steps. If the files are clean, walk away at no cost. Walk-away condition: no critical findings and no material exceptions means no engagement.
More questions? See the full FAQ  ·  hello@financeocr.com
Scope & Limitations Structural screening only. Entity labels are harmonized for comparison; elections, attributes, and numeric values are never inferred or altered. No tax advice, nexus determinations, or filing recommendations provided. Professional judgment remains with your firm.