Neptune Data / Applications / K-1 Extraction Suite
Upload whole Schedule K-1 (Form 1065) packages — faces, supplemental statements, Schedule K-3, state equivalents — and receive a reconciled, review-ready workbook, an interactive data room where every figure traces to its page, an entity map and an analyst who knows the data.
What goes in
A K-1 package is 40–60 pages and most of it is noise. The suite reads what matters and provably ignores what doesn't.
The rules
Four hard rules govern every run — encoded from how experienced preparers check the work.
Every box's components must sum exactly to the printed total — on the face, or the statement's own TOTAL line. No component ever stands in for a whole box.
K-3 foreign amounts are the sum of the foreign category columns — never the U.S. column, never the grand total. A classic FTC error, eliminated by construction.
Beginning capital + contributions + income + withdrawals must reach ending capital; anything unexplained surfaces as an explicit adjustment line.
Where the package includes an Item L income reconciliation, extracted items must reproduce it before the run is accepted — even $1 items count.
What comes out
| Box | Fund I | Fund II | Totals |
|---|---|---|---|
| 1 · Ordinary income | 812 | 390 | 2,148,300 |
| 5 · Interest income | 340 | 185 | 96,410 |
| 9a · Net LT gain | 275 | 310 | 483,920 |
| 13H · Inv. interest exp. | (420) | — | (31,750) |
Illustrative — all names and figures fictional.
Running at volume
Select the packages already filed to the matter — no re-uploading, no renaming.
Extraction continues if you close the laptop. Progress, per-document results and cost are visible throughout.
Each document's extraction is cached — re-runs only pay for what changed, and cancelling discards cleanly.
One matter holds one run. Clear it and the matter returns to a blank slate — records untouched.
The suite was built for the messy ones — that's the demo worth seeing.
Request a demo