How to import a private loan book from Excel without importing the mess

Migration is a decision, not a copy

Moving a book into servicing software is not a copy-paste. It is a sequence of decisions about which records are authoritative, made before the new book begins. The workbook you are leaving behind accumulated years of one-off fixes, hidden columns, and "I'll clean that up later" cells. The goal of a good import is to bring across what is true and leave the ambiguity behind — not to faithfully reproduce the mess in a nicer interface.

Everything below assumes a current-operating-book import. Paid-off loans, dead deals, and historical documents belong in your source archive unless you have a dedicated historical-record path.

Choose the cutover date

Pick a clean cutover — almost always a month-end. Everything on or before that date is history you carry as opening balances; everything after it is serviced in the new system. A mid-month cutover forces you to reconcile a partial period in two places, which is exactly the confusion you are trying to escape. Board as of the cutover, then run the first full month forward.

Define the book you are importing

Separate the loans you want to service or retain for reporting from records that never became serviced loans. A paid-off or sold loan can still be worth importing when you need its servicing history, payoff record, investor reporting, or year-end records. Any current or completed loan retained in LoanConsole counts as one tracked loan for plan capacity. Voided records and superseded predecessor versions do not count twice.

Separate identifiers from display labels

A loan number is an identifier; "Smith – Maple St flip" is a label. Investors, payments, reserves, and documents all attach to the identifier, so decide the identifier scheme first. If your workbook keyed everything off a borrower name, you will discover the problem the first time one borrower has two loans. Assign stable loan identifiers, then let labels be labels.

Reconcile principal and investor positions

For every loan, the sum of investor principal must equal the outstanding balance. Reconcile this in the workbook first — an import cannot fix a book that does not foot.

If the investor principal on a loan sums to $681,500 against a $680,000 balance, that $1,500 is a real discrepancy — a rounding drift, a missed paydown, or a fat-fingered contribution. Find it now. After boarding, that gap becomes a reconciliation exception on every future close.

Bring opening reserve and receivable balances

Opening interest-reserve balances and any outstanding receivables must come across as starting values, or the first invoice and the first reserve draw will be wrong. A reserve that boards at its original funded amount instead of its current balance will over-draw for months before anyone notices. Board the current funded balance, the draw behavior, and any amount already billed and unpaid.

Validate before commit

Review a reconciliation total — loan count, aggregate principal, aggregate reserves — before you confirm anything. You review the validation results and reconciliation totals before anything commits, and commit is all-or-nothing, so a single malformed row never lands half a book. A file that validates is not a file that is correct: validation confirms the data is well-formed and internally consistent, not that the numbers are economically or legally right. That judgment stays with you.

Preserve the original workbook and run one close in parallel

Keep the source workbook, read-only, as your before-picture. Then run the first month-end in both the workbook and the new system and compare. When the close packet, the investor distributions, and the trial balance tie out to the penny, the migration is done and you can retire the workbook with confidence.

Common failure vs. correct record

The most common failed import is the one that "worked" — every row loaded, no errors — but boarded reserve balances at their original funded amounts and investor shares as static percentages. Three months later the reserves are wrong and a non-pro-rata paydown has silently broken the participation math. The correct record boards current balances, reconciles principal to investor capital, and preserves the identifier relationships so later events attach cleanly.

Where LoanConsole fits

LoanConsole's guided import walks a single workbook through upload, validation, review, and an all-or-nothing commit, with forgiving money and date handling and a reconciliation total before you confirm. You decide the cutover and which records are authoritative; the import preserves what you commit and shows you the totals first.