Financial records reconstruction

A reconstruction engagement covers the full period at issue, rebuilt transaction by transaction from whatever source documents survive. We deliver a reconstructed transaction history, a profit and loss statement and balance sheet for the period, and the assumptions log that documents every judgment call made along the way. Where the reconstruction needs to support a later tax filing, we build it to the standard a credentialed preparer can rely on, with the understanding that the preparer, not us, signs and files any resulting return.

We also include a document inventory listing exactly what survived and what did not, since that inventory is often the first thing an opposing party or an adjuster asks to see. If certain months or accounts have no surviving records at all, we say so plainly in the inventory rather than presenting a reconstructed figure that implies more documentation exists than actually does.

We start with an intake conversation covering exactly what period and entity is at issue, what records the client believes still exist, and what deadline a court, insurer, or filing requires. From there we request bank and card statements directly from the institutions if the client's own copies are incomplete, since most banks will produce several years of statement history on request even after an account has closed.

Reconstruction proceeds statement by statement. Each transaction gets matched to a source document where one exists, an invoice, a contract, a vendor confirmation, or a prior filing, and coded into the rebuilt chart of accounts. Where no document exists, we look for a secondary source, such as a vendor's own records or a pattern in surrounding transactions, before falling back to a documented assumption.

Once the rebuild is complete, a senior reviewer reconciles the reconstructed books to bank totals for the period, and reviews both the financial statements and the assumptions log before delivery. We deliver on the timeline the matter requires and remain available to walk counsel, an adjuster, or a mediator through any specific line.

Reconstruction is engaged most often when a dispute or claim needs financial statements that do not currently exist in reliable form: a partner dispute where one side never saw the real books, a divorce where the family business's records are commingled or incomplete, an insurance claim where the policyholder's own accounting lapsed before the loss event, or a fraud investigation where the person who kept the books is no longer cooperating or no longer employed.

It is not the right service if your books mostly exist and just need to be brought current for this year's tax filing. Our regular catch-up and cleanup bookkeeping work covers that at a lower cost and without the assumptions-log discipline a dispute-grade reconstruction requires. Tell us upfront whether a court, an insurer, or a filing deadline is driving the request, since that changes both the standard we build to and the price.

The most common problem is a client who assumes reconstruction means guessing, and pushes back when we insist on documenting every assumption instead of just producing a clean-looking set of statements. We explain early that a reconstruction with an honest assumptions log survives scrutiny far better than one that looks polished but cannot answer a single hard question about where a number came from.

A second common problem is records scattered across five different systems: a defunct QuickBooks file, a box of paper invoices, a spreadsheet someone kept on the side, and years of bank statements in PDF form. We inventory everything first, before starting the rebuild, so we know what we are working with and do not discover a gap halfway through.

A third problem is a deadline that arrived before anyone realized how much documentation was actually missing. Where that happens, we tell the client immediately what a realistic timeline looks like given what survives, rather than promising a full rebuild by an unrealistic date and delivering something thin.

The rebuilt books live in QuickBooks Online or Xero, whichever the client already uses or prefers going forward, so the reconstructed file is usable after the engagement ends and not locked into a proprietary format. Microsoft Excel handles the transaction-matching working papers and the assumptions log itself, since that log needs to be readable by a non-accountant reviewing it line by line.

Dext helps pull invoice and receipt data out of scanned paper documents faster than manual entry when a client hands us a physical box of records, and we scan and preserve the originals before working from the digitized copies. Where the underlying accounts had a prior QuickBooks or Xero file that still exists but is unreliable, we export whatever transaction history it retains as one more source to cross-check against the bank statements, rather than starting completely from zero.

Reconstruction is quoted per engagement based on the number of periods, accounts, and entities involved, and on how much documentation survives versus needs to be requested from banks or third parties. A single tax year with reasonable bank records available costs meaningfully less than a multi-year reconstruction across several entities with significant gaps.

We give a written estimate after reviewing what documentation the client already has, and we flag upfront if the scope looks likely to grow once we start pulling bank statements and finding more accounts than the client initially remembered, which happens often enough that we build in a checkpoint before continuing past the original estimate.

Every dollar figure in the reconstructed statements needs to trace to either a specific source document or a specific, dated entry in the assumptions log; there is no third category. Before delivery, a senior reviewer reconciles the rebuild against bank totals for the period, and reviews the assumptions log specifically for anywhere a judgment call was made without being written down.

We treat the assumptions log as the real deliverable, not a footnote. If a reader cannot tell, for any given figure, whether it came from a document or an estimate, the reconstruction has not met our standard yet, and it goes back for another pass before it ships.