Nobody went to law school to rename scan001.pdf at 4pm.
A client mails in a stack of bills, bank statements, invoices, and letters. The scanner turns them into 40 PDFs named scan001 through scan040. Someone has to open each one, figure out what it is, and type a filename. That task requires no legal judgment and takes an afternoon. DilloLex reads each scan on the firm's own appliance and proposes a clean, dated, findable name so that task becomes a quick review instead of a slow grind.
General information for legal professionals. This is educational material rather than legal advice. Read the full disclaimer.
Document naming is one of those tasks that every firm does differently, costs real time each week, and never gets better on its own. A Las Vegas family law practice handling a financial disclosure under NRCP 16.2 may receive dozens of scanned bank statements, pay stubs, tax returns, and credit card bills in a single intake. A personal injury practice may receive a months-long run of medical bills and itemized provider invoices. A probate matter may open with a box of scanned estate documents, letters, and old bank records. Every one of those files lands in the firm's system with a name that says nothing: scan001, IMG_4832, document(2), or a fax cover sheet that the scanner grabbed first.
Getting from that pile to a folder with clean, dated, searchable filenames is entirely manual work. There is nothing legally difficult about it. It is exactly the kind of high-volume, low-judgment task that eats a paralegal's morning and keeps the firm from handling more matters without adding headcount.
What the manual process actually costs
The per-document cost is small. Open the PDF, read enough to identify it, type a name in the right format, save, move on. Thirty seconds to a minute per file, depending on how fast the scan loaded and how legible the document is. Across a 60-document intake, that is 30 to 60 minutes of a paralegal's time, before any quality check on whether the naming is consistent with the firm's convention for that matter type.
The hidden cost is inconsistency. When naming is manual and done by different people at different times, the same type of document gets different names across different matters. A medical bill from the same provider appears as "St. Rose Bill 4-12-2025," "St Rose Dominican 04.12.25 invoice," and "Bill April 2025 medical." Three documents that should be findable by the same search require three different searches. Inconsistency compounds over time, and it compounds across staff turnover.
A third cost is the occasional miss: a scan that names itself after the second page instead of the first, a faxed letter where the cover sheet got filed as the document, or a statement that spans two scan jobs and looks like two documents. These are not errors a paralegal makes from inattention; they are errors that are easy to miss at volume, when you have 60 more files queued up behind this one.
How DilloLex reads and names each scan
When a scan lands in the firm's designated intake folder, the appliance picks it up and runs OCR on the firm's own hardware. No document content goes to an outside AI. The text the OCR engine extracts stays on the appliance and is used to identify the document and compose a proposed filename. Here is what that looks like in practice, across the document types a Nevada practice sees most often:
Invoices and medical bills
The appliance reads the vendor or provider name, the invoice date, and the total amount. A hospital bill with a legible header and a statement date becomes something like "2025.04.12 St. Rose Dominican Invoice $4,280.00." The vendor, the date, and the dollar figure are all there. A paralegal scanning down the matter folder can identify the document without opening it.
Bank and financial account statements
The appliance reads the institution name and the statement period. A Chase checking statement covering March 2025 becomes "2025.03 Chase Checking Statement." Multiple months of statements from the same account sort in date order automatically, which matters when you are pulling a 12-month financial history for a NRCP 16.2 disclosure.
Letters and correspondence
The appliance reads the sender name or firm, the date of the letter, and the subject line if one appears. A letter from opposing counsel becomes "2025.05.03 Letter from [Counsel Name]." An insurance adjuster's letter becomes "2025.04.28 Letter from [Insurer] Re: Claim." The date and sender make the document identifiable without opening it.
Tax forms and government documents
The appliance reads the form number and tax year. A W-2 from an employer becomes "2024 W-2 [Employer Name]." A Nevada state tax document carries whatever identifier appears on its face. The firm's convention for how these are named in a specific matter type is set at onboarding.
Configuration, not training
The naming rules the appliance applies are configured for the firm during onboarding, not trained on the firm's data after deployment. If the firm names medical bills with the provider first and the date second, that is the convention the appliance follows. If the convention for financial statements in a family matter is different from the convention in a probate matter, that distinction is part of the configuration. The system does not learn new conventions on its own over time; the conventions are explicit, and the firm controls them.
This matters for two reasons. First, it means the firm's naming convention is consistent from day one, not after a training period. Second, it means the convention is something a supervising attorney or office manager can inspect and change if the firm's needs evolve. There is no black box: the rules are explicit and documented.
Low-confidence documents go to review, not to the matter folder
Not every scan produces a confident name. A scan where the OCR engine cannot read a date, where the vendor or sender is not identifiable from the document text, or where the layout is unusual enough that the appliance cannot determine the document type with reasonable confidence goes to a separate queue. The paralegal sees it there, labeled as needing review, and names it manually. Nothing is filed under a guess.
The review queue is a feature, not a failure. It means the matter folder contains only documents the appliance named with confidence, and the documents that need a human look at them are clearly separated. The paralegal does not have to check every file; she checks the ones the appliance flagged. That is a better use of her time than checking every file, whether the appliance touched it or not.
Common reasons a document goes to review: the scan is too light or skewed to read cleanly, the first page is a fax cover sheet with no useful content, the document is a form type the appliance has not seen before, or the date is handwritten rather than printed. The appliance surfaces these rather than guessing.
The proposed name is a proposal
The filename the appliance generates is a starting point. Staff can adjust it before the document moves to the matter folder. If the proposed name has the right date and vendor but the wrong account type, the paralegal corrects it in one click. If the naming convention the firm uses requires a matter number prefix that was not in the document, that can be added. The appliance does the reading and the first draft; the human does the review and has the last word.
This is the same model DilloLex uses everywhere: the tool proposes, the attorney or staff member reviews, and nothing is filed without a human having looked at it. Document naming is a lower-stakes version of the same principle, but the principle is the same.
How DilloLex helps
DilloLex is software that assists the firm's staff. It does not practice law, does not make legal determinations, and does not file documents on behalf of anyone. What it does for document naming is run a task that requires no legal judgment, at high volume, consistently, on the firm's own appliance, and surface the exceptions that need a human. The paralegal reviews proposed names, adjusts the ones that need adjustment, and handles the documents in the review queue. The matter folder ends up with clean, consistent, searchable filenames instead of a pile of scan001 files. No client content goes to an outside AI. The documents, the OCR output, and the proposed names stay on hardware the firm owns.
Because the naming convention is configured at onboarding, the output is consistent across every person who handles intake at the firm, not just the one paralegal who remembers the convention from the last time this came up.
For more on how the appliance handles document management alongside naming, see how DilloLex works end to end, the firm's security posture, and the guide to Bates numbering and production for when named documents go into a production set. If you want to see the naming workflow on a real intake, book a demo.
Common questions
Does the auto-naming tool work on documents that are already in the matter folder?
The appliance processes documents that arrive in the firm's designated intake folder. Documents already in the matter folder can be routed through intake if the firm chooses to reprocess them, but the typical use is at the point of intake, before the document is filed in the matter. Ask about your firm's specific configuration during a demo.
What happens to a scan where the first page is a fax cover sheet?
A fax cover sheet as the leading page is one of the common reasons a document goes to the review queue rather than receiving an automatic name. The appliance reads what is on the first page; if what is there is a cover sheet rather than the document itself, the confidence level drops and a human reviews it. The paralegal can then name it based on the content pages behind the cover sheet.
Can the naming convention be different for different practice areas within the same firm?
Yes. The naming configuration is set at onboarding and can include different conventions for different matter types. A family matter intake can follow a different convention from a probate intake or a PI intake. The distinctions are part of the configuration the firm controls, not something staff has to manage document by document.
What does the review queue look like for staff?
Documents in the review queue appear separately from documents the appliance named with confidence, labeled as needing review. The paralegal sees them listed, opens them to confirm what they are, and assigns a name. The queue is a shortlist of exceptions, not a second pass over everything. The goal is to reduce the time staff spend on naming overall, including the time spent checking the appliance's work.
Is the naming tool governed by any Nevada court rules?
Nevada court rules, including NRCP and NEFCR, address the format and content of filings submitted to Nevada courts. They do not generally dictate how a firm names documents in its internal file system. Internal naming conventions are a matter of firm practice and professional judgment. Nothing in this guide is legal advice; confirm current Nevada rules before relying on any specific requirement described here. Nevada court rules are amended from time to time.
A note on accuracy
This guide is general information for legal professionals. It is not legal advice, it does not create an attorney-client relationship, and it does not address the requirements of any specific matter or court. Nevada NRCP, NEFCR, and related procedural rules are amended from time to time; confirm the current text before relying on any specific requirement described here. DilloLex proposes document names based on OCR and configured naming conventions; those proposals are subject to staff review and are not guaranteed to be accurate in every case.
See it work on your intake.
Bring a scan pile from a recent matter. We will show you the naming workflow, the review queue, and how the configuration maps to your firm's own conventions, on your hardware, before you commit to anything.
Book a demo