Digital Mailroom Software
AI sorting for scanned mail
Last updated:
Most digital mailroom providers scan and forward; sorting still falls on you. Docusplit takes the scanned PDF and splits, classifies and renames every letter. Pair it with your own scanner or a virtual mailbox.
A digital mailroom turns incoming paper mail into digital files. Docusplit is the software step after the scan. It splits a scanned batch into separate documents and names every file by type, sender and date, invoices by number and company. It works with any scanner or mailbox provider that delivers PDFs; the upload returns a ZIP plus a CSV overview.
Scanning solves the paper. It doesn't solve the sorting.
Why scanned mail still costs time
Unsorted incoming mail
Invoices, contracts and letters arrive in one pile. The important ones sit between advertising, and deadlines are noticed late.
Sorting by hand, every day
Opening, reading, classifying and filing each document by hand. How long it takes depends on volume, but it is the same routine every morning, and it only scales with headcount.
Scans without names
A scanning service or your own scanner delivers scan_0412.pdf. Finding one letter later means opening files and paging through them.
In-house mail by hand or scanned and uploaded
The digital column describes scanning in-house and uploading to Docusplit, not an outsourced service.
| Criterion | In-house, by hand | In-house scan + Docusplit |
|---|---|---|
| Setup | A tray, folders and someone who sorts | A scanner you already own and a free account |
| Cost per page | Staff time for opening, reading and filing | €0.10 per page on Starter (€9.99 for 100 pages); the first 30 pages are free |
| Turnaround | Depends on who gets to the tray | About 10 seconds per page of AI processing after the scan |
| Filename | Typed by hand, if at all | Type_Sender_Date, invoices as InvoiceNumber_Company |
| Findability | Paper folders | Named PDFs in folders by month or sender, plus a CSV overview |
| Unreadable pages | Noticed while reading | Named Document_Page_N.pdf and listed as "Not recognized" in the CSV |
How it works
From scanned mail to named files in three steps.
Scan or receive your mail
Scan the day's stack as one PDF, or download the scans your virtual mailbox delivered.
Upload and pick Documents mode
The AI reads each page, detects where a new document begins, keeps multi-page letters together and classifies each one. Expect about 10 seconds per page.
Download the sorted ZIP
One PDF per document, named by type, sender and date, optionally in folders by month or sender, plus a CSV overview. Copy it into your DMS or cloud folder.
What Docusplit adds to a scanned mail stack
The part after the scan, described without embellishment.
Named files instead of page images
Each file gets a name from its content, Type_Sender_Date or InvoiceNumber_Company, so it can be found without opening it.
Folders by month or sender
Choose folders by month (2026-03/) or by sender when you download; pages without a recognizable date go to Unknown/.
Deleted after processing, no AI training
Files are transferred over TLS, kept only while processing and then deleted from our servers. AI analysis runs through the OpenAI API and is never used for training; see our privacy policy.
A digital mailroom receives, scans, classifies and files mail
A digital mailroom is the process of turning incoming paper mail into digital files that reach the right people. It has four parts: receiving and opening the mail, scanning it, classifying and naming each document, and routing or filing it. Each part can be done in-house or by a service.
Outsourced providers such as Xerox or Iron Mountain, and virtual mailbox services, handle the physical side. They receive your mail at their facility, scan it and deliver the scans. What most of them deliver is a PDF per envelope or per day, with the classification and naming left to you.
Docusplit covers the third part only. It takes a scanned PDF from your own scanner or from a service, splits it into individual documents, names each file by type, sender and date and returns a ZIP. Filing into your DMS remains a copy step on your side.
Virtual mailboxes give you an address and scans, not sorted files
Virtual mailbox providers give you a real street address where your business mail is received and scanned. iPostal1, Anytime Mailbox (2,500+ locations as of August 2026), PostScan Mail and Earth Class Mail all offer receiving, opening, scanning and forwarding, with plans that start at a few dollars a month.
What you get from them is a scan per mail item, viewable in a portal and downloadable as a PDF. Sorting those scans into your own structure, naming them so they can be found later and keeping multi-page letters together is still your job, and it is where the workflow stalls once volume grows.
That is the step Docusplit takes over. Download the day's scans, combine them into one PDF or upload them one by one, and the AI splits, classifies and names each document, for example Notice_City-Utilities_2026-03-15.pdf for a letter or INV-2026-0481_Acme-Corp.pdf for an invoice. A virtual mailbox for the paper plus Docusplit for the naming covers the whole chain without manual sorting.
When outsourcing makes sense, and what it still leaves to you
Outsourcing the mailroom makes sense when nobody is at the office to open the mail, when several locations should feed one inbox, or when you want a registered address that isn't your home. The providers are good at the part you can't do remotely, which is receiving, opening and scanning.
What they don't do is decide what each page is and what it should be called. An outsourced mailroom hands you a PDF per envelope or a daily batch; the sorting is still yours, and it costs the same whether the scan came from a service or from your own scanner.
So the question isn't outsourcing versus software. It is who handles the paper, and that depends on your office situation. The naming step is the same in both cases, and it is the one Docusplit does for €0.10 a page on the Starter plan, or for free on the first 30 pages.
How Docusplit fits into your digital mailroom
Docusplit doesn't replace your mailbox or your scanner. It sits after the scan, with a PDF going in and a ZIP of named single documents coming out. Your mail gets scanned, by you or by your provider. You upload the multi-page PDF and pick Documents mode. The AI splits the batch into individual documents, keeps continuation pages together, classifies each document and names it. You download a ZIP with one PDF per letter, optionally in folders by month or by sender, plus a CSV overview.
The CSV lists filename, page, document type, sender, date and a confidence value per file, so you can see at a glance which files deserve a second look. Fields the AI could not read are marked "Not recognized", and pages it could not classify are named Document_Page_N.pdf.
There is no connector to a DMS, ERP or cloud drive. You unzip the archive into the import folder of your DMS or the sync folder of your cloud drive; the folder structure is kept. That is the handoff most virtual mailbox providers leave to you, and it is a copy step rather than a data-entry step.
What Docusplit does for your mail
The functions that complete a scanned mailroom, and their limits.
- Splits multi-page mail scans into individual documents, with boundaries found from content
- Classifies each document by type: invoice, contract, letter, form, notice
- Names files by type, sender and date; invoices by invoice number and company
- Keeps multi-page documents together (a 3-page contract stays one file)
- CSV and JSON overview per upload; import into a DMS by copying, no connector
- One PDF per upload up to 50 MB, page count limited only by your plan
Who runs scanned mail through Docusplit
Four inboxes where mail arrives daily and has to be findable the same day.
Law firms
Court notices, contracts and correspondence scanned once a day. Each document is named by type, sender and date, so the notice with a deadline shows up in the CSV before anyone opens a file.
Accounting firms
Client mail with invoices, receipts and tax assessments. Invoices are named by number and company, everything else by type, sender and date, ready to drop into the client folder.
Insurance agencies
Claims letters, policy documents and correspondence split and named. Which file goes to which desk is decided from the filename, not by opening every scan.
Healthcare practices
Patient correspondence, insurance forms and lab reports filed with sender and date in the name. The AI reads the pages through the OpenAI API, so check that this fits your confidentiality obligations first.
FAQ: digital mailroom and virtual mailbox
What is the difference between a virtual mailbox and a digital mailroom?▼
A virtual mailbox gives you a physical address where mail is received and scanned for you. A digital mailroom is the broader process of digitizing incoming mail, in-house or outsourced. Docusplit is neither; it is the software step that splits and names your already-scanned mail.
How much does it cost?▼
Virtual mailbox and outsourced mailroom services set their own prices; check the provider's page. Docusplit's part starts free with 30 pages and then costs €9.99 a month for 100 pages (Starter), €29.99 for 300 (Pro) or €99.99 for 1,000 (Max), billed in euros, cancel anytime, with a 14-day money-back guarantee.
Can I digitize my mailroom without outsourcing?▼
Yes. With a scanner that produces PDFs and a free Docusplit account you can do it in-house. Scan the day's mail as one PDF, upload it, download the named files. Nothing is installed and no provider is involved.
What happens when the AI splits wrongly or can't read a page?▼
Handwriting, very faint pages, scans below roughly 200 dpi, blank back sides or attachments without a letterhead can be assigned wrongly. If the AI reads nothing on a page, the file is named Document_Page_17.pdf and, with folders by month, lands in Unknown/. Check those files and the confidence column in the CSV first; the rest of the stack is unaffected.
What happens to my files?▼
Files are transferred over TLS, kept only while processing and then deleted from our servers. AI analysis runs through the OpenAI API (data processing agreement with OpenAI Ireland, Standard Contractual Clauses, never used for training; OpenAI retains API data for up to 30 days for abuse monitoring). See our privacy policy.
Related guides
How splitting and naming work in detail.
Split & Rename PDF
Upload one batch scan of mixed invoices, letters and receipts. The AI splits it into separate documents and renames each file by content. Sorted ZIP with CSV.
Read guideAuto Rename PDF Files
The AI reads type, sender and date on each scanned page and names the file after it, e.g. Contract_Northwind-Ltd_2025-01-15.pdf. 30 pages free, no credit card.
Read guideSources & References
- Anytime Mailbox — Lists services at 2,500+ locations and plans from $9.99 per month, with opening and scanning of mail contents as PDFs. As of August 2026.
- OpenAI — Your data (API platform) — Data sent to the OpenAI API is not used to train OpenAI models; abuse-monitoring logs are retained for up to 30 days.
Already scanning your mail?
Upload a scanned mail stack and look at the ZIP. The first 30 pages are free.
Try Docusplit free30 free pages · no credit card · deleted after processing
Last updated: