Preparing a Consignment
What to put in the manifest, how to name the files, what the scans need to look like, and what happens to your paper afterwards
How a consignment works
The sequence end to end
A consignment is one physical unit of work: the boxes you are sending in this round, tracked together from scan to sign-off. It is deliberately a different thing from a client-list import, which is a spreadsheet of rows you typed rather than paper you shipped.
- You send a scoping enquiry giving volume in boxes or linear metres, and we quote the project.
- You download the manifest template and complete one row per document.
- The paper is scanned, either by a scanning bureau or on your own equipment if you already have it.
- You upload the consignment: the manifest plus the scanned content.
- We reconcile what the manifest declared against what actually arrived.
- Each item is classified and its key fields extracted on our own servers.
- Your firm works the review queue, correcting anything the extraction got wrong.
- A quality-assurance sample of the auto-accepted items is spot-checked.
- An authorised member of your firm signs the batch off, and a certificate is filed to your vault.
- You record what happened to the original paper.
Where the documents end up
Everything lands in your firm’s own vault space, under the same access rules, audit trail and encryption as documents your staff upload by hand. If your firm belongs to an organisation, the documents belong to the organisation rather than to the member of staff who ran the upload, so a consignment survives that person leaving.
The manifest
Required columns
The manifest is a CSV with one row per document, not per box and not per page. Download the current template rather than building your own file: the template is the authoritative column list, and a manifest that does not match it is rejected before anything is written.
- Box reference: your own label for the physical box or bundle
- File reference: your own label for the file or matter folder within the box
- Client identifier: the email address or client reference we resolve against your own client book
- Matter reference: your practice-management reference for the matter, where you have one
- Document description: a short human description, for example "attendance note, first appointment"
- Page count: the number of physical sides the document runs to
- Date: the date on the document itself, not the date it was scanned
Reconciliation
When the consignment arrives we compare what the manifest declared against what was actually received, row by row, and produce an unreconciled report listing anything that is missing, unexpected or ambiguous. A batch whose counts disagree does not quietly proceed. The mismatch has to be resolved first, because a silently short consignment is exactly the kind of gap that surfaces years later in a disclosure request.
Client identifiers
Client identifiers are resolved against your own client book and nothing else. An identifier that does not match one of your clients comes back as unresolved on the report; it never tells you whether that person exists anywhere else on the platform.
Naming, boxes and batches
Naming conventions
- Every scanned file name must match the file reference in its manifest row exactly, including case: that pairing is the only link between paper and row
- Keep box references stable across rounds; re-using a reference for a different box makes the chain of custody unreadable
- Use plain ASCII letters, digits, hyphens and underscores in file names; avoid spaces, slashes and accented characters
- One document per file. A single PDF holding an entire file folder cannot be reconciled against per-document rows
- Multi-page documents stay as one multi-page file, in page order, right way up
Box and batch limits
Limits on archive size, entry count and per-item size are enforced on our side and published alongside the manifest template, so that the figures you plan against are the figures actually applied. Send a large back-book as several consignments rather than one enormous archive; each is tracked, reviewed and signed off independently.
Storage is checked before any byte is written. If a consignment would take your firm over its storage allowance, the upload is refused up front with a storage-limit error, rather than failing part-way through with half a box ingested. Within a consignment, one unreadable file fails on its own; it never brings down the batch around it.
Resolution and colour
What the scans need to look like
- 300 dpi is the floor for anything going through extraction; higher is fine, lower is not
- Scan in colour anything carrying ink: signatures, initials, manuscript amendments, coloured tabs, certification stamps
- Greyscale is acceptable for clean typescript with no manuscript content
- Deskew and de-speckle, but do not crop into the page edge or auto-rotate text-light pages
- PDF is preferred; single-image-per-page TIFF or JPEG is accepted
- Do not apply a downsampling or lossy compression profile that softens handwriting
The scanning device and the resolution used are recorded against the consignment and printed on the certificate at sign-off, which is part of what makes the record worth keeping.
Scanning on your own equipment
What we can and cannot read
Generally read well
- Typed correspondence, typed attendance notes and typed file notes
- Printed forms with handwritten entries in the boxes
- Certificates and formal documents with a consistent printed layout
- Clear block-capital manuscript annotations
Often need a human
- Dense cursive handwriting, particularly older hands
- Faint carbon copies and thermal fax paper that has aged
- Heavy bleed-through from double-sided originals on thin paper
- Pages skewed or obscured by staples, tape or archival repairs
- Documents photographed rather than scanned, at an angle or under uneven light
Nothing is silently dropped. An item the classifier cannot read confidently is routed to your review queue rather than guessed at, and an item it cannot process at all lands in a failure list where you can enter the fields by hand or send it back for a re-scan.
Archive content never leaves our servers
What happens to your originals
We never take custody
We never take custody of your originals. The paper stays with your firm or with the bureau you appointed, under whatever arrangement you have with them. We are not a storage business and we do not become a link in the physical chain.
Recording the disposition
Once a batch is signed off you record what happened to the paper (retained, returned, or destroyed), the instruction your firm gave and when, and the date it happened, plus a destruction certificate reference for anything destroyed. That instruction is never assumed and never pre-selected: a batch cannot be marked destroyed without an instruction on the record.
Deeds and original wills are surfaced separately in the disposition record, because documents executed as deeds may retain legal significance that a scan cannot replicate. Retention periods and the standing of a scanned copy vary by document type and by jurisdiction, and this page is general information rather than advice on your own files. We never tell a firm that it may destroy anything. That decision is yours, and your own compliance lead is the person to take it with.
Timescales
We do not publish a turnaround figure, because we have not measured one yet. Throughput depends on how much of a consignment is handwritten, how clean the scans are, and how quickly your firm works the review queue, and a number invented before those are known would be a number we could not stand behind.
Once we have measured real consignments end to end we will publish indicative ranges here. They will be ranges rather than a guarantee, and your quote will carry the estimate for the specific consignment you are sending.
Related topics
Can’t find what you’re looking for?
Our support team is here to help. Contact us and we’ll get back to you as soon as possible.
Contact Support
