How to Clean Up Files Before Moving to Clio, iManage, or NetDocuments

You signed the contract. The DMS sales rep is calling about kickoff. And now you are staring at twenty years of folders on a Windows file server, a Dropbox two paralegals share, and a partner's desktop with files named Smith Final FINAL v3 (use this one).docx. If you are wondering how to organize files before moving to a legal DMS, the honest answer is that the migration itself is the easy part. The cleanup before it decides whether your firm spends the next three years cursing the new system or quietly billing more hours because everything is finally findable.

This guide covers pre-migration file cleanup for the three platforms most US law firms evaluate: Clio Manage, iManage Work, and NetDocuments. It is platform-neutral. The goal is to make sure that on day one, the documents you imported reflect the matters and clients you actually have, not the mess on the share drive.

Every DMS vendor will tell you their import tool is fast. True. What they will not tell you is that the import faithfully reproduces whatever you feed it. If your source has six copies of an engagement letter, four mis-filed under a closed matter, your DMS will contain six copies, four mis-filed. Forever. With versioning on.

This is why the highest-leverage work in any law firm DMS migration happens before a single file is uploaded. A legal document migration checklist that prioritizes cleanup, dedupe, and naming consistency beats one that prioritizes speed. Firms that rush the import pay twice: once for the migration, and again for the consultant who untangles it eighteen months later.

Law firm file cleanup before DMS import is also one of the rare moments when you can systematically apply security classifications, confirm retention compliance, and create an audit trail. After import, retroactive cleanup is harder because every document is wrapped in DMS metadata, version history, and shared links.

Each platform imports differently. Knowing the target shape tells you what your source folders need to look like.

• Clio Manage is matter-centric. Documents live under a Matter, which belongs to a Client. Bulk uploads happen at the matter level, via the web app's document upload, the Clio Launcher, or third-party migration partners. Source folders already organized by client and matter make Clio the most forgiving destination.
• iManage Work (including iManage Cloud) uses workspaces and folders inside them. A workspace usually maps to a matter. Documents are imported via drag-and-drop in the desktop or web client, the iManage Drive sync agent, or partner import tools. iManage leans hard on profile metadata (client, matter, document type, author), so prep pays back fastest with a clean folder-to-matter map.
• NetDocuments uses cabinets, workspaces, and documents, with ndDrive (a Windows file system mount) plus a bulk import utility. NetDocuments asks for profile metadata at import, so cleaner source names mean less manual profiling at cutover.

All three reward the same source-side cleanup: predictable folder structure, predictable file names, deduped content, and a clear inventory of what should and should not migrate.

The short answer is to treat the source files as if they were already inside the DMS. Decide what a matter folder should look like, name documents the way the DMS will expect, and remove anything that should not have followed you out of the Windows 7 era.

In practice, the prep work falls into seven steps. If you do only three, do the inventory, the naming standardization, and the dry-run. Those catch most of the disasters.

Step 1: Inventory your current sources

Most firms underestimate sources by half. A typical mid-sized practice has documents on a Windows server, in Dropbox or Box, in OneDrive accounts, on partner laptops, in old PSTs, and on an external drive in a drawer.

• List every source: server shares, cloud drives, personal devices, retired laptops.
• Estimate file count and total size per source (drives scope and import time).
• Identify the source owner who can answer content questions.
• Flag read-only, encrypted, or legacy-format sources (WordPerfect, old PSTs).

Step 2: Map source folders to target matters

This decides whether your DMS will be navigable. For Clio that means existing matter IDs; for iManage and NetDocuments, workspace IDs and matter numbers.

• Export your matter list from your practice management system as a CSV.
• For each top-level source folder, assign a target matter, or flag it administrative, archive, or skip.
• Decide how to handle folders that span multiple matters (most firms split with a deduped copy in each).
• Flag orphan folders for partner review. Do not let them become a DMS dumping ground.

Step 3: Standardize filenames

If you do nothing else on this list, do this. Consistent file names are the single highest-ROI prep step.

• Adopt YYYY-MM-DD_Matter_DocType.ext and apply it consistently.
• Strip junk suffixes: final, FINAL, v3, (copy), USE THIS, partner initials, stale dates.
• Remove emoji, special characters, trailing spaces. They break iManage profiling and confuse NetDocuments.
• Limit path length. Windows servers tolerated 260+ characters; DMS APIs sometimes do not.

Step 4: Detect and deduplicate

Most firms have 10 to 30 percent duplicate content across sources: engagement letters saved into matter, client, and personal folders; contracts emailed back and forth; PDFs re-scanned.

• Run a content-hash dedupe (not filename dedupe) so renamed copies still get caught.
• For near-duplicates, keep the version with the latest modified date in the correct matter folder.
• Log every dedupe decision. You will be asked about one in six months.
• Treat email attachments as a separate dedupe class, especially when migrating Outlook archives.

Step 5: Identify privileged and confidential documents

The DMS can set ethical walls and read-only profiles, but only on documents tagged correctly. Tag in the source first.

• Search for privilege markers: privileged, attorney-client, work product, do not disclose, confidential.
• Identify documents tied to ethical walls (lateral hires, conflict screens, sealed matters).
• Pre-tag privileged documents in a metadata sidecar your import tool can map to DMS security classes.
• Confirm alignment with the ABA Model Rules, particularly Rule 1.6 on confidentiality.

Step 6: Decide what to migrate vs. archive

Not every document deserves a new home in the DMS. Closed matters past retention, expired CLEs, old admin spreadsheets, and files of departed staff are migration debt that bloats your license and slows search.

• Apply your firm's document retention policy. If you do not have one, write it now.
• Send sub-threshold content to a cold archive (cheap blob storage, write-once if possible).
• Record the archive location in matter records so disclosure obligations remain enforceable.
• Get partner sign-off when archive vs. migrate is ambiguous.

Step 7: Dry-run import on a single matter

Pick one closed matter with moderate volume and mixed file types, and run a full import end to end. This surfaces every issue you missed in planning.

• Profile the matter with the same metadata your real matters will use.
• Test search after import: can you find documents by client, matter, document type, and full text?
• Confirm versioning, check-out, and check-in behave correctly for at least three users.
• Roll back, fix the cleanup process, then begin the live migration.

Saying no separates a clean DMS from a digital landfill. Most firms can cut total file count by 20 to 40 percent without losing anything they will ever need. Default-skip list:

• System files: .DS_Store, Thumbs.db, desktop.ini, ~$ Office lock files, .tmp drafts. Zero value, clogs DMS indexes.
• Duplicate PDFs: the same court filing scanned three times for three associates.
• Old admin junk: party invitations, expired holiday schedules, payroll from a prior accounting system.
• Expired CLE certificates from departed attorneys, unless your jurisdiction requires firm-level retention.
• Personal documents from prior owners of laptops and accounts. HR and privacy problem, not a DMS problem.
• Software installers and ZIP archives: nobody needs the 2014 copy of WinZip in the DMS.
• Email PSTs without a retention decision. Do not silently drop PSTs into matter folders.

Run a literal blocklist during your iManage migration file prep or NetDocuments migration cleanup. Filtering before import is far easier than cleanup afterward.

If your firm adopts one habit during this migration, make it filename consistency. 2023-04-12_Smith-Estate_Engagement-Letter.pdf is searchable, sortable, and self-describing. smith stuff FINAL v2 (Karen edits).pdf is a tax on every future associate.

A defensible convention:

YYYY-MM-DD_Matter_DocType_Descriptor.ext

Examples:

• 2024-08-14_Jones-Acquisition_Engagement-Letter.pdf
• 2025-02-03_Acme-Trademark_Office-Action-Response.docx
• 2023-11-30_Estate-of-Riley_Inventory_v2.xlsx

Why this works inside Clio, iManage, and NetDocuments:

• The leading date sorts chronologically in every file browser and DMS list view.
• The matter token aligns with DMS profile indexes, lifting search hit rates.
• The document type token maps into iManage and NetDocuments profile fields at import.
• Junk versioning suffixes disappear because the DMS handles versioning natively.

For matter folder cleanup migration, do renames in the source before import. Renaming after means rewriting DMS profiles, which is slower and adds audit-log noise.

Privilege is not a metadata afterthought. Under ABA Model Rule 1.6 and parallel rules in most US jurisdictions, lawyers must take reasonable steps to safeguard client information. DMS access controls only protect what you classify correctly. Documents that arrive in the wrong workspace or security class can be visible to the wrong attorneys on day one.

A defensible pre-migration approach:

• Run keyword and pattern searches for privilege markers. Capture results in a CSV with file path, matter, and detected marker.
• Cross-reference matters subject to ethical walls, sealed proceedings, or conflict screens. Put those in a separate import batch.
• Tag privileged documents in a metadata sidecar (CSV or JSON) the import tool can read. iManage and NetDocuments both support import-time profile assignment.
• Have a partner review the privilege-tagging report before any import touches the destination DMS.

The goal is not just correct permissions. It is being able to show, after the fact, that you applied reasonable care to client confidentiality at the moment of transition.

The most underrated artifact of a DMS migration is the move log: a CSV (or small table) with one row per file processed. At minimum:

• Source path: original full path on the file server or cloud drive.
• Target matter / workspace: where the file landed in Clio, iManage, or NetDocuments.
• Target name and ID: new canonical name and DMS document identifier.
• Action: migrated, deduped (kept), deduped (skipped), archived, skipped (blocklist).
• Confidence: matter-mapping confidence (high, medium, low).
• Privilege flag: marked privileged or confidential before import.
• Timestamp and operator: who ran the step, when.

This log saves you when a client calls eighteen months later asking where their 2022 amendment went. You search the log, see the file was deduped against an identical copy in the master matter folder, and walk straight to the canonical DMS document. Without the log, you are guessing while the partner waits on the line.

Keep the move log inside the DMS, in a non-client administrative workspace, with appropriate retention. Preserve it for the life of the underlying documents.

Most of the prep work above is local. The files live on Macs, Windows laptops, file servers, and locally synced Dropbox and OneDrive folders. Sortio is a Mac and Windows desktop app that runs that local cleanup layer before the DMS import: it sorts source folders into a matter-centric structure and applies a consistent naming convention so the import into Clio Manage, iManage Work, or NetDocuments starts from a clean source. Sortio is not a DMS. It is the layer that runs after you decide your law firm Dropbox to DMS plan and before the import tool runs. For larger drives, the AI Rule Builder approach is faster than a single full AI sort because rules run locally and can be reapplied as new content arrives.

How long does a law firm DMS migration take?

Most small to midsize firm migrations run four to twelve weeks. The active import is one to two weeks; the rest is discovery, cleanup, mapping, dry-runs, and training. Compressing the schedule usually shows up later as cleanup consulting bills.

Should I clean up files before or during migration?

Before. Cleanup during migration means rushed decisions, inconsistent naming, and a "we'll fix it later" backlog that never gets fixed.

Can I migrate Dropbox to iManage?

Yes. Dropbox sync folders are file system folders, and iManage Work imports from local paths via the desktop client, drag-and-drop into the web app, or partner-built migration tools. The constraint is preparatory: Dropbox folders are usually shared with informal structure, so matter mapping and naming standardization matter more, not less.

What's the biggest mistake firms make during DMS migration?

Importing the source folder structure verbatim. The DMS faithfully reproduces the mess, and the firm spends the next year reorganizing inside a system that costs significantly more per seat than the share drive did. Treat the import as re-classification, not a copy.

Do I need professional services to migrate to NetDocuments?

For small firms with well-organized sources, ndDrive plus the bulk import can be enough. For firms with mixed sources, legacy servers, or real volume, services partners are worth the cost. Either way, the cleanup above happens first, and the cleaner the source, the lower the services bill.

A DMS migration is one of the few moments in a firm's life when every document gets looked at. The firms happiest after a Clio, iManage, or NetDocuments rollout are not the ones with the fastest import. They are the ones who used the migration as a forcing function to inventory, rename, dedupe, and re-classify, so the new system starts honest. Spend the prep time. The import takes care of itself.