Moving off Doculex without losing a thing

The documents are the easy part of leaving an old document system. The hard part is everything attached to them: the index values people search by, the folder logic only one person understands, and the routing that quietly moves invoices to the right desk. This guide is for the IT lead or records owner planning the move.

Why now

• The server or operating system it needs is out of support and hard to patch.
• It no longer gets along with current Windows, Office or browsers.
• Nobody left in the building knows how to administer it.
• People cannot reach documents from home or a phone.
• Sign-in and permissions do not meet what your auditors or insurers now ask for, such as multi-factor authentication.

A failing old system is the worst time to plan a migration. Do it while the old one still runs, so you can export cleanly and check every record against the original.

Inventory what you have

Map the index fields

Every searchable value in the old system needs a home in the new one. Write the mapping table before anything moves, and decide what to do with each awkward case: fields nobody fills in, free-text fields that should be pick lists, dates stored as text.

A worked migration

A regional distributor’s accounts payable archive: about 400,000 documents across 12 years, four document types (invoice, credit memo, purchase order, vendor statement), six index fields, and one routing rule that sent invoices over a set amount to the controller.

1. 01

   Export from Doculex while it still runs

   Each document as a file, plus an index file listing its type and field values. Export by year, so each batch is a size you can check.
2. 02

   Build the target first

   In Content Central: one catalog for AP, the four document types, the six fields with their types and pick lists, and permissions per type.
3. 03

   Load one year as a test

   Bring the files in with their index values, for example as XML capture jobs, where a small descriptor sets the catalog, type and field values for each file. Image-only files get text recognition on the way in.
4. 04

   Check the test like an auditor

   Counts by type and month match the export. Twenty random documents opened side by side with Doculex. Ten searches people run every week give the same results.
5. 05

   Rebuild the routing

   The amount rule becomes a workflow: invoices at or over the threshold go to the controller, with a deadline and escalation.
6. 06

   Load the rest, newest years first

   People ask for recent documents most, so those are searchable soonest. Older years follow in the evenings.

Cutover and checks

• Freeze date: from this day, new documents go only into the new system. Point scanners and email capture there first.
• Delta load: anything added to Doculex between the main export and the freeze is exported and loaded once more.
• Reconcile: total counts by type and year, plus the spot checks, signed off by the department that owns the records.
• Training: one short session per team on searching and opening documents, and a longer one for whoever will administer it.
• Retire: read-only period, final archive copy, then decommission the server.

Fix what Doculex could not do

A move is the cheapest time to add what people have been missing. Content Central runs in the cloud or on your own servers, your choice, with full-text search, workflows and approvals, e-signatures, permissions down to document type and field, an audit trail, version history, check-in and check-out, retention schedules and legal holds, Active Directory and SAML sign-on and MFA, and connections to QuickBooks, Sage 50, Sage X3, Workday, Epicor and Dynamics CRM. Go-live in 30 days is a real commitment. Once the archive is across, put CapturePoint 6 at the scanner so new paper arrives already split, sorted and named. Ready to leave the old server behind? Run Content Central in the cloud instead.