Use caseContent migration

Migration that makes content AI-ready

Procurement is leaving OpenText: 3.8 million files, over half untyped, many in three copies. A straight copy would hand that mess to the AI.

The content is measured before it movesEach file gets a class and its properties on the wayA person reviews every file the AI is unsure about

The operations engineQuality profile, procurement

3.8 million files read in place. 2.4 million carry no category at all, and 410,000 are exact duplicates of another file.

source · discovery scan, Content Server
Actions on this campaign
The verdict on each file

The AI is below the threshold · a person confirms or corrects

01

Where it stalls today

The core idea

A search index fed with untyped duplicates returns three versions of the wrong contract, and an assistant grounded in them cites the draft instead of the signed copy. A straight copy moves the problem into the new repository.

The cleanup rides along with the move. The pipeline classifies each file, extracts the properties its class requires and flags what is redundant, and anything below the threshold goes to a person. The files land with a type, a supplier and one main copy.

Source profile, procurementOperations engine
The mess, measured. More than half the files sit in folders called Misc, Scans or Old, with no category at all.
Ask the repositoryAI console
The signed copy, cited. After the move, the answer comes from the signed contract, because the draft was set aside.

If you keep one thing from this page, keep the order: measure the content, define what a usable file looks like, fix each file in the pipeline, and send the doubts to a person. Tomas measures, Kevin builds the pipeline, and Keisha settles the doubts before she asks the first question.

02

The measure

A quality profile comes first: it tells the team which stages the pipeline needs and how much review to plan.

Step 1Discovery

Profile the content before moving it

Tomas Lindqvist, the content architect, runs a discovery scan on the procurement workspace of the Content Server. It reads formats, volumes, categories and folders in place, and nothing moves.

Missing metadataInconsistent valuesDuplicatesObsolete content

More than half the files sit in folders called Misc, Scans or Old. The profile becomes the baseline the project will be measured against.

Source profile, procurementOperations engine
The folders that hold the most files are the ones with the least metadata.
The cleanup is sized before it startsThe team knows how many files need a class and how many copies to sort out, so the review is planned on real numbers.
Step 2Repository

Define the classes and the properties each one needs

Tomas writes the target model with the procurement team: eight document classes, each with its required properties. A supplier contract needs a supplier, a contract number, a start date and a signed flag.

A file that cannot get its required properties is never loaded with blanks. It goes to review.

Target modelRepository
A class is created or changed with no schema migration.
Usable is defined before the moveThe team agrees in advance what a file must carry to be found. The pipeline then checks every file against it.
03

The pipeline

Four stages sit between extraction and loading, and each one writes what it did on the file.

Step 3AI check

Classify each file

Kevin Brandt, the migration engineer, adds a classification stage after extraction. The AI reads each file and proposes one of the eight classes with a confidence level.

Above the threshold Tomas set for this use, the class is applied. Below it, the file goes to the review queue with the AI's proposal attached.

Stage: classificationOperations engine
The threshold is set per use, and can be changed between two batches.
The AI only applies what it is sure ofEverything else reaches a person with the proposal already made. Nobody types a class into two million files.
Step 4AI check

Extract the properties the class requires

Once a file has a class, the next stage extracts that class's properties: supplier name, contract number, dates, amounts. A business rule normalizes the supplier name to the spelling procurement uses.

SupplierContract numberStart dateSigned
Stage: enrichmentOperations engine
The values are written as properties of the target class.
The file can be found by what people knowA buyer searches by supplier and contract number. Those values now exist on files that had only a scan name.
Step 5AI check

Flag what is redundant, and what is sensitive

Exact duplicates were counted in the profile. The AI also flags redundant content the scan cannot see: the same contract scanned twice, a Word draft next to its signed PDF. It flags personal data in a scan at the same time.

The rule keeps the signed copy as the main document. Every other copy is set aside and listed in the report, with its original path.

Stage: redundant contentOperations engine
Two versions of one contract, compared before anyone chooses.
Picture a buyer asking your assistant for a rate. How sure are you that it read the signed version, and not the draft sitting in the same folder?
One main copy per documentSearch and the assistant return the signed contract. The draft is set aside, listed, and still traceable.
Step 6Review

Work the review queue

Keisha Brooks, from procurement, opens the review queue every morning. Each file shows the AI's proposal and what is missing. She confirms SCAN_0412_0004 as a supplier letter, picks the supplier, and marks two old fax covers as not worth keeping.

Review queueOperations engine
Each decision is recorded on the file, with its author.
A person decides every doubtThe AI made 5,000 proposals. The 212 it was unsure about were settled by someone who knows the suppliers.
04

The landing

After each batch, the team checks the result against the baseline, then tries the content the way users will.

Step 7Supervision

Follow a batch through every stage

Kevin runs the Scans folder in batches. The counts per stage show where files leave the main path: set aside as redundant, sent to review, excluded by a reviewer. Every exit has a reason that can be listed and exported.

Campaign: Scans folderOperations engine
What left, what arrived, what was set aside and why.
Nothing disappears without a reasonA file that did not land is on a list, with the stage that stopped it and the person who decided.
Step 8Reporting

Compare the quality report with the baseline

Tomas opens the quality report. It uses the same measures as the profile in the first act, so the two read side by side: files with a class, required properties filled, redundant groups settled. He sends it to the AI project lead with the files still open in review.

Quality report, procurementOperations engine
The same measures, before and after the move.
Ready is a number the AI lead can readThe report says what is clean, what is not yet, and where the open files are.
Step 9Assistant

Ask the new repository a procurement question

Keisha tries the result the way a buyer will. She asks the assistant for the handling rate signed with Haulmark Freight. The answer cites the signed contract and its page, and reads only what she is allowed to see.

Supplier contract PA-2017-0388AI console
The answer points to the document and the page it comes from.
The assistant cites the right copyBecause the draft was set aside and the class is set, the answer comes from the signed contract.
05

The verdicts

When a file leaves the pipeline, one of three things happens to it: it lands, it waits for Keisha, or it stays behind.

Loaded ready

Class and properties found

The AI is above the threshold and every required property has a value.

Next: the file lands in its class, with its trace in the report.
Sent to review

The AI is unsure

A low confidence level, or a required property it could not find.

Next: a reviewer confirms or corrects, and the file runs again.
Set aside

Redundant or not worth keeping

A copy of the main document, a draft, a fax cover.

Next: listed in the report with its path and the reason, decided by a person.
What the AI does not do

It deletes nothing. It does not write the classes, and it does not apply a class it is unsure of. Across the nine steps on this page, every doubtful file is settled by a person, and the decision is recorded on it.

The AI classifies, extracts and flags. The threshold and the required properties are the team's, so the rule holds when the volume is millions of files.

06

Time back

The Scans folder, 1.1 million files on its own, is where these counts come from. Your content and your rules will set yours.

0 filesloaded with a required property blank. They go to review instead.
1 main copyper document, so search and the assistant return the signed version.
1 migrationinstead of a copy now and a cleanup project later.

Where these numbers come from. They are counted on the scenario described on this page, with example data. They are not a customer result.

The rules behind the gains

Measure firstDiscoveryThe profile and the final report use the same measures, so progress is a number.
Fix on the wayTransformationClass, properties and redundant copies are handled in the pipeline, before loading.
Doubts to a personHuman reviewThe threshold sends what the AI is unsure of to someone who knows the content.
07

For your teams

Before migrated content feeds an AI project, the AI lead, the content architect and the people who know the suppliers each have to trust it. What they get:

IT and architecture

CIOs and AI leads

The AI project needs clean content, and the migration is about to copy the old mess.

  • One project handles the move and the cleanup.
  • The quality report says how ready the content is.
  • The model is chosen per use, hosted where your data requires.
  • Every AI call keeps its question, answer and model.
Content architecture

Content architects

Design a model for files that carry no metadata at all: that is your brief.

  • The profile shows what is missing, folder by folder.
  • Classes and required properties are yours to define.
  • The threshold is set per use and can change between batches.
  • Transformation rules are tested on a sample first.
Business operations

Procurement reviewers

If the assistant quotes the wrong contract, it lands on your desk.

  • You only see the files the AI is unsure of.
  • Each file arrives with a proposal to confirm or correct.
  • Drafts and copies are set aside, never lost.
  • Your decisions are recorded on each file.

Where to start

With one folder. Take the one that worries you most, profile it, and run it through the pipeline as a pilot. One folder is enough to watch the numbers move.

The source is read in place during the profile, and nothing lands in the new repository until the batch report has been read.

Next step

Profile one of your folders with us

Point us at a part of your Content Server that nobody can search. We will profile it and show how much the pipeline can classify and enrich, and how much a person would review.

The screens on this page are drawn to show one step at a time, with example data.

Uxopian Software helps organizations put AI to work across the systems they already run, without giving up control of their data, their infrastructure or their critical processes.