Paper, PDFs, Emails and Legacy Databases: Who Can Bring Them Into One Searchable System?

A former employee sends a subject access request. You have one month to respond^[1], and their information is in a filing cabinet, a shared drive full of scanned PDFs, three colleagues’ inboxes and an HR database that nobody can export from since the vendor stopped supporting it. Each source needs a different search by a different person, and there is no reliable way to know when you have found everything.

That is the practical cost of information spread across paper, PDFs, emails and legacy systems. Bringing it into one searchable document management environment fixes it, but the work is rarely done by a single type of company.

The short answer: Four kinds of specialist do parts of this work – document management software vendors, IT and Microsoft 365 partners, document scanning and data capture bureaux, and data migration specialists. The software platform is usually the easy part. The hard part is getting content out of each source, making it readable and consistently indexed, and loading it with the right permissions, so the specialist you most need is one that handles capture and migration across all your sources and delivers into whichever platform you choose.

This guide explains why each source is different, which type of company does what, how a consolidation project runs and what to ask before you appoint anyone.

Why each source needs different handling

Organisations often assume that once they have chosen a document management system, the content will follow. In practice, each source has its own problems, and a provider that handles one well may not handle the others at all.

Source	Typical problem	What has to happen	What to ask a provider
Paper archives	Unreadable by any system; often indexed only by box	Preparation, scanning, OCR, indexing and quality checks	Do you scan in-house, and to what standard?
PDFs and shared drives	Many are image-only scans with no searchable text; duplicates and inconsistent file names	OCR, de-duplication, classification and metadata extraction	How do you identify duplicates and image-only files?
Emails	Records sit in individual inboxes, mixed with everything else	Decide what counts as a record, then export it with sender, date and attachments intact	Can you preserve email metadata and attachments as linked records?
Legacy databases	Files stored as binary objects, or in a system with no vendor support and no clean export	Extraction, format restoration and mapping each record to its owner	Have you extracted from this type of system before?

The common thread is consistency. A searchable environment only works if a contract scanned from paper, a PDF from a shared drive and an attachment from an email are all indexed in the same way – the same client reference, the same document types, the same dates. That classification scheme needs agreeing before any content moves.

Which type of company does what

Document management software vendors

Vendors supply the platform: The repository, search, version control, workflow and permissions. Many offer implementation services. Their strength is configuring their own product; their weak spot is often getting messy legacy content into it, which is frequently passed to a partner.

IT and Microsoft 365 partners

If your destination is SharePoint, an IT partner will typically design the site structure, permissions and retention settings. They are well placed for born-digital content already in Microsoft 365, but paper archives and extraction from non-Microsoft legacy systems are usually outside their core work.

Document scanning and data capture bureaux

Bureaux turn paper into searchable digital files and extract data from documents. The better ones also classify content and add metadata, rather than simply producing image files. Check whether they scan in their own facility, work to BS 10008 for the legal admissibility of electronic information and can deliver straight into your chosen platform.

Data migration specialists

Migration specialists extract content from legacy systems and load it into new ones. For document-heavy migrations, look for experience with systems that store files inside the database, reconciliation reporting that proves every record arrived, and a clear process for items that fail.

Combining them

Because the four sources need different skills, many organisations end up managing three or four suppliers and the gaps between them. A provider that covers capture and migration across paper, files and legacy systems – and works alongside your platform vendor or IT partner rather than replacing them – removes most of that coordination risk.

How a consolidation project runs

A well-run project follows the same broad sequence, whatever the platform:

1. Inventory every source: List the paper archives, drives, mailboxes and systems in scope, with rough volumes and who owns each.
2. Apply retention first: Don’t migrate what you no longer need. UK GDPR requires that personal data is not kept longer than necessary^[2], and every expired record you destroy is one you don’t pay to scan, index and store.
3. Agree one classification scheme: Document types, mandatory metadata fields and naming conventions that apply to every source.
4. Capture and extract: Scan paper, run OCR on image-only PDFs and extract content from legacy systems into the agreed structure.
5. Map permissions: Decide who should see what in the new environment, rather than copying old access rights across unchecked.
6. Load, reconcile and validate: Migrate in batches, with a report matching each batch back to its source and an exceptions list for anything unreadable or unmatched.
7. Keep new content consistent: Route incoming post, email and documents into the same structure, or the problem rebuilds itself within a few years.

Steps 2 and 5 are the ones most often skipped, and they matter more than ever once AI tools are involved.

One searchable environment is also the first step to AI readiness

Organisations consolidating their records now are increasingly doing so with AI in mind, whether that is Microsoft 365 Copilot or a specialist tool. The same work that makes records searchable for people makes them usable by AI, and the gap is wide: In a Gartner survey of 1,203 data management leaders, 63% said their organisation either did not have, or was unsure whether it had, the right data management practices for AI^[3].

Three parts of a consolidation project decide whether AI can use the result:

• Readable text: An AI assistant cannot read an image-only scan. OCR quality on paper and legacy PDFs sets the ceiling on what it can find.
• Permissions: Copilot only surfaces content a user already has at least view access to, which means broadly shared files can appear in answers to people who never knew they could open them^[4]. Bringing scattered records together without reviewing access can turn a forgotten folder into an exposure.
• Provenance and metadata: If every record carries its source, date and document type, AI answers can be traced back to the document and page they came from.

Building these in during consolidation costs far less than retrofitting them once content has been loaded. Dajon’s AI data readiness service covers this preparation for organisations planning to use AI on their records.

Questions to ask before you choose a provider

1. Which of our sources can you handle directly, and which would you subcontract?
2. Where is the work done? In your own UK facility, or sent elsewhere?
3. Which security and compliance certifications do you hold? ISO 27001 for information security and BS 10008 for legally admissible electronic records are the most relevant.
4. Are you tied to one platform? A technology-neutral provider can deliver into SharePoint, your existing document management system or a new one.
5. How do you prove completeness? Ask for an example reconciliation report and exceptions log.
6. How do you carry permissions across? And how do you flag content that looks overshared?
7. What happens to the originals? Return, secure storage or certified destruction once the digital copies are signed off.

How Dajon helps

Dajon scans paper and extracts content from shared drives, unsupported document management systems and databases that store files as binary objects, then runs OCR and classification and adds metadata. Dajon is technology-neutral: It migrates prepared records into SharePoint, your document management system or a data platform with permissions intact and a reconciliation report for every batch. Work is carried out in its Orpington and London facilities under ISO 27001 and BS 10008.

Frequently asked questions

If your records are spread across paper, drives, inboxes and old systems, talk to Dajon about what you hold and where you want it to end up. We’ll scope the sources, volumes and steps involved.

---

References

1. A guide to subject access ICO[↩]
2. Principle (e): Storage limitation ICO[↩]
3. Lack of AI-Ready Data Puts AI Projects at Risk Gartner[↩]
4. Microsoft Copilot for Microsoft 365 – Best practices with SharePoint Microsoft Learn[↩]
5. Data (Use and Access) Act 2025, section 78 legislation.gov.uk[↩]