Document intelligence and cloud migration
Matching lease records to physical documents across an estate of hundreds of thousands of files, migrating to cloud and proving nothing was lost.
Business-critical lease documents were spread across a large SharePoint estate, with no reliable link between the lease record and its corresponding file. The company needed to migrate that content to a cloud platform without losing track of any document — and with no way, at the end, to assert that the migration was correct.
The starting inventory held 437,225 files and 109,406 folders across 171 properties. Filenames and folder structures followed no consistent convention, making it impossible to match lease to document by naming pattern. Migrating blind was not viable: without reconciliation, any gap would surface months later, in an audit.
- Analysis of 546,631 inventory rows and 176,513 lease-related files
- Matching of 2,928 out of 3,582 lease records with high confidence (81.7%)
- Built the property, tenant, lease and final destination-path mappings
- Consolidated files into a SharePoint staging area using ShareGate
- Tracked an earlier batch of 6,823 files: 6,247 copied and 576 exceptions handled
- Final transfer to Azure Blob Storage via AzCopy
- Path and file-size validation on every file, with exceptions held separately
The final run took roughly 13 hours and moved 27,821 PDFs, covering 165 properties and 3,490 tenants. The validated batch closed with zero missing files, zero unexpected files and zero size mismatches. A further 306 files unavailable at source and 13 zero-byte PDFs were documented separately rather than absorbed into the result — giving the business an auditable number instead of a claim of success.
Have an operational problem to solve?
Describe the process that needs to evolve, or the decision your company needs data to support. We come back with an initial read and the next steps.