Scope the evidential collection

Email work for a dispute can involve preservation, collection, targeted analysis and preparation for wider document review. The appropriate sequence depends on the issues, custodians, sources and instructions. An examination of a disputed message and a large disclosure exercise require different outputs even when they begin with the same mailbox.

Inputs and review requirements

The instructing team should identify the relevant questions, date ranges, custodians, jurisdictions and deadlines. Review format, search criteria, potentially privileged material and confidentiality restrictions should be agreed before processing or production.

  • Preserved source material with its collection history.
  • Instructions for searches, exclusions and review boundaries.
  • Required metadata fields, native files and attachment relationships.
  • The intended review environment and delivery specification.

Possible deliverables

Agreed outputs may include a collection inventory, searchable review exports, an evidence schedule, search documentation and technical reports on specified issues. Processing decisions should document duplicate handling, conversation grouping and the relationship between messages and attachments. Broader hosting, review and disclosure services are described on the separate D3 eDiscovery website.

Keep meaning through processing

Deduplication can reduce repeated records without establishing that every source is complete. Threading can assist review while omitting important variations if treated uncritically. Export conversion may change representation. Retain the original sources and explain the processing steps so that relevant findings can be traced to the material supplied.

Boundaries and first discussion

Technical processing does not determine relevance, privilege or disclosure obligations. Those decisions remain with the authorised instructing team and advisers. Bcc, deleted-item status and envelope recipients are source-dependent fields. Begin with the review purpose and requested output, rather than assuming a standard spreadsheet can contain every possible metadata field.