| Screenshots | Fast review, visually readable messages or posts, initial client intake | Visible content, interface context, timestamps or identifiers shown on screen | Hidden metadata, material outside the crop, complete thread/account context |
| PDF / copied text | Readable packets, notes, correspondence, selected excerpts | The text and layout included in the created document | Native metadata and direct linkage to the underlying source may be reduced |
| Email / secure upload / cloud link | Transporting files already collected | Depends on the files being transferred | The transfer channel itself does not fix a weak collection method |
| Native files | Documents, spreadsheets, media and other file-based evidence | File contents and available native metadata | May not capture account-level or application context around the file |
| Account exports | Cloud services, messaging history, AI chat history and platform data | Provider-supplied structured data and broader account context, depending on the service | Can include substantial unrelated private data; formats vary by provider and over time |
| Forensic collection | Matters where source integrity, deleted data, device state or detailed metadata matters | Potentially extensive source information using documented technical processes | Cost, scope and privacy impact can be higher; not automatically necessary for every dispute |
| eDiscovery workflow | Larger litigation collections and controlled review/production | Structured processing, review decisions, production metadata and auditability | Quality still depends on what was collected and how source data was preserved |