Fronting e-discovery with ECM

Enterprise Content Management systems enable organizations to work effectively with their unstructured content. ECM typically takes a more holistic view of an organization’s documents than e-discovery and enhances the ability to retrieve and analyze documents beyond what e-discovery is typically able to achieve. ECM classifies unstructured content, provides controlled access to it, and assigns granular retention schedules to it.

Here are some ways that having a fully functioning ECM system pays huge dividends in limiting litigation risk and expense:

More Precise, Richer Initial E-Discovery Collections

The ECM system can be used to collect documents for review, and document types that are clearly inapplicable to specific litigation can be excluded based on their document classification. The richness or prevalence of the resulting collection is much higher than brute force collection and this improves the performance of other technology like that used in technology-assisted review. With role-based access rules for content in ECM systems, attorneys can focus on the content accessible by or used by key actors.

Much Lower Volume

With granular retention periods, documents that are no longer needed can be disposed of before litigation arises. Further, documents that are not “records” can also be disposed of prior to the commencement of litigation.

In addition to excluding irrelevant document types, an ECM system will dedupe files and track key document attributes so that even when a document type has responsive information, only responsive documents within the classification need to be selected. For example, if only invoices from select companies during a stated period were relevant, invoices from other companies or from outside the time frame could be excluded. This may be difficult to do without accurate document-type classifications and extracted document attributes.


Because document-type labels have been assigned to documents, producing parties can give detailed accountings of what they reviewed and what was or wasn’t produced. The receiving party does not have to depend on the efficacy of some black box technology-assisted review program. Questions about the responsiveness of individual document types can be analyzed by sampling within those document types.


The producing party’s use of the ECM for its day-to-day operations lessens any concerns that the collection or production process is somehow engineered to lose or hide relevant content.

Accumulated Intelligence

Special processes such as extracting specific attributes of some document types for certain types of litigation can be automated and carried forward to subsequent similar litigation. The fact that individual documents were produced in individual cases can be tracked and managed within the ECM. Also, the intelligence accumulated to classify and attribute documents owned by the producing party can also be used to analyze incoming productions from other organizations.

If the ECM system uses visual classification, there are also other advantages that can benefit e-discovery:

Non-Textual Files or Documents

Visual classification reliably classifies even documents that do not have any appreciable text associated with them. By contrast, non-textual files are essentially invisible to text-restricted TAR technology, and in some collections that can account for 30 or more percent of all files.
Furthermore, by identifying non-textual documents or textual files with embedded graphics, visual classification can add text layers to these documents for use in text-restricted systems.

Document Unitization

TAR often implicitly assumes that there is one document per file. However, because of the way that PDFs are created, faxes are sent, or pages scanned, this assumption is often invalid. Visual classification creates logical document boundaries in multi-document files or in single-page image collections at an accuracy higher than manual unitization. Proper unitization improves the performance of all downstream functions in e-discovery.

Consistent PII Protection

The document classification and attribute extraction process inherent in content management will do a more nuanced job of identifying PII or other sensitive data, even handwritten forms. By contrast, text-restricted TAR systems may be limited to using text search/replace technology to identify or redact PII.

More information on how to manage unstructured content is available in my new book, Guide to Managing Unstructured Content. Free personal-use copies are available at

Comments are closed.