The disclosure of all relevant documents has traditionally been treated as a right of the parties to the litigation, but Australian courts have increasingly placed limits on that right. The courts generally attempt to encourage the efficient disposition of cases and contain legal costs by focusing on the real issues and limiting discovery in a manner proportional to those issues and what is at stake for the parties.
In that context, the use of modern technology including electronic discovery has long promised costs savings and other efficiencies, but has it delivered?
Court approach to electronic discovery and hearings
Most superior courts adopt a similar (although not identical) approach to the Federal Court where electronic discovery can be ordered under rule 20.14(4) of the Federal Court Rules 2011(Cth). The Federal Court’s Technology and the Court Practice Note (GPN-TECH) introduced on 25 October 2016 and a Standard Document Management Protocol are also relevant, but are not as specific or detailed as its prior electronic discovery practice note and related materials. That may have been intentional to provide for flexibility of practice.
Generally, in any proceeding in which a significant number of relevant documents have been created or are stored in an electronic format, in order to facilitate the quick, inexpensive and efficient resolution of the matter, the court may order that:
- discovery of documents be given in an electronic format and in accordance with a discovery plan, and
- hearings be conducted using documents in an electronic format.
The court expects the parties to use technology in the management of documents and conduct of the proceeding, in particular for:
- creating lists of discoverable documents, giving discovery by exchanging electronically stored information and inspecting discovered documents and other material;
- lodging documents with the court and delivering court documents to, and otherwise communicating with, each party, and
- presenting documents and other material to the court during hearings.
Efficient document management and metadata access
The courts also encourage efficient document management by:
- discouraging printing electronic documents and photocopying paper documents;
- encouraging or requiring the parties to exchange documents in a usable, searchable format or in the format in which the documents are ordinarily maintained, and to allow the party receiving the documents the same ability to access, search, review and display the documents as the party producing the documents, and
- requiring the parties to plan for appropriate discovery as early as possible in the proceedings with discovery plans and agreed protocols for the electronic exchange of documents.
Disputes sometime arise as to the form of access to documents and extent to which access to electronic document metadata should be permitted. Cases are fact-specific (see, for example, Jarra Creek Central Packing Shed Pty Ltd v Amcor Limited  FCA 1802, Tamberlin J at -), but generally:
- discovery requires production, not creation, of documents;
- production can require production of documents including metadata, but there can be technical and/or hurdles to obtaining full access, and
- a discovering party will not generally be required to expend significant sums to change or create documents in the most usable form merely at the request of the party seeking discovery.
Challenges to efficient electronic discovery
Two main issues arise in relation to electronic documents and discovery:
- the time and resources of the parties and legal costs required to conduct all reasonable appropriate searches for relevant documents and collate them, and
- the legal costs involved in giving and obtaining discovery effectively.
Reasonable search and predictive coding
Predictive coding technology promises effective searches of electronic documents in conjunction with significant time and costs savings.
While predictive technology has not yet been widely approved for use in discovery by courts in Australia, decisions of American, Irish and English courts have endorsed the use of the process. For example in the United Kingdom see Pyrrho Investments Ltd v MWB Property Ltd  EWHC 256 Ch reviewed the process in detail and approved of it.
Approval to use predictive coding was also sought in Australia and approved in McConnell Dowell Constructors (Aust) Pty Ltd v Santam Ltd & Ors (No 1)  VSC 734.
The process involves a review of electronic documents by software rather than lawyers. The software codes the documents for relevance based on a pre-determined data set which, in combination with a predictive coding protocol is created by lawyers in conjunction with IT support and agreed between the parties. The data set and protocol address numerous variables including sample size, batches, control sets, reviewers, confidence level and margin of error.
Criteria are agreed for the selection of documents as relevant by the software, for example specified custodians of documents, date ranges or the presence of certain keywords, or some combination of variables. A representative sample of documents selected by the software is then used by an experienced lawyer who has mastered the issues in the case to train the software to apply the criteria for relevance consistently. If any documents have been selected by the software but are in fact irrelevant, these are removed so that the software does not learn from these and will not code them as relevant in the future. The results are analysed, and the process is repeated until an agreed margin of error is reached.
Predictive coding software is likely in time to be accepted more generally in Australia in cases involving voluminous material requiring review for the purpose of discovery, in particular because:
- the use of a single, consistent algorithm in some cases might be more accurate than having numerous junior lawyers or paralegals reviewing documents, and
- it becomes progressively more cost-effective as the number of documents in a matter increase once provider costs have been paid.
Giving and obtaining effective discovery
However, on the other hand there remain concerns about duplication and the unintentional disclosure of privileged information in electronic discovery, and it is prudent to seek and obtain court approval for predictive-coding based discovery to ensure that “reasonable search” and related obligations under court rules and otherwise to the court are complied with.
It is also important not to assume that electronic discovery is a panacea which can be delegated to third party providers, or that the technology is necessarily efficient and/or foolproof. Advances in predictive coding are promising, and it is now routinely used in some jurisdictions for some matters. However, there have been many examples over many years of parties embarking enthusiastically on electronic discovery as a time and costs saver only to find that:
- searching fails to discriminate adequately between emails, email chains and other identical documents and multiple copies of identical documents can be discovered and multiply costs;
- searching can fail to identify privileged documents;
- searching can yield voluminous objectively irrelevant documentation which needs to be reviewed at considerable cost by lawyers, and
- errors in loading and/or coding documents electronically can give rise to difficulties properly listing, searching for and using relevant documents – the “rubbish in, rubbish out” principle applies.
Active supervision of the setting of all search parameters, monitoring of all search results and comprehensive accurate loading and coding of only truly relevant unique documents, by the senior lawyer(s) conducting the case, is essential for electronic discovery to be friend rather than foe.