Disclosure is a growing headache for companies as the overall cost and risks have increased significantly in recent years. This is the result of three key issues: the increase in the volumes of electronic documents, the wide range of locations, formats and devices in which these documents are stored, and the number of languages in which the company communicates.
Large scale document collection and review exercises are not unique to common law jurisdictions where discovery is part of the litigation process. Even in countries with a civil code, very similar issues will be encountered in respect of regulatory investigations and the demands for information made by regulators are growing in civil and common law jurisdictions alike.
The more complex a process, the more likely it is that something will go awry. Courts and regulators will take a dim view of the failure to produce relevant documents in a timely fashion, while in the opposite direction, the consequences of disclosing privileged information can be severe.

Getting it wrong

Even those companies at the forefront of electronic information management can get it wrong. Last year, Google, in patent litigation with Oracle, inadvertently disclosed a crucial document after its discovery approach failed to identify that it was privileged. So pivotal to the case did the document turn out to be that the judge described it as ‘all that the opposing counsel would need’ to win the case, which it duly did (although Google is appealing).
A lawyer’s role in these exercises is increasingly focussed on project management. The lawyer must consider the type of matter, the desired result and the strengths and weaknesses of a number of approaches, and to select the right combination of human resources, technical solutions and workflow to minimise cost and risk for the matter in question. In some cases, a higher level of risk will be accepted as significant cost reductions can be delivered, and in others the opposite will apply. Therefore, a lawyer should have some understanding of the options that are appropriate for different types of matter.
Document review makes up the lion’s share of the cost of litigation, and technical solutions have evolved to reduce the amounts requiring review. The standard approach -- which is generally accepted by the courts and regulators -- is to perform keyword and date range searches over data collected from relevant individuals (often termed ‘custodians’) and remove any duplicate documents.
This process reduces volumes significantly, but care should be taken to perform at least some analysis of the performance of the selected keywords, either through sampling or early case assessment (ECA) tools, to ensure the results are worth reviewing. As well as providing the ability to see the search results, ECA-type solutions also provide useful advance information about the number and type of documents to be reviewed, which can be useful when negotiating the scope of disclosure with other parties and understanding which resources are likely to be required.

Predictive coding

A more recent development is the use of technology assisted review, often referred to as predictive coding. By applying sophisticated statistical algorithms to a controlled sample of a document set to be searched, it can accurately assess the likelihood of the remaining documents containing useful information by identifying the context of the document. This information can be used to either predicatively categorise unreviewed documents as irrelevant/relevant or to prioritise the review of documents the system has identified as more relevant.
Removing a large part of the human review element brings significant cost benefits, but there are risks. If the controlled sample is not representative of the wider data set then accuracy will suffer, so a sampling process should be built into the workflow to mitigate risk. Consider whether this technology is appropriate if the base data set includes documents in multiple languages. Be sure to ask the technical service provider how its solution deals with this issue.
Another option for reducing the cost of a disclosure exercise is the use of outsourced document reviewers for the early stages of the review where the focus is on sifting out irrelevant material, allowing the case team to focus on case strategy. Outsourced review costs vary, but it is rare to see a saving of less than 50 per cent against a law firm’s hourly rate for similar staff.

Risks management

Again, there are significant risks to consider here. Is this the type of case where a relevant document can be easily and clearly defined? If there are multiple languages involved, will the outsourced review vendor be able to find staff proficient in those languages? How will feedback between the legal team and the document review team be shared if they are separated by thousands of miles and several time zones?
As the industry has matured, the solutions become better aligned to the problems being faced by consumers. It is now possible, for example, to use outsourced document review providers that provide multilingual staff, and are based in the same city as the legal team.
So while e-disclosure technology and outsourcing can significantly reduce the cost and decrease the time taken to complete a disclosure exercise, or respond to a regulator’s information request, lawyers still need to select carefully from a range of options. Making the time to understand the strengths and weaknesses of the solutions in advance will ultimately pay dividends.

Drew Macaulay is a Director in the London office of technology company, First Advantage Litigation Consulting, a global specialist in computer forensics and e-disclosure for corporations and the legal profession