eDiscovery in SA – South Africa using eDiscovery technology more than ever before, more tips, and reviewing for investigations versus litigation

Quite apart from my own knowledge of cases here in SA I am hearing from others of increased activity in our country so a huge congrats from me to all those involved in using and hosting eDiscovery technology in SA! One of my longstanding global eDiscovery contacts contacted me recently to say that he was aware of more activity in the last 9 months or so in SA than ever before. He is Steve Couling, MD and VP of Sales, EMEA at Relativity. For those who do not know, Relativity is the global market leader of eDiscovery technology solutions and their superb solution Relativity One is well and truly making a mark in SA. Steve wanted my advice and thoughts on how things were developing in SA as he continues to look at emerging markets in his region. I can tell you that there is a huge amount of data in SA currently hosted in Relativity (and other solutions) which is extremely gratifying. Litigation cases, arbitrations, investigations and enquires such as the Zondo commission as well as criminal investigations are all making more use of eDiscovery technology than at any time and there are literally millions and millions of electronic documents in SA currently in use.

SA has also been in the eDiscovery news as it featured heavily in a session of Relativity’s recent annual global fest. The session explored “emerging markets” of RelativityOne Partners and one of the panellists was Waseema Harrison, Head of eDiscovery at Insiox. She did a great job as I would have expected as she is my wife!

All of this led me to think about some of the questions I am often asked, as well as the latest news, tips, and comments on the eDiscovery global stage.

In his blog, my good friend Andrew King in New Zealand, reminded us of some of the more basic, simple and effective methods available to us in eDiscovery solutions. He makes the point that there are fantastic and powerful features within these solutions these days but sometimes it is the more basic ones which have been around for years that can make a difference. There are numerous examples and any service provider worth its salt would almost automatically perform these functions. Andrew reminds us of some of these features including ascertaining appropriate date ranges, key people, using keyword searches and others. He also mentions good practice such as; name normalisation (where the system is asked to regularise variations in the same person’s identity e.g Terry Harrison, TJ Harrison, Mr Harrison, Terry, are all one and the same person); domain analysis (where the system can ignore domains that are not of interest such as eBay, Netflix etc.; file types where certain ones are not required such as video files; and mapping communications which is a method of identifying who was communicating with whom and when. These are extremely effective features in the quest to filter and reduce the number of documents that need actual review and should always be considered (and used) before the more complex stuff like technology assisted review, concept searching, email threading etc.

Speaking of technology assisted review (TAR) or predictive coding, reminds me that we are seeing so many changes globally whereby these techniques continue to develop at a rapid pace and are in constant use. For more than a year now we have come to a point where there are different versions of TAR and the original concept (which I wrote about at length in my book ) is now phasing out in favour of Continuous Active Learning (CAL). I have to say CAL is truly phenomenal AI whereby the eDiscovery technology system is so sophisticated that it updates and applies each and every coding decision (relevant, not relevant etc.) to the remainder of the documents in the collection. Further, this happens literally every few minutes. ?It is therefore fantastic for prioritisation, privilege reviews, QA and review generally and much more. I suspect that in the new year I will devote an entire post to CAL!

?People often ask me about the differences, if any, of reviewing documents for investigations as opposed to reviewing for litigation or arbitration cases. This is very important as service providers who are adept at investigations are not necessarily adept at dispute resolution simply because they use the same technology, and they may not be able to properly advise on reviews etc. for dispute resolution. Even before review there are different tactics and processes to be employed. For example, those of us who are experienced in the industry would almost always recommend different methods of de-duplication of data in investigations from dispute resolution. In the former we generally advise what we call custodian de—duplication whereby in the latter we normally would advise global de-duplication (unless the specific case warrants a different approach). With custodian de-duplication the duplicates are set aside ONLY within the data of that custodian and not anywhere else in the total collection. Not only does this save time but it leaves those documents in other areas of the collection as they may be required for investigations of other custodians. In dispute resolution, on the other hand, a global de-duplication will have the effect of markedly reducing the total volume of the data which is of course the whole point.

Moving forward to review, investigations usually require a quick charge through the data to determine if there is a problem and if so, what it is. The features of the eDiscovery solution are employed to aid this deep dive into the data especially as there is often an almost impossible timeframe. Remember that the aim here is finding facts and not defensibility. With dispute resolution the methods and requirements are quite different. A more methodical and strategic approach is required which usually involves the management of a team of reviewers – hence the term “managed review”. Defensibility is key and why you need experts involved throughout this process to safeguard your discovery process in case it is called into question and has to be defended. This review often requires a number of different workflows and the whole focus is broader and inevitably takes longer. Leveraging technology is crucial, again, but the emphasis is about efficiency and proportionality.

Whilst on the subject of review it is well established globally, and I have referred to it on many occasions, that the cost of review in a case, accounts for approximately 70% of the total. That being so, how reviews are conducted and priced to the client are of huge significance. This is a big subject and again one in which I may devote an entire post in the new year. For now, I will leave you with this thought. The traditional method of charging the client by the law firm or by a provider to the law firm is on a per hour basis. An excellent and speedy reviewer would then generate less revenue than a slower less efficient one. Can that be right? There is a fast-growing global belief in the eDiscovery space that a per document model is more appropriate, more accurate and more “right”. Global corporations are wising up on this issue and will not pay for inefficiency. As I say, think carefully about this as it is another attack on the much maligned “billable hour”.