Are your eDiscovery processes pandemic and recession-ready? Are coronavirus-related cutbacks shrinking litigation budgets and forcing you to do more with less? Looking for ways to lower your eDiscovery spending in good times and bad?
Join us for our webinar about how to make your eDiscovery processes pandemic- and recession-ready. We’ll be discussing how you can leverage technological advances and improved workflows to save time and money, preparing your organization for whatever comes next.
Learn how today’s eDiscovery tools are particularly well-suited for this moment in time, and how they’ll continue to be effective in the future:
- Establish a Standard eDiscovery Business Process
- Centralize Your eDiscovery Processes
- Leverage Custodian Questionnaires
- Use Remote Data Collections
- Use Targeted Data Collection Methods
- Leverage Data Processing & Early Analytics
- Utilize the Power of Machine Learning Technology
Watch the Webinar
Webinar Q&A Session
When is the best time to issue a custodian questionnaire?
While some clients will issue at least a preliminary custodian questionnaire at the same time as their legal hold notice (many software solutions enable that), most often we see clients wait a bit after the initial legal hold notice is sent out, to allow more time to develop a truly effective custodian questionnaire. That said, custodian questionnaires are by no means a one-time event. While you don’t want to inundate your custodians with new questionnaires every week, as the case develops and the need presents, you can always issue additional focused questionnaires to the entire custodian group or a subset as needed. Remember: custodian questionnaires are an excellent way to gather case facts and other information, so don’t be afraid to use them whenever you need to gather such information.
Who should be on the eDiscovery steering committee?
This was discussed in the first part of the webinar, so you might want to view that portion as well. That said, you should include internal resources (an executive sponsor, IT, HR and Legal as available), your outside counsel and your preferred vendor. It’s important that you include all of the people who normally would be involved in an eDiscovery process, otherwise, your plan will fall short of being truly comprehensive.
You mention creating a written eDiscovery plan; does BIA offer this service and can you share a sample?
Yes, we do. If you would like to discuss this further, please contact us and we’ll be happy to show you a sample of the plan.
When should we image computers as opposed to collecting live data?
We recommend that you only make forensic images of hard drives or other data sources in cases of potentially criminal acts, employee theft, certain human resources complaints, and cases where you suspect that a given custodian may have tried to destroy data. Essentially, only do it in those cases that require a deep-dive investigation into the activities of that custodian on that computer. For most civil matters, collecting the live data will absolutely fulfill your obligations.
What’s the advantage of imaging a HDD, if imaging and live data collection are both forensically sound?
By imaging a hard drive, you can do a comprehensive forensic investigation of the custodian’s activities, as well as potentially recover deleted data. An image will preserve all of the system files, logging, event records, registry settings, browser histories and the like. So, for example, if you are investigating a possible employee data theft, a forensic image will allow you to determine what external media devices were recently attached to that computer (i.e. USB thumb drives or external drives where data may have been copied to as a way to steal that data). You can also reconstruct a user’s activity through a blend of event tracing, log reviews, registry settings, browser history review and more. Essentially, the forensic image gives you the ability to see *how* a custodian used their computer, which is not usually subject of most civil litigations.
In addition to the legal industry, what other industries use eDiscovery?
Most Discovery is eDiscovery, so any industry faced with litigation is likely to encounter eDiscovery and will need proper tools and solutions to handle it. But again, the benefits of having the right eDiscovery tools extend far beyond their usefulness in litigation. Over the years, we have helped the legal departments inside corporations use eDiscovery technologies for a number of other functions. For example, customers have leveraged their legal hold notice software to also track other important legal and corporate notices such as HR Manual updates, Security/IT notices, Corporate Policy changes and more. Anywhere you would like to have a full record of who received a given notice or information circular (and even have the ability for the recipient to acknowledge those notices), you could leverage your legal hold solution for those purposes very easily. Similarly, we’ve helped clients use their Custodian Questionnaire software (usually a part of a legal hold software solution) for any number of confidential or important internal surveys or tests, like sending out a survey to measure employees’ knowledge on how to recognize and prevent email phishing attempts. One client used it to gather feedback on their benefits programs during their annual renewal efforts. Anytime you’d like to either better understand your employees’ knowledge on a given subject or gather their thoughts on a particular topic, you could leverage the Custodian Questionnaire solutions for that too. We’ve also helped customers utilize data collection solutions not just for litigations, but for their exiting employee processes, remote system gatherings and the like. We’ve even seen companies utilize machine learning (TAR) and analytics solutions for such things as tagging documents for retention categorizations and more. The tools used in eDiscovery are powerful data-capturing, handling and analytic solutions, and you can put them to good use outside of litigation as well. Using these tools to answer everyday needs is a great way to leverage your legal spend across more than just your legal matters and internal investigations.
Mike: Hello everyone, and welcome to the webinar channel of the Association of Certified eDiscovery Specialists. My name is Mike Quartararo, and I am the president of ACEDS. Thanks for joining us today. We are joined today by our partner BIA for a webinar entitled, “Seven Steps to Make Your eDiscovery Process Pandemic and Recession Ready.” Before we get started, please know that we love questions. We’re always happy to take questions. You can submit your questions using the Q and A widget at the bottom of your screen. Also, if you’d like a copy of the slide deck from today’s presentation, you may also download that at the bottom of the screen. Alright. Without further delay, I am pleased and very excited to introduce our presenters from BIA. BIA is an eDiscovery and digital forensics firm, protecting the enterprise legal risk and unnecessary extra costs, trusted experts for two decades in the industry. Without further delay, I will turn it over to Mark MacDonald for the presentation today. Mark, take it away.
Mark: Thank you very much, Mike, and thank you to everybody who’s attending today’s webinar. We are super excited to have you and appreciate your time. As Mike said, please feel free to submit questions during the interface. We’re going to try our best to answer questions live as we go through the presentation. We’re also going to try our best to keep 10 or 15 minutes towards the end to answer additional questions. If we don’t get to your question, just know that we will answer it. It will be on our blog, and we will send everybody who attends a copy of the deck, as well as answers to all the questions. Today’s presenters are Barry Schwartz, and Barry is our SVP of advisory services. Lisa Prowse, our SVP of technology and managed review, and Brian Schrader, one of our original founders and president of BIA. As Mike said, our seven steps, even in the absence of the world’s current situation, embracing these strategies and workflows, will absolutely have a positive effect and possibly an immediate one on your discovery budgets and how you manage your workflows. So without further ado, Brian, take it away.
Brian: Well, thank you, Mark, and thanks to everybody for joining us today. This is an important topic, really kind of like as Mark said, regardless of the times. Even before the Covid-19 pandemic hit us, a lot of economists were predicting that we were overdue for a recession. And so whether it’s a pandemic or a recession, or just really even in good times, there’s no reason to spend more money than what you should. And there are some great ways – some of them are workflow techniques, some of them are approaches to how you handle eDiscovery, some of them are better technologies that we’re going to discuss as ways to reduce your legal spend. And it’s at every step of the workflow. We’re going to start out at the very beginning of the EDRM model and move towards the end. We’re going to cover – everybody talks about TAR – we’re going to hit on machine learning and how that can really help reduce your legal spend. But it’s important to think about it from the very very beginning as well. And throughout your entire process, examine every single step. From how you’re issuing legal holds, to how you’re collecting data and how you’re processing it, to where you’re hosting it, how you’re reviewing it, and what steps you can take along the way at each one of those steps. And we’ve got about seven steps today that we’re going to go through, and address how you can look at saving money from the very outset, all the way through to the end of your eDiscovery process.
Maybe not too surprisingly, where all of this starts is establishing a standard eDiscovery business process. So a lot of companies have standard processes for how to handle data breaches, or even how to handle security, or all sorts of other – and you know a lot of companies have now formed pandemic response task force. The idea of a task force like this isn’t necessarily something that has to meet on a monthly basis or whatever. But the idea is anybody who has ever looked at any kind of business efficiency processor or anything like six sigma or anything else like that, the very basic idea that kind of crosses all of those to a large extent at least is this idea of if you can’t have an efficient process, you’re going to end up wasting a lot of money in anything you do. Whether it’s something that’s a daily workflow in your business or a relatively rare event like a lot of litigation is for a lot of companies, whether it’s not something they handle every day.
But the simple thing that kind of cuts across all of those is this idea of planning. Without a plan, without having this established policy. I’m sure everybody on this call has experienced this – jumping into an eDiscovery without any real plan, and you’re constantly jumping between resources, finding out things you didn’t know existed now you have to deal with, kind of dealing with each phase of the process as it comes up without really thinking about it in a larger scale plan. And this goes for corporations and law firms. We’ve helped both actually create kind of default written protocols. So it’s not just something that you may think a corporate or an organization should worry about, but also law firms. And it all boils down to one basic idea. If you create this plan and you follow the plan, and you perfect it over time, you’re not reinventing the wheel every time you walk into litigation. And even if you think ‘oh we don’t reinvent the wheel all the time’, unless you really have a documented plan and a process, you are reinventing the wheel to some extent as you go along.
So how do you establish that? How do you create that standard process? Well, the first thing you want to do is create what we call a legal event team. You can call it a legal committee, you can call it whatever you’d like. We call it a legal event team. And the reason it starts there is that you have to look at the members of that team and make sure you have the right people. Because if you just bring in two or three people who may be involved in litigation, but you don’t think about bringing in the entire spectrum of inside resources. And when I talk about inside resources, it’s not just legal. It’s also having somebody on the executive team, so they understand the impact, having somebody in HR, having somebody in IT, because those people know who the people are and where the data is. Outside counsel, even your litigation support and vendors, if you have a vendor that you use on a regular basis. Bring those people together as a group to put together your plan in the outset. It doesn’t really cost a whole lot of money, but having all of their input and coming up with your standard process is key. And if you don’t include those individuals, you’re just cutting yourself short. So it’s definitely something. And once you have that team, you can put together the standard business process. And the last point on this is education is key across the board. Once you have that team put together, then you can talk about creating your written plan. And your written plan should cover the who, what, why, where, how of every step. Who’s going to identify custodians? Who’s going to issue and track legal hold? Who’s going to conduct and monitor data collections? Who’s going to ensure automatic processes like deletion and things like that, automatic deletions, get suspended where needed? All of this stuff. If you literally just use the EDRM as your model and walk through it and say okay, well, the first step obviously is issuing legal holds and custodian questionnaires and data collections – and how are we going to handle each one of those steps? You don’t have to necessarily get extraordinarily granular. And we recommend that you don’t get too granular on these plans because again, they’re model plans that you use as a starting point for every litigation. If you haven’t gotten the idea from the last five, six minutes of me going on on this topic, the single most important thing you can do to control your eDiscovery spend is to come up with a standard process. And to boot, if you come up with this standard process, it also helps make sure not only is your process more efficient, more effective, but it also is more defensible. Because if you’re doing it the same way every time and you’ve used that process over and over again, it’s much easier to go into court and say, your honor, look, this is how we’ve done it in the last ten litigations or whatever, and here’s our process with written documentation. And it also helps dealing with opposing counsel in cases and being prepared for your meet and confer conferences and things like that, and even 30V6 depositions.
And the last kind of thing in putting together this process is to think about both how you’re going to approach. Not only who is going to be involved in each one of those steps, but how you’re going to approach it as well. So for each one of those steps, whether it’s legal holds or data collections – are you going to be using manual processes? Are you looking for enterprise-level software solutions for those things? Like I said before, legal holds, data collections, analytics and review solutions, and even your document review process – is there a preferred process for you to use for document review? We’ll talk a little bit about that more in-depth later. All of these things, if you kind of walk through them, they’re not really that specific to an individual case. Meaning if you’ve got them – you don’t have to create like one process for your intellectual property, one process for your employee issues. You want it to be broader than that. But these big questions if you can answer them and make them part of your standard process, to begin with, you’re going to save a ton of money as you walk through. Barry and Lisa, do you want to throw anything before we go on to the next step here?
Barry: Sure, thanks, Brian. A couple of points I’d like to add to our step one is that having the written plan gives you continuity between matters over matters, and it minimizes disruption within your organization. When you have an event that causes the need to implement your plan, you know where to start. And even when the organizer identified that that event has occurred, the triggering event. And we have webinars on our website that discuss that in a fair amount of detail. But knowing when to implement the plan is key because then the steps that Brian outlined – issuing a legal hold, identifying custodians, and so forth – come into play as a result of that. And having a plan does minimize disruption within your organization. Lisa?
Lisa: Yeah, the only thing that I would add to that is also having that plan allows you to bring other people into that process. For example, if you’ve got people who aren’t able to work because they’re out sick or something else. If you’ve got everything really well documented, especially when it comes to document review portion, you can bring other people on and try to get them up to speed quickly.
Brian: Yeah, that’s a great point, Lisa. I even say that’s all the more important in modern times like this where having someone suddenly unavailable, whether it’s because they personally have an experience with Covid-19 or because they have a family member that they need to deal with – or not deal with it – to help I should say, to take care of. That just sounded really strange. That kind of absence is happening a lot more, and it’s something we need to deal with more. So having that plan definitely makes it easier for people to step in without, again, trying to reinvent everything all over again.
So something that kind of goes hand in hand with the first step, but it is something to consider separately as well, is the idea of centralizing your eDiscovery benefits. Barry, you’ve talked a lot about this in the past, so why don’t you tell us a little bit about the idea behind this?
Barry: Thanks again, Brian. The keyword in the first bullet point is defensible in our opinion because doing what you need to do on a repeatable basis consistently matter over matter allows for you to not miss information that needs to be addressed during dependency of a legal matter. And having the process where you know exactly what the steps are that need to be followed. And as Lisa said, having backup support who can step in on an as-needed basis if people aren’t available aids in reducing your risks and increases your security and control over the data. And interestingly, yesterday in Legal Tech News and today in Law 360, was an article about an old database that was made available online that contained data from over 200 different law firms. And one of the things that we talk about repeatedly at BIA when we’re talking with our larger corporate clients that have multiple matters spread over many different law firms is the need to corral your data so that you know where it is in the unlikely, but in this case, it did happen to several hundred firms, have your data everywhere, so you know where it is so that the risk of it getting out into the wild. If you’ve even forgotten that you used law firm ABC years ago, you know where your data is, and it’s not at risk of loss or misuse. And again, reusing your data matter over matter increases your efficiency, particularly in a couple of significant areas. One of them is you have serial custodians who have similar documents that are collected matter over matter, and they’re reviewed multiple times. Well, you can use that review to aid in the prior review in your current matters—the same with privilege considerations. Again, you have corporate counsel, you have senior management to receive documents, email, correspondence and so forth from your counsels, and those coding calls can be reused in a secure database matter over matter so that you’re not reinventing your privilege calls, your redactions, and so forth. It increases efficiency. It increases accuracy. It helps prevent any later privilege considerations. Brian?
Brian: I was just going to throw in there, this, like I said, goes hand in hand with creating a standard workflow. You really kind of have to standardize on the solutions you’re using, right. And the side benefit is not only does it blend in and make that workflow that you’ve created more defensible when you have everything centralized, and you do it using the same tools and processes to a large extent. But whether you’re buying your review tool for example, whether you’re buying it from a vendor or you have your own that you use, you have a direct contract with a software provider – volume in this industry, at the end of the day it’s all talking about data storage. And the simple fact of data storage, the more data you have, the more per unit model pricing you get. And that’s true not just for the discovery software. It’s true with Azure and AWS and everybody else in the world, right. The more you can consolidate and centralize and commit to and bring in all that stuff together. This isn’t just the legal industry. We use Azure quite a bit, and the more data we centralize there, the cheaper our per-unit cost got. And so you see the same thing here. And like I said, it doesn’t really matter whether you’re buying it direct, you’re buying through a vendor, or doing it yourself. Centralizing all that stuff is going to reduce your legal spend significantly just because you get volumes of scale, you get the economics of scale. So just throwing that out there. I know you touched on it a little bit, but I just wanted to highlight that.
Barry: Good point. And one other thing on the data breach that I mentioned. The technology company, the security info, the security company stated in their report breach data of this nature usually ends up in the wrong hands. So again, as Brian was just saying, having a centralized location where you’re not worried about corralling your data on a long-ended project is in one spot increases security, increases control, and helps manage your costs when you need to retire that data as well.
So I think we can move on to step three, which is leveraging custodian questionnaires. And when you’re dealing with eDiscovery, it’s important to focus your attention on the custodians who may or may not have data to know which ones have the data and which ones you don’t need to talk with. Using a custodian questionnaire is one of the most efficient means of ferreting out which are the important custodians and which ones aren’t, and also identifying data that you may not have been aware of or weren’t thinking about when your legal event team met at the outset of the matter. And using a well thought out custodian questionnaire today helps you in identifying that information. In the old days, it used to be that custodian questionnaire was a PDF that was sent to custodians, and they would handwrite it, and maybe in more recent times, they could type into the form and send it back. And then, perhaps those were collated and reviewed and maybe summarized and may be utilized in live interviews. With today’s technology, a custodian questionnaire could be distributed through a web portal, and the form completely electronically, and the responses collated in such a way that you can compare and contrast one custodian’s response to the next. Not so much – well, I’m Joe Smith, and this is my email address and so forth – but exactly what Joe Smith’s role is and what the involvement was with the matter at hand and any other pertinent information such as unique databases that may have been utilized. In the construction industry, there’s a lot of project management software in play, in the medical, in the pharma world, lab notebooks are in play. And to have the custodians identify those resources at the outset of the matter again goes to the notion of efficiency and identifying the key information early in the case so that as the attorney’s counsel that need to actually dig into the data and realize what the strong points and the weak points of the case are and the important parts of the case are, the custodian questionnaire helps do that. It also helps you.
Brian: I’m just going to say really quick – very sorry, didn’t mean to cut you off there. I think that’s one of the most critical points is people think of custodian questionnaires – the first thing that pops into your head is like, oh yeah, we’re going to send out a custodian questionnaire to find out where data is. And that’s important. And like you said, it’s a great way to find out about data, especially in non-standard systems like those industry-specific processes. And even a great way to find out whether or not custodians are using unapproved. Like are they using Signal? Are they using Onenote? Or Evernote? Or something that’s not necessarily corporate sanctioned but would still fall under the corporation’s duty if their employees are using it. So there’s that kind of stuff. But the other thing you touched on is this idea of using these things to gather facts in a case. And people don’t really think about using custodian questionnaires that way. Finding out and asking people about their involvement in a particular lawsuit or a particular claim or an event or was there involvement in the development of a new product that might be involved in an IP case. This idea of using these custodian questionnaires for a much broader purpose than just finding data. That’s great, that’s easy, but there’s a much more powerful leverage. And not only does that help you gather a centralized database of all this information for all these different custodians, but like you said, it helps the legal team then focus on those key custodians that matter and not waste their time doing insight or even remotely over the phone interviews that can get much more costly than the custodian questionnaires. They don’t have to waste their time – and I don’t want to minimize somebody – but they don’t have to waste their time with the people who don’t necessarily, weren’t necessarily really involved in a particular case or a matter. So it allows you, think of it as kind of a filter. It allows you to gather enough facts about all your custodians so that you can then counsel can filter down and get to the ones that they really care about, and they really need to talk to. And they can even spend more time doing that, spend more time with those key custodians. And so I think a great way to think about this when it comes to the fact thing is it allows you to filter down, and it also avoids custodians from having to be kind of intimidated by oh you’re going to be interviewed by an attorney tomorrow if it’s not really necessary for them to be interviewed. So employee disruption concern there goes away too. Sorry, I just wanted to kind of throw that in there and make sure we got to that idea that it’s not just about finding data, it’s actually better to use for fact gathering.
Brian: Yes, Mark?
Mark: Sorry, we had a question come in that I think is probably appropriate for this point, which is at what point do you actually issue a custodian questionnaire? Would you do it before a legal hold goes out? Or traditionally after a legal hold goes out? At the same time? Maybe you could just take 30 seconds and answer that.
Brian: Legal hold definitely comes first if you have to prioritize. The very first thing you ever want to do is get a legal hold out. And we have a whole webinar on our website, “The Ethics of Legal Hold,” that really talks in detail about how to issue legal holds, how to do it right, triggering events, and all that kind of stuff. So that’s an hour and a half discussion in and of itself. But most clients that we see will send out the legal hold first and send out a custodian questionnaire later because it gives them a little bit more time to think about the substance that they want to include in that custodian questionnaire. Now, most legal hold software solutions out there will allow you to do both at the same time. You send them a legal hold notice, and they get a link to answer the custodian questionnaire. But we found that most companies like to do the legal hold first, and then think a little bit about the custodian questionnaire and do that. Again, that does bring up an interesting point too. Don’t think of custodian questionnaires as a one-shot deal. You can issue additional custodian questionnaires even to the same custodians multiple times. I mean it’s not like you want to hit them every day or every week, but you know six months into litigation you might have some other questions that you’ve discovered from all your discovery, and research, and working on the case that you’d like to ask some more questions. Feel free to send them. We see people do that all the time. Go ahead, Barry.
Barry: And we see that different groups of custodians may get different sets of questions. The marketing team gets one set of questions, the production team gets another, and the engineering team gets a third set of questions. Some of the questions obviously will be overlapping between the various iterations of the CQ. However, they can be more pointed to the audience, the custodian base, the set of custodians that you’re sending it to. One thing we have found over the years is we can generally trust the custodians to do the right thing. And another point with custodian questionnaires is that from an ethical and legal responsibility basis, counsel at the end of the day has to sign off on the completeness of discovery. And one of the ways to help be assured that you’ve not left any stone unturned is to issue the appropriate custodian questionnaires, whenever you do that either at the outset of the matter or throughout the term of discovery, as Brian was saying. And rule 26G requires that certification. And the one case we have cited here, A & M Florida Properties, counsel must affirmatively act to communicate with the client to identify all sources of information. And that is one of the primary uses of the custodian questionnaire is you’re asking those questions; you’re getting responses, you’re evaluating the responses. And maybe you have to go back and ask again, or during the folks’ interviews that may happen. And a very valuable tool. And again, one point we like to make is CQ’s are attorney-client, attorney-work product, and so they’re protected by the privilege considerations. So I just want to put that point on the table as well. And back to you, Brian.
Brian: And so the next step is the first part where you really get into how do you control the costs? And for anybody who’s familiar with the electronic discovery reference model, or the EDRM, one of the background pieces in that graphic, like literally the background graphic, is that you start out that workflow with a large amount of data, and you reduce that data every step along the way. And one of the things that you can do from the very beginning to help reduce the amount of data you’re collecting as well as reduce the cost in collecting that data itself is doing remote data collections. This is especially true today when nobody’s in their office, to begin with. And you’ve got to collect data from a lot of remote places. Now, sure, a lot of stuff is now cloud-based, and it’s a lot easier to collect cloud-based data because of the centralization of it, but you still have to collect a lot of time from individual computers and even from servers in other locations. And it is incredibly easier to do that remotely today than it ever has been before. And when I talk about remotely, we’ll talk about some of the solutions available here in the next slide. But when I talk about remotely, I’m not necessarily talking about shipping out hard drives, which is kind of what most people think of when they think of remote collections is you have to ship out hard drives to the individual custodians, and they have to plug them in, and put in passwords, and run some software, and do all sorts of other stuff. And even that in the era of Covid-19 is not something that most people want. I mean they get this idea of people still aren’t very comfortable with exchanging even physical goods in a lot of circumstances because they’re concerned about whether or not they’re contaminated. So I don’t want to like put a lot of alarm bells, I’m just saying, even more, today, remote collections not only are they cost savings, but that allows you to get collections done, but otherwise you might not be able to. Because right now, you can’t send people on-site to do hard drive images, which is kind of the old fashioned way of doing things. We’ve been arguing for remote collections for years. But at the end of the day, if you compare the two, I have yet to see a single case where somebody was sanctioned because they did forensic collections or because they did forensically sound – we’re still talking everything’s forensically sound – I’ve never seen somebody get sanctioned or even slapped on the wrist or anything because they did remote-based collections versus forensic imaging. And why is that such a kind of such a big factor is because today you’ve got these solutions that are cloud-based and there’s a number of them out there where you can literally collect data from anywhere in the world, from any computer in the world, as long as it’s connected to the internet. And you can do it securely, you can do it defensibly, you can do it with data encryption, and you can do it in a way that follows the same type of forensics processes that your boots on the ground collections would. And so if you’re still paying for teams of people to fly all over the country, or even go to one location on-site and image hard drives, you gotta think not only are you incurring a lot of extra hourly costs, probably extra equipment costs, probably travel costs, but you’re also significantly disrupting your employees. You’re creating chaos in the organization, and you’re also literally kicking people off their computers and cutting into the productivity of your employees, and there’s no need to do that. You know, like I say, compared to, you wouldn’t send out your paper vendor to copy every single piece of paper in every filing cabinet and on every desk in every office of your client. That’s what you’re doing with forensic imaging. If you’re doing remote imaging, you can do it a lot cheaper, a heck of a lot faster. With the technology we use today for remote collections, I literally can have one project manager collect from 500 custodians with two clicks of a button and watch it all proceed while he sleeps. It is so much more efficient, and it is so much more cost-effective than if you’re not doing remote collections, you definitely, definitely, should look into it.
And hand in hand with that here’s a little quick guide. There’s how to do remote collections. There’s everything from drag and drop, which is extraordinarily low cost because there’s really no cost to it, but it’s not defensible at all because dragging and dropping is hopefully as everybody knows going to mess with metadata. But this gives you kind of some idea of defensibility versus costs and the various options. I mean, you have very highly defensible, very costly solutions like the old autonomy, enterprise vault, or things like that. You have DiscoveryBot, which full disclosure we created it, but we sold it off years ago, so we don’t own it at all anymore, but it is one of the tools we use quite a bit. And there’s other remote collection tools out there as well. At the end of the day, when you’re talking about remote collections, this little box on the left-hand side here says, ‘forensic principles should guide you, not control you.’ This isn’t a criminal case. If you’re talking about criminal cases, sexual harassment cases, things where you think people might have deleted data, then sure go out there and image hard drives because you might need to do data recoveries or do intensive investigations. And if you’re doing an investigation in forensics, you need that image. But if you’re just responding to standard discovery requests, collecting data, think about how you would have done it in the paper world. It’s no different here. You should be going out, talking to custodians, figuring out where that data is, and collecting the data you need, and not doing full forensic images of everything. Because all you do is end up with so much more data than you need that you’re going to have to pay somebody to hold onto, you’re going to have to pay somebody to filter it all out. You’re going to have to pay somebody to process it all. You’re going to have to pay somebody to hold that data and hold that original archive stuff. Even though you can filter it out along the way, it does increase your costs. The earlier you can reduce your data and eliminate costs in the process, the faster it ripples throughout the rest of the process.
What goes hand in hand with this is not only using remote but using targeted collection methods. In other words, again, not imaging and entire hard drive, not doing a remote collection of an entire hard drive, but focusing in on user data files is what we always say. There’s no need to collect a bunch of files that you’re going to end up paying someone to get rid of later on. And in some circumstances, if it’s an HR complaint, there’s no need to go collect every file on an HR department server if you can narrow down and identify the server if you can narrow down and identify the files that may be an issue in that case. Same way with intellectual property. If you have intellectual property patent dispute, there’s no need to collect every single intellectual property thing in the company, or all the files of anybody involved. You use those custodian interviews to narrow it down to the right people, and then to the right data. And the more you can do that, every case is different, of course, but if you start thinking at the very outset at the point of collection – how can we target the data we need? Some of the easiest ways are data filters. We always say you never want to necessarily collect data based on search terms unless you’ve got really good enterprise systems that allow you to do that easily. But you never want to collect based on search terms because search terms change, and it’s hard to do those collections in a really effective way without pre-indexed data. But date range filters, if you have a case that goes form like 2018 to current, maybe collect from 2017 to current is a way to easily narrow down some of that stuff. There are a lot of different things you can do. And talking to the custodians and understanding your facts before you jump in and start collecting data just from everybody and every source is a great way to make sure you’re focusing on what you need and not gathering a ton of stuff that you’re never going to use. Especially now, when data storage is so cheap, inside of corporations, you no longer have a lot of companies that do automatic deletions every six months or a year or whatever. So you end up with custodians with years and years and years – we have companies that we deal with that have email going back to the early 2000s and beyond. And if you’re talking about a more recent litigation, the last thing you need to do is collect all of that stuff. And of course, using targeted collections, using those things, you get rid of all the system files, and things that you might collect otherwise from more old fashioned data collection methods.
Mark: Hey, Brian?
Mark: I’m sorry, another question just came in that I think might be interesting to answer right now. This one’s from Sarah. I’ll paraphrase. An ongoing case has just been, the timeline has been extended. And apparently, the email data has been deleted, but it exists on the backup tape. Are they required to go to the backup tape because the lengths now in question has been increased? Or can that be avoided?
Brian: Good question. So basically, your timeline or time limits or date filters have gotten extended. And the only place to get that data from is on backup tapes. You can get into, whenever you talk about backup tapes, the very first thing that comes up is you, going back to Judge Schinland’s cases from the early 2000s, are still some of the best on those topics. And it’s all about balance, right? You’ve got to be able to show that there’s a likelihood that there’s information there that might be relevant to the case. You’re going to get into this cost balancing, cost analysis. And it’s one thing to say well we think there’s data there, and we think there’s something – we’ve done this a lot for our clients. Help them go to courts and work out a process where somebody says look, we’re not getting any information. We think there’s stuff on these backup tapes. But of course, nobody knows what’s on a backup tape until you restore it. You may know general time frames and things like that, but you don’t know whether there’s anything relevant there. And the best way to handle that is to work out a process where you do limited restorations. You do like, you choose a couple of short periods or a couple of tapes or whatever. Restore those, do your searches, find out how much information in those sources are actually valuable to the case. And then at the point you can say look, we spent 10 thousand dollars and we found two emails, your honor, it’s ridiculous. The cost-benefit is like 5 thousand dollars an email right. It allows you to make those kinds of arguments. So it’s very hard to make it in a vacuum, especially when the opposing side comes and says look, the only place this important information exists is on those backup tapes that don’t exist anywhere else, you have to get it there. It’s a very compelling argument and the way to address that is to say okay, let’s test it. Let’s create a little model test; let’s come up with a process and a procedure. We can figure out what stuff is going to be valuable going forward.
So the next step is leveraging early analytics. Lisa, these are – steps six and seven are really your territory because you focus on all this stuff. So I will shut up for a minute and let you talk.
Lisa: So the fun part of this will be our power went out, and so did my internet. So I’m going to go blind on this. But from memory here, I know that one of the slides is basically using analytics to reduce the data set, and then the next one is using analytics during review. So assuming that I’m still correct, I think–.
Brian: The first one is leveraging the data analytics and process. This is the idea of okay you’ve collected all the data, and now you’ve got this – you’ve still got a large amount of data. Before you just throw it all up on a review tool, or throw it all up or put it all through processing, even after a basic search, there are other ways that you can – before, between your finish processing, and before you get all the review, other than just search terms right. And that’s the first step.
Lisa: Exactly. And the biggest issue with that is because search terms are where everyone is basically comfortable. Most of us are comfortable with search terms at this point, thank Google for that, I guess. But most of the time, parties will go back to the drawing board and say okay, we ran search terms, and it’s a million documents. We don’t want to review a million documents. Let’s just start cutting down the search terms. That’s really not only inefficient to go that route. You have no idea what you’re cutting out when you’re cutting out those search terms. So back to the analogy that Brian used earlier where he said that he wouldn’t send somebody in to just photocopy everything on everyone’s desk. Don’t just toss out everything because it was on Joe’s desk, and you no longer thing Joe has any relevant information. You’ve already collected all the data, you’ve already done the hard work, so you might as well use analytics to at least try and understand what it is that you have. You can use analytics to, again, so that you don’t have to just get rid of your keywords. Use analytics to narrow down what search terms are actually hitting on which domain, which custodians have the bulk of which search terms. Once you’ve got all of that information into analytics, you can really pare down and do a lot of reporting that is simply too complex for our brains to handle that many different connections. But it always makes me just cringe when I have a client that says well we’re going to tell the other side we don’t want any documents that have anything to do with their kids’ school. So you just run the search term school and kick everything out, which to me, makes no sense. If I can imagine an email that says you know, I need to pick up my kids from school, and then I need to log into my Seychelles account to see if all the money that I’m laundering has gone through. If I can imagine that email, it’s because I’ve probably seen an email similar to that in the past. People talk in email very loosely. And there’s no reason to think that somebody would say well this email is about my kids in school so I definitely won’t talk about any naughty stuff that’s not. So you know, use the analytics to pare down what it is that you have and get an understanding of it.
Brian: Just to be clear, Lisa, you’re not talking about – when you say analytics, you’re not necessarily talking about TAR? You may leverage some of the same kind of things, but analytics can cover anything from concept filtering to clustering to the timeline filtering, all that different kind of stuff. There’s a never-ending amount of tools that once you’ve run your search terms, you can then start digging in and say oh well, by using these and kind of grouping things and looking at it from different ways. Like an attorney I used to work with said, there are a million ways to slice up the data picture. And depending on how you look at it, you’re going to be able to identify oh well, here’s a group of stuff that we knew is never going to be relevant, so we can get rid of it right away. That’s really what you’re talking about, is trying to identify those clearly irrelevant things without throwing the baby out with the bathwater.
Lisa: Exactly. There are two types of analytics that we use for eDiscovery. There’s structured analytics and then conceptual. Structural analytics is, for anybody who uses Power BI and Excel and all those, it’s essentially the same thing. It’s taking all the metadata and just dicing it up a million different ways to give you more insight into what you’re looking at. Because frankly, looking at all the metadata fields that get extracted from a million documents, you can’t see those connections and stuff. Analytics helps you do that. Like I said, it’s very similar if anybody uses Power BI, it’s very similar to using Power BI on your data. And you can get a lot of insight into it. And that’s what if you give me a report that says okay so, for all these search terms, here’s where school actually appeared, and I look at it, and I go okay yeah, then I’m totally fine with kicking out every email that has that term because now I have a picture as opposed to just the word school as it appears to go ahead and pull those out. To me, it doesn’t make any sense.
Brian: Yeah. I mean, and one of the earliest examples I remember seeing of this was the case, this is probably 15 years ago before a lot of the great analytics came up, was we were doing a huge project, it was millions, I think 9 million documents to start off with for a very large corporation. And it was hundreds and hundreds of custodians. And one of the things we – and this is obviously just a very simple and I think very kind of easy way to understand this – is one of the search terms was coming out with a lot of hits. And after we kind of dug into it and really figured out the question, is you can’t just take those search terms on its face value of number of hits. Dig in and figure out why. And when we dug in and figured out the why, we realized that the daily lunch menus that every employee was getting from the corporation cafeteria was hitting on search terms and adding hundreds of thousands of documents. That’s a crazy simple example, but this is what hiring somebody who can really dig into the data and use not just technology-assisted review, but before you even get there, use these even simpler methods that are a lot cheaper and a lot faster, to get rid of these big chunks of data.
Lisa: Right. Like there’s another, like a lot of companies will want to use certain emails to help narrow down the set as far as to segregate the potentially privileged documents and stuff. And they’ll end up running a name that will hit every single document because they’ve collected their general counsel or something and it hits every document because every document has the general counsel’s name on it for their collection. And that’s where as long as you’re dealing with a vendor that understands how to do the analytics and the indexes and stuff, you can have your vendor go in and just remove that portion of the name from the index of certain fields so that when you run those terms, you’re actually getting terms in the body of the email as opposed to just the periphery. You know the other metadata fields that you’ve collected.
Brian: And one of the things I kind of compare this to, and as mentioned on the slide here, is think of this how companies – and we had this experience – use search engine optimization experts or SEO experts. For those of you who don’t know, an SEO expert is somebody who you bring in to help you develop your website in such a way so that it gets recognized by Google and it raises you up to the top so when somebody searches for your industry or your products or whatever; you come up to the top of the listing. And so for years, and this is from personal experience, for years we never thought about doing that at BIA because we put out loads of content, and we do lots of articles, and blog posts, and webinars, and all of this stuff. And we always thought well, we’ve got all this content. It’s gotta be coming up on search rankings and so on and so forth. When we hired the SEO expert, the first thing he did is come in and say here’s the top 10 things that you’re doing wrong, and a lot of them were simple stuff. And the same thing here, the analogy persists here. We may all, like you said at the beginning, be much more comfortable with search because of Google. And a lot of people on the phone are probably attorneys who were trained in law school a lot of how to search Westlaw and Lexus and all that kind of stuff as well. And so, but even so, there’s a big difference between doing this kind of searching and just being familiar with a search. And so it’s that thing, it’s important that the person that you bring in somebody who really understands all these different things because search is not just about search. It’s about how you deal with those results, and how you can dig in behind them. It’s like expertise. You can spend your life on Google and Westlaw and Lexus, but it’s not the same as doing really effective searches here, right.
Lisa: Yeah. It’s definitely very different. The big difference is when you’re searching like Westlaw and Lexus and stuff, you are searching for, and typically you’re searching for case law. Which is typically written by the same group of people. It’s written by your judges and their law clerks who had the exact same training, they’ve had law school training, etc. In eDiscovery, we’re searching people’s emails. And not only is it less formal, everything from the janitor all the way to the CEO, they speak differently. They use different words. They just have different manners of speaking. And if you don’t understand the different manners of speaking for a company, you won’t be able to find the documents that you want.
Brian: And we’re running a little bit short on time so why don’t we jump over to the next – unless there’s anything else on that topic, I think we—
Lisa: No. Yeah, go ahead. I can go through it. As far as for document review, if you’re not using analytics for document review, it’s definitely not going to be helpful. You’re definitely going to be wasting money. Analytics is cheap enough now that it almost doesn’t make sense not to use it for every review. I typically will use it for a review all the way down to a few thousand documents just because it’s helpful. You can do email threading, which is the most basic, even if we’re not talking about the conceptual analytics. Just email threading will allow you to review the most inclusive email, the email that has all the pieces of the rest of the thread. Most of the tools now will actually let you visualize the email thread so that you can literally look at the entire thread and you can see the dates, and who’s on them, and you can actually dissect into it and say okay, the thread up to this point is responsive, and from this point backward it was non-responsive, or the other way around. You can look at a thread and say okay, this thread was not privileged up until this point, but then they brought in outside counsel and now it’s privileged from this point forward. You can actually pinpoint those areas and tell the machine go ahead and code everything about this as privileged; everything below is not privileged. And you can wipe out an entire thread with one little sort of coding as opposed to reviewing 10, 15 different emails.
And another big way of using it is for QC. If review managers are still using just random sampling to do their QC, they’re wasting time, they’re wasting money, and there’s really no good benefit to that other than you might catch some mistakes. The way that we always handle QC is we run analytics, and then we pull up documents, and we say, here’s this document and here all the near dupes, or here are all the textual dupes, and here’s the email thread. And they should have pretty consistent coding. If they don’t, that’s the ones that we pinpoint for QC. Basically, we throw up a lot of what we call overturn reports, meaning something where the machine thought it should be coded one way, and we thought it should be coded the other way. Those are the things that we want to narrow down our time for QC as opposed to just kind of wasting time on random documents.
Brian: And in fact, Lisa, I think it’s safe to say that even for some of our kind of stubborn clients who – I hate to use that word – but there are stubborn clients out there who still want to do serial reviews for whatever reason. In some pieces, it may be needed because TAR doesn’t work great on certain files, certain number intensive documents, or CAD intensive documents, so you still have to do serial review sometimes. But it’s safe to say that even if we’re not using TAR for the actual review process itself, we are using it for every single QA process, and for all sorts of privilege and other stuff like that. And I think that’s part of the thing here is it’s been drilled into everyone’s head at this point that TAR can save you money. I think everybody thinks it’s on the big cases, but as you say, it’s gotten so commoditized and so straight forward to use that even on the small cases, it pays dividends and savings. But I think that one of the biggest points from what you just said is it’s not just the reduction in the amount of review costs, but it’s making it more efficient and more accurate by using it in these other things. Not only will you benefit from that, but you’ll also see reduced costs associated with that as well. So TAR is not just for review anymore, it’s for QA, it’s for all this other stuff.
Lisa: And you know, quite often one of the nice things about using the QC is if we’re able to just catch even one document that should have been tagged privileged and wasn’t tagged privileged before it gets produced, we are presenting an entire clawback process that will then have to happen later. And that’s a costly process. So anything that we can do to catch something before it gets out the door incorrectly will always save money because the analytics and the QC tiny price compared to the clawback process afterward. Which even if there’s no arguing on either side, even if it’s very clear and the other side’s happy to give the document back, it’s still tech time to go in and do that whole process in both databases and everything. So it will always save money to use those analytics during the review for the QC and stuff before you start producing documents.
Brian: Yeah. And I mean one of the things that I always use is back when I was a young associate and dealing with paper, in the 90s was still very much paper world when I was in litigation, one of the first things I would do when I came across a document that was redacted or even getting a privilege log and seeing descriptions of documents as we would go in. And we’d oftentimes find the exact same document or a different version of it that was either unredacted or produced because people just wouldn’t catch those similar documents across. And when you talk about privilege, you can almost entirely prevent that through the use of analytics and near dupes and things like that.
Barry: You can’t unring the bell.
Lisa: And the other part of that, just the redactions, I can’t even think about how many cases we’ve had in the past where clients have had to come back and re-redact or change redactions. And analytics is one of the things where if you run the analytics you can see – are all of our redactions consistent? If not, let’s fix them first before one side has to argue about why the redactions were different.
Brian: Yeah. And I think I do love the thing you can’t unring the bell. Sure, you can get the document back, but counsel, if they’ve seen it, you can’t absent some borg technology. You can’t take it out of their mind. The last thing, and since you can’t see it, Lisa, I’ll just run through it was a document or a little case study we did on a small document. And everybody can read this for themselves, and you can all get copies of the slide deck. But basically, it gives you an example of how reviewing 115 thousand documents. One attorney in six weeks was able to do it at a cost that was 127 thousand dollars less than a serial review would have cost. And I think when you look at that cost savings differential, and again this is not just using TAR alone but using all these methods together. You can go through – I’ve always, I’ve gotta say, Lisa’s probably the best at this I’ve ever worked with. I’ve always been amazed at how what do you have 20 years’ experience in document review and analytics and all this stuff? I mean, you can’t beat that.
Lisa: I started when I was five. No.
Brian: Oh yeah, I don’t want to add too many years there. But you know, just because you have that experience, you have this ability go in. And I kind of make the comparison to the guy in the movie ‘A Beautiful Mind’, where the guy saw numbers and formulas because the things started floating around. And that’s kind of where you get, right. You start seeing these patterns and recognizing these things and ways to cut data, just because you have that level of experience. And again, I think that’s just as important to talk about when we’re talking about all this TAR and analytics and everything else. All of those tools are only as good as the people who are using them. And so it’s really important. And I’m not saying you all have to go out and hire Lisa right away, or us necessarily. But you gotta vet the person who’s doing this and who’s running those tools, and they really need to know what they’re doing because as I like to say, those tools simply magnify either the quality or the stupidity, or bad decisions or whatever you want to say. I can’t think of the right word there. But the technology just amplifies that individual input. So if you don’t have somebody that really knows how to use the tools, as well as knows the subject matter that you’re dealing with on that particular case, then you’re going to end up with a really crazy result. And so I wanna just make it clear, it’s not just about the tools. You gotta have the right people running it as well. And I’m sure Lisa, you’ve probably seen examples of it going very badly when the tools are used by people who don’t have the training.
Lisa: Yeah. And I mean there was one where we were brought in to advise during their hearings, and one side had run a TAR process, but they didn’t understand – not only did they not understand the TAR process completely, they didn’t understand the reports themselves that they were reading from which led to our side basically winning very handily because all we had to do was ask a few pointed questions and they couldn’t answer them. And the answers they did have were clearly wrong, and it really did not take much for us to win that case simply because they didn’t understand the reports. The tools are complex; they really are. And that’s why you should use somebody who is in the tools 40, 50, 60 hours a week instead of somebody who just got them. They just got their training a couple of weeks ago. The tools are easy to use. They are designed to be easy to use. That doesn’t mean that you are fully able – you are not able to fully use all of the pieces of the tools just because you have a quick training in it. They’re very complex.
Brian: Yeah, absolutely. It’s not like you would go to somebody who’s never used Westlaw or Lexus and throw him on there and ask him to research your next Supreme Court brief.
Lisa: I mean, for an example, last week, my kid was in high school, and now he’s home. I had him inventory our kitchen and put it into a spreadsheet in Excel to teach him how to use Excel. He used Excel last week, and he used a spreadsheet. He’s not an Excel expert, and I would certainly not think that he could do a whole heck of a lot in Excel. But he has used Excel, but I’ve been using Excel for 20 years. I know a lot more.
Brian: I gotta say I love it. I love the way that you figured out how to like not only to teach your kids at home during this but come up with a useful way to get your kitchen cleaned and organized at the same time. That’s brilliant.
So I know we’ve got a couple of minutes left. Mark, we’ve answered a lot of questions as we went along, which is great. But do we have any other questions are out there that we can answer in the last few minutes?
Mark: We have a couple more, Brian. We’re at 2 o’clock right now. So there’s one I’ll put out there, even though we’re going to run a little bit over, and I apologize for that to everybody. Time is valuable.
The last question I think is relevant is if you’re a small organization with an eDiscovery steering committee, do you need to create one? And who should be on it? If you can answer that in a minute or less, that would be awesome. And then you can wrap it up.
Brian: Sure. I mean, just like any other kind of business process out there, if you’re a small office, you’re not going to go buy a 10 thousand dollar copier. But if you’re a huge office, you might go buy a 10 thousand dollar copier. It’s the same thing you have to think about here. There’s a balance to everything in this. And so if you’re dealing with a smaller company that has limited resources, and you have a limited amount of exposure to eDiscovery generally speaking, then you can probably go to a vendor and get some standard draft forms – like we have some standard stock forms for all of this stuff – and hire a single outside consultant for a small organization to help you put together that plan. And then you revisit it once a year. And that’s a really inexpensive way to kind of get a handle on it. Up to we have some organizations that literally have dozens and dozens and dozens of litigations going on at any given moment in time. And for those groups obviously having a standing committee that meets on a regular basis and reviews their process, has that protocol, and you’re constantly looking at it, refining it, just like you would any other business process. So it’s just a matter of scale. And so if you’ve got very little litigation or you’re a very small firm without a lot of internal resources, then bring in that one extra resource, focus on some of the bigger pieces of that without getting into too much granular detail, and you can kind of develop it as you go along. And just like anything else. The first case you’re going to put that test, you’re going to refine it. The next case you have, you’re going to refine it further, and the next case. And by the second, third, fourth case, especially by the second or third case is where you really start to see those refinements and that kind of efficiency come into practice. So you gotta look at it. And of course, it’s like anything else. You’re not going to go spend a lot of money to get a general electric or a Walmart level protocol and process and team if you’re a small regional or a much smaller company than that. So that’s what I would recommend.
Mark: Great. On that note, I’d like to thank the presenters, thank ACEDS, and of course, thank everybody who attended today’s webinar. Of course, without you attending these, then there’s no use to do it. At BIA, we love education. So thank you again for attending. To wrap it up, I think to say that people, processes, and technology is the secret to success for recession ready, recession proof discovery. If you can do those things and do them well, you’re going to end up better, faster, and cheaper. And isn’t that really what this is all about? Recession, no recession, good economy, bad economy. So again, thank you all for attending, and have a great day.