Moving to the Cloud for eDiscovery

By Christopher Hunt posted 09-13-2018 12:26


The cloud is often referred to as the great equalizer when it comes to technology. Organizations that can learn to leverage the flexibility and efficiencies of cloud-based platforms will quickly realize numerous benefits, from cost control to unlocking capabilities never-before available (and in some cases, may never have been available at all). Smaller firms and companies in particular find the appeal of the cloud tantalizing – gain modern technologies and efficiencies, literally jump years ahead in your process, with very little up-front investment in hardware and software. This is the position my firm found itself in a couple of years ago as we embarked on a project to replace and enhance our litigation support and document review tools.


Our firm is a small litigation boutique. We do not have a dedicated litigation support group, and with limited resources available we would often lean on third-party providers to handle discovery processing. As the industry evolved toward digital discovery, the old tools built around paper became a detriment, and we recognized it was time for a drastic change in how we handled litigation support. We took advantage of a confluence of events – the end of life of our on-premise review platform, our lack of in-house processing capabilities, and growing user frustration with the limitations of what was available to us – and embarked on a journey not just to replace our discovery review tools, but to really look at the entire litigation support process and improve it from the ground up.


Replacing a review platform and reshaping the process associated with litigation support and eDiscovery offers an exciting opportunity, and a herculean task in trying to find the right solution for your users. Even though our user population was screaming for change, we wanted to be very methodical about it and ensure that we were making the right changes to benefit our users. This is a balancing act: if it’s too different, or too hard to use, you run the risk of losing adoption; if it’s too similar to your current environment, have you done enough to justify the change?


Since this change would directly impact our end-users, we understood immediately that the user experience would be a key element of the selection process. We formed an internal task force comprised of a number of different stakeholders – attorneys, paralegals, and IT – and began the process of laying out the goals of our project. Before we even started talking about products and features, we defined our major pain points and what action would remedy or remediate those issues. This allowed us to really focus on what we were looking for in a new platform, and what we expected to be able to do once we made a selection and transitioned. From there, we began the process of mapping out the features of our existing system that were liked and should be available in a new tool, along with talking about the wish list of features that would be must-haves moving forward, and some things that would be nice but that we could live without, at least initially.


With a requirements list completed, we began identifying potential platforms. Initially we discussed and reviewed the typical names in this space, and spoke with other firms to get their opinion and feedback, as well as solicit opinions from ILTA and other professional groups. As we progressed through the process of bringing in vendors for discussions and demos, we started to talk about non-traditional providers, particularly cloud-based providers, to see if there was a different path for us to move forward.


At the conclusion of our initial round of discussions, one thing became very clear to us. Transitioning from one on-premise system to another was likely to be a very costly investment, between infrastructure and licensing requirements and the purchase, installation, and setup of the new system, along with the resources required to migrate existing data to the new platform. We also began to realize that many of the solutions we were considering weren’t as complete as we would prefer, leaving some gaps to be filled by additional software or third-party support.


And so we pivoted to the cloud. What jumped out immediately to us with the cloud providers was the broad toolset offered by many, and the dedication to continuous improvement and development in their offerings. This is a pretty important advantage of the cloud model – the entire user base benefits from enhancements and improvements almost immediately, and with very little, usually no, involvement from the end user - no technical support, testing, piloting. The other major advantage, particularly to smaller organizations, is the reduced implementation investment. Typically the only cost is a monthly subscription charge per user or per GB – no annual maintenance fees and no additional costs for different levels of access or for additional tools – which allows the firm to budget much more reliably with known quantities.


In addition to reviewing cloud providers for the functionality we would expect from an on-premise tool, we were looking for value adds from each provider – what exactly was included in the monthly fee. This varied quite a bit depending on the provider, the model being offered, and the complexity of the tools. Some questions we asked through this process:

            Is user access included in the monthly fee, or is there an additional license and access fee?

            Is any storage included in the monthly fee?

            What types of data processing are included, and what is not included?

            Is support included in the fee, and what level of support?

            What training is included, both initial onboarding and long-term?

            Is data import and processing a self-service model, a vendor-supported model, or a hybrid of the two, and are there any additional costs for import and processing if done by the vendor?

            What analytics are included?

How is data shared with co-counsel, experts, and, in the case of productions, opposing counsel?

            Are there any costs for preparing data for production and exporting data?

            What does the end-of-matter process look like – how is data archived when not in use?


Being a small firm, we were excited to see a number of offerings that provided all-in-one solutions, from data collection through production. Some even offered transcript management and trial preparation tools as well, which meant potentially a reduction in the number of third-party vendors that would need to be involved in our lit support process.


The cloud model forced us to think much more strategically about the data that should live in our new tool. With an on-premise tool, we could throw as much storage as we wanted at it, and it didn’t impact the bottom line in the grand scheme of things. A more thoughtful approach is required with the  cloud model, as most platforms utilize a per GB per month subscription fee, and the firm must decide whether that is a cost that is passed on to clients or absorbed as firm overhead, or some combination. While our attorneys and paralegals were already mindful when it came to structuring search terms for discovery, a much greater attention to the data being delivered is required in order to avoid processing and paying for data that is irrelevant to the matter at hand.


Another important consideration of the cloud model is to what extent your clients approve of having their data stored in the cloud. In our case this wasn’t an issue as we have already moved systems to the cloud and jumped that hurdle, but if this is your first foray into the cloud, then one of the first things you’ll need to do is review client guidelines to see if any clients have restrictions in place, and determine from there how that may impact your path to the cloud.


With all of these considerations in mind, we ran through a new round of product demonstrations with our key stakeholders and solicited their feedback after each. By the end of this process we narrowed the field to two options, and eventually selected the platform that we are using today. We opted not to do a proof of concept to aid our selection, but should we repeat this process in the future that is one step I would recommend. Have the providers spin up a matter with your actual data and let the stakeholders interact with the program and perform real-world functions to see how each proposed solution performs. Demonstrations are typically very polished and can gloss over the finer details.


With a new platform selected, we began the process of training our users and identifying what data, if any, was to be moved from our legacy system to the new platform. Aside from the system selection process, training may be the most important part of a transition to any new tool, and in particular one that lives in the cloud as there are unique challenges to a cloud-based system. We required on-site training prior to our official go-live on the platform, broken down into admin training, training for those users who would be processing data, and a more simplified training for reviewers. Users worked with live data with their own credentials during each training phase to get a feel for how the new platform functioned, and this gave us all a chance to refine the workflows needed for working with data in the cloud.


One of the biggest challenges in this transition was how to move data into the platform. Our provider initially had two options for uploading, a web interface upload that was limited to a relatively small data size, or a direct upload to AWS using third-party tools, which required additional training to learn how to setup the tool and access the storage locations for each matter. Large data sets required time to upload before they could even be processed, a new wrinkle to keep in mind if you have time-sensitive data to handle. Our provider has since built out a more robust file transfer facility within their interface, and now almost all functions can be completed through the browser, with an eye toward completely removing the need for third-party tools.


Once we moved through training and setting up the new environment, the benefits of our choice were immediately felt. Self-service import and processing of native data required nothing more than a few mouse clicks, and within minutes we had review sets available for attorneys. Gone are the days of needing to farm data out to a third party for processing! We also gained the flexibility of being able to access the platform from anywhere, through any browser. No remote access back to the office, or clunky exports or other methods to be able to access data on the go. In addition to our own flexibility, we gained the ability to easily add additional parties to our databases, so co-counsel or experts could review data specific to them without having to do an export of any kind. Productions are processed and a link shared with opposing counsel for download, removing the need to put data into an FTP site or a flash drive for transfer. In short, the impacts to our workflows have allowed us to work much more efficiently.


Has everything been perfect? Of course not. We knew going into this process that tweaks would be needed on both sides. When we signed our agreement, it wasn’t just a license agreement, but more of a partnership. The cloud model is one where continuous improvement is built into the development cycle, and we have taken full advantage of that with our provider. We opened a feedback loop with regular meetings to discuss areas we feel need improvement, hear from our providers team about what they are working on, discuss limitations and see if any potential roadblocks can be shifted. The key to this is the willingness of both sides to provide feedback and to act on it, to be active participants in the process, and sometimes to have hard conversations when things aren’t working the way you think they should. Having a partner who is open to this insight and willing to make changes to improve the product has been tremendous.


Looking back at our journey, moving our eDiscovery/document review platform to the cloud has unlocked the potential of scaling up our operations without costly overhead or increased resource requirements, allowing us to provide cost efficient, high-quality service to our clients.


Chris Hunt is the Director of Technology & Operations at the Boston law firm Sugarman, Rogers, Barshak & Cohen, P.C. With over 20 years of experience in small firms, he leads the strategic vision and practical implementation of process and technology. Chris is also a co-chair of ILTACON 2018 and a member of ILTA’s Program Planning Council.

Also Read Aaron Crockett's Blog and John Sawyer's Blog