In 2021, CDS launched a proprietary data visualization platform, CDS Vision, designed to help users harness the full power of cloud-based eDiscovery platform, RelativityOne. CDS Vision leverages robust visualizations along with CDS’ proprietary workflows, technology and expertise.
During our recent webinar Seeing is Believing: Why Visualization Matters in eDiscovery, Chris O’Connor, Cory Logan, Michael Milicevic, Esq. and Sue-Deelia Tang, of CDS discussed how, with minimal training, users of all technical proficiencies can easily navigate Vision’s intuitive dashboards to streamline the eDiscovery process.
Enjoy part 3 of our 3-part blog series on how CDS deploys data visualizations to empower clients to raise their eDiscovery game.
Mike Milicevic:
CDS Vision is a proprietary set of dashboards that CDS has created in RelativityOne leveraging Relativity visualization combined with CDS proprietary workflows, filters, and technology, to really improve the way our clients interact with, understand, and act upon their data during the various phases of eDiscovery.
Right now, CDS Vision is composed of four main dashboards that roughly follow the EDRM, those dashboards are called First Look, Data Intelligence, Communications Analysis, and the Scoreboard. And at a high level, CDS Vision allows our clients to quickly organize, classify, and prioritize relevant document sets, leverage Relativity analytics to visualize key concepts and communications networks, and monitor review progress and attorney work product in real time.
Each CDS Vision deployment is fully customized to each of our client’s needs. So, we also offer the ability for our clients to select CDS visualizations that they would deploy on an à la carte basis, or we can work to create a completely new visualization, widget or dashboard based on their unique data sets and requirements. Our hope is that CDS Vision really unlocks the power of visualizations in RelativityOne while also making everyone’s job a little easier.
Chris O’Connor:
How do clients take advantage of this stuff if they have a new matter they want to bring to CDS and they’re ready to go? Do we just turn the button on? What happens?
Mike Milicevic:
So much of this can be templated for clients so that when they login they have the visualizations that they’re used to, or use most frequently. Typically, it starts with a design meeting which can be a pretty quick process. And that’s to say, okay, one of the things that you’re really interested in focusing on for your case is what’s your coding criteria? We look to build out the fields and create what we call the scoreboard to track coding progress. So again, it starts usually with that initial design meeting, and then if we agree upon a set of visualizations that make the most sense, even across all of their matters, we can pretty easily deploy them as new matters are spun up on RelOne.
Chris O’Connor:
It could be done on a single matter basis though, correct?
Mike Milicevic:
Of course, yes.
Chris O’Connor:
What about customization? If a firm has a baked-in idea of what they’d like to see if they’re using visualization, can they bring those to the table?
Mike Milicevic:
Yes. Those are welcome. A lot of what we’ve done in terms of our proprietary visualizations has been informed by our clients. We’re looking to provide them with solutions. So any feedback that we receive, usually gets incorporated into a menu of visualizations that we offer.
Boost productivity and accuracy from Day 1 by automating tasks and streamlining workflows
Chris O’Connor:
So Sue, combining the power of visualization with traditional investigative review processes is important, but what are the primary benefits that CDS Vision addresses here in the version one release, which was on March 1st.
Sue Tang:
There’s a ton of benefits with these dashboards. Mike has been so great with the development team, working this out for CDS. Regardless of whether you’re conducting an internal investigation, you’re doing early case assessment, or preparing for depositions, there’s really no time to waste a lot of time. You’re in a high-stakes race to find the information. One of the keys to an efficient and successful fact-finding mission is selecting the right data analytics tools that will help reduce the noise and boost the ability to find what you’re looking for. And what CDS Vision does is it takes all of that analytics information and puts it into these dashboards so that you can focus on getting real work done. Right from the beginning.
As soon as we start getting data loaded into the system, you have these visualization charts. They’ll reveal what you need to focus on so you can start working immediately. You can build out your case and address errors from the very beginning. And then as the review is going on, you’ll want to see the review progress that’s being made by your team so that you can be proactive, provide additional training if needed or perhaps modify your team.
Chris O’Connor:
Okay. So, for example, for Personally Identifiable Information (PII), can we automate the way it’s handled? In other words, after data has been published, can I get a PII flag to automatically attach?
Mike Milecevic:
So yes. I would say right now it’s partially automated. As soon as data is ingested, we have a PII widget, it’s a wheel with different categories of PII, the things that you would expect like social security numbers, credit card numbers, date of birth. But we also have categories of PII from other countries as well, UK telephone numbers, Swiss telephone numbers, that sort of thing. So you can click through and select the categories of PII hits that we have. And then you can automate redaction through Relativity Redact now, in RelOne, which is pretty neat. So you’re having the system draw a lot of those initial PII redactions for you, and then you can have a review team come back and QC it, which tends to save a ton of time.
Early on, I was a document review attorney and I happened to be the team lead for redaction projects. So my job was to draw boxes all day, and that was very tedious and time consuming, and it’s very easy to make mistakes doing that sort of thing. And so I think by leveraging the PII visualization that we have along with automated redaction, you can take a lot of the time and effort out and reduce the number of documents you might miss.
What types of data and information are best suited to visualization?
Chris O’Connor:
So, here’s a question from the audience about structured data. Can we put trade data in here? I assume they mean stock trading to look at examples of SEC malfeasance or potential crimes. I could see that possibility. I know pulling structured data is a challenge.
Cory Logan:
Yeah. I think that’s the ever-evolving conversation. And I think that goes back to the point I was making earlier about the collection piece of it. Again, the visualizations and the cloud-based structure and framework of Relativity One allows you to ultimately put everything into it as you can based on privacy and security concerns. However, you just need to make sure it’s collected in the right way and converted to be able to process it properly and visualize it in the way that we’re depicting it here.
Sue Tang:
Visualization can really be built on anything that you can capture in a field. For example, if there are metadata fields that are getting attracted from your datasets, we can build a dashboard around it. If there are reporting metrics or any kind of metrics that are getting automatically put into fields, we can build dashboards around them.
Cory Logan:
And we’re not just talking about email data. We’re not just talking about share drive data. We’re talking about chat data, mobile data, all types of data regardless of where it’s being collected. As long as it’s collected in the right way and it’s converted properly in conjunction with RSMF it can also be visually appealing via a dashboard or a widget in RelativityOne via CDS Vision. So as data points continue to change, as the adoption of mobile and chat data especially since the pandemic continues to evolve, that’ll also be incorporated into this as well.
Mike Milecevic:
I’m personally very excited about the prospect of better utilizing visualizations. It’s not a new concept, but the advanced technology to power them and to do a lot with them is a new thing. That’s RelativityOne and being in the cloud and being able to scale up resources in the way that Cory spoke about earlier has really unlocked a lot of opportunity for us to get very creative and very tactical with the use of visualizations across eDiscovery matters. I think we’ll be seeing a lot more of this in the industry, and I’m excited with where we can take it.
Cory Logan:
Mike, I think you hit the nail on the head. Just going back to my statement at the beginning here about purposeful innovation, RelOne being built in Azure and being able to scale gives us these data points much, much quicker. RelativityOne is super powerful. If you rewind a few years when we were doing training for attorneys and we were talking about how to get to these data points, we weren’t talking about these types of dashboards and visualizations. We were talking about nested searches and that can be a lot. That’s cumbersome. And if you’re not very familiar with the tool, you’d probably throw up your hands and be like, “I have had enough.”
And now you don’t have to do that. Now you can templatize these dashboards and have them available every single time for every phase right after processing. Yes, there are some customizations that we can deploy, there is some button clicking that has to take place but for the foundational pieces of this, it’s ready to go, and there’s not a heavy lift here. So it should be very intriguing for folks that haven’t been exposed to dashboards to know that they don’t have to be experts in the tool to be able to take advantage of this.
Next up: Forecast the trajectory of your matter. See billable metrics in real time.
Chris O’Connor:
Cory, can we give our audience a little preview of what’s in development? I know we’re not going to promise a delivery date yet.
Cory Logan:
One thing we prioritize at CDS is to make sure we provide a ton of transparency to our clients. We make sure they have all the metrics from a billing and reporting standpoint. So, what we’re doing is taking our reporting portfolio, and we are putting that into CDS Vision via RelativityOne.
Corporate legal departments are often facing fiscal pressures and concerned about, “Where am I going to be in the next three months? Where am I going to be next year?” So understanding their data footprint, understanding user seats and hours are probably the three main buckets, but those are big drivers to budgetary decisions. And if they don’t have that information and the metrics readily available, it makes it very, very difficult to make crucial decisions.
Instead of just giving clients the ability in real-time to see where they’re at, we can also show them where they’re going. We want to be able to forecast and give them the ability to see, “Okay, in one, two or three months from now, if I continue on this trajectory, this is where I’m going to end up.”
Mike, you probably have a more accurate timeline than I do but I would say probably within the next six to nine months, we’ll be deploying these elements. And I think it’s very powerful, and whether you’re a corporate, law firm or a government entity, it’s going to give you the ability to see your billable metrics in real time and to forecast and make those educated decisions as you need to.
Chris O’Connor:
Right, we spoke about case matter dashboarding. The goal of cross-matter dashboarding would be to figure out the average cost of litigation, right? Whether a law firm or a corporation is doing it, they can take a look at their matters hosted in the products and be able to see what they’re paying for? Determining value. Like, $100,000 dollars in investigative costs for a $50,000 value matter is probably a bad spend. So, how do we make sure we don’t stray too far? That’s something we can also feed this system, right? Things like reviewer hours and cost?
Mike Milicevic:
For sure. I think one thing that’s worth mentioning is that all of these widgets, all of these visualizations within a workspace, let’s say they are on the same page, act in unison with each other. If you’re selecting one widget, the next one updates and the next one, and the next one. All the information is linked. It’s something that builds on itself, and you can do that across matters as well. That’s a little more complicated. We’re looking at incorporating aspects of that in later versions, but certainly looking at things like billing across the portfolio of matters, and even things like reusing some review coding that may be relevant across matters – those are all things that I think visualization can help with.
