HaystackID® Acquires Business Intelligence Associates, Inc.Read More

Advanced Analytics Value for Small Document Review Cases

Advanced Analytics Value for Small Document Review Cases

Can I Use Advanced Analytics For a Small Case?

There is a common misconception that advanced analytics value is derived only on large document sets and that not all cases are big enough to truly benefit from them. So many times, I’ve been on calls with outside or in-house counsel and heard the questions: “Will advanced analytics be useful on such a small set?” or “Is this really enough documents to make advanced analytics worth it?”

In short, the answer is YES!

I understand and can appreciate the misconception; TAR, in its earlier days, used to require a good amount of documents in order to feed and seed them. If we’d had this conversation three years ago and you didn’t have that initial volume to start, then the answer may well have been that we couldn’t make much use out of advanced analytics on your small document set. However, to borrow (clumsily) from Bob Dylan:

The Tools They Are A-Changin’

TAR 1.0 takes a set of documents reviewed by attorneys and then uses those decisions to make a prediction on the remaining docs—end of story. Since 2017, Brainspace and Relativity have introduced Active Learning and CMML (Continuous Multimodal Learning), also known as TAR 2.0 or supervised analytics, which takes the docs we feed and continuously reviews and continuously scores. Every scoring decision is applied to future documents, so the predictions get smarter and more accurate as the review moves forward.

Today we have advanced analytics capabilities that extend beyond TAR, and it is these tools that can result in the biggest savings for your smaller cases. Most advanced analytics tools come in a bundled deal, meaning that if you use one tool in the platform, you have access to all of the tools. Email Threading, Near-Duplicate Identification, and Textual Duplicate Identification are three of the primary tools you’ve probably already encountered.

Here are some lesser-known advanced analytics tools that can come in handy as well:

  • Name Normalization will gather all variations of someone’s email and put their name in a field. For example, I have my work email, a google email, a yahoo email, and probably some others. The analytics would identify all those emails as mine and put my name in a field. This is especially helpful for privileged logs.
  • Language Identification is just that: a tool that identifies the primary and secondary languages used in your data set. This is great to run at the start of a case to figure out staffing needs, like whether you’ll need to hire Spanish, German, or Japanese reviewers for a foreign language document review.
  • PII Identification will flag documents for social security account numbers or anything else that may be confidential so you can redact that information.

Since these tools come in a bundled deal, some clients may think: I only need one tool. If I’m not going to use all the tools, is it worth having the whole bundle?

The short answer is YES. But to address this question with another question, let’s return to the original discussion topic: Does this case have enough documents to use analytics?

Some Easy Math…

Let’s look at a case of 1,000 documents. Let us also assume that this is a sole practitioner, so only one person is reviewing the documents. I will use some industry averages to show the advanced analytics value.

  1. The average attorney will review 50 documents per hour.
  2. The average attorney will charge $200 per hour.
  3. The average document reduction from email threading, if looking only at the unique threads, is 20%.
  4. The average increase of review speed, from viewing threads together, is at least 10%.

Let’s say you have a vendor who charges 4 cents per doc for advanced analytics.

Cost of Review Without Email Threading:

  • 1,000 documents reviewed at 50 documents per hour is 20 hours.
  • 20 hours at $200 per hour is $4,000.
  • Total cost of review: $4,000.

Cost of Review with Email Threading:

  • 800 documents reviewed at 55 documents per hour is 14.5 hours.
  • 14.5 hours at $200 per hour is $2,900.
  • Cost to run advanced analytics is 1,000 x $.04 which equals $40.
  • Cost of a tech to run advanced analytics $200 per hour x 0.5 hours is $100.
  • Total cost of review: $3,040.

Total Savings: $960

As you can see, advanced analytics value can have a tremendous impact, regardless of the size of your case. Keep in mind that the above scenario uses only one of the many tools available through the advanced analytics bundle. Our BIA team of lit tech experts working on your document sets might find, for instance, that we are able to mass tag a number of near or textual duplicates together. Using other tools in the bundle could easily increase the average review rate to 60 documents per hour, saving you or your client another 300 dollars.

Of course, the definition of “a small case” is different for everyone. It could mean one thousand documents for one client or ten thousand for another. But those smaller cases are where you’ll start to see the biggest savings and the highest advanced analytics value when you choose a review company that will use advanced analytics on your document review sets. At BIA, our tools are top-shelf, and our people have the top levels of training and certifications needed to run these tools at their maximum capacity for speed, accuracy, and efficiency.

So the next time you find yourself wondering if you have enough documents, or if you can convince your client of the advanced analytics value for your case, I hope you’ll stop wondering and give BIA a call. Our answer is YES.

Aaron Boone, Esq.

Aaron Boone, Esq.

Aaron Boone is part of BIA’s Litigation Technology Expert team. With over ten years of experience working in litigation support, he has a vast knowledge of the various industry-standard analytical tools such as Brainspace, Relativity Analytics, NexLP, and Matter Analytics. Aaron holds a JD from Western Michigan University’s Thomas M. Cooley Law School and is certified as a Brainspace Administrator and Milyli Blackout Administrator.