Ricoh eDiscovery

Five Easy Steps to Get Started Using Analytics

Posted by Jennifer Johnson |4 minute read

Mar 12, 2015 4:30:00 PM

Needle-in-Haystack_smAnalytics tools have been around for a long time. You've probably even used these tools whether you know it or not...

  • Have you grouped similar documents together (near duplication)? Then you have used analytics.

  • Have you reviewed emails by thread rather than chronologically? Then you have used analytics.

The critical barrier to expanding the adoption of analytics is changing the way we work.

The reality is that when we use these tools in a more comprehensive way, we transform our workflows. We are no longer simply opening a database, sorting chronologically by family and clicking through document by document. Now, we must pause to assess the case and look at it in a more holistic manner. It is undeniable that analytics helps us to recognize key concepts and patterns that a human being could never have the time or mental capacity to assess. From this perspective, analytics mimics, and likely exceeds, the combined insights of a whole team of experts.

But, who has time to change the way they work on an important case? 

The reality is that we don't have a choice.The real question to ask is at what point will I not be able to do my job effectively without it? We can't continue to work in the way that we do with growing data and downward pressure on time and cost.


So, how can you feel more savvy on your next case? And more importantly, how can you save time, effort while delivering additional value to your team and your client? Following are some accessible entry points that will help you to continue stretching your competencies and ultimately master analytics technology on your cases:

Step 1: Strategic Sampling

Before you dive in to your database, do some preliminary analysis and perform strategic sampling to valid your assumptions across subcollections of documents.

Step 2: Look at Key Concepts

Spend some time looking at key concepts. This may shed light on the pertinent issues in your case and help you to assign key sets of documents to specific reviewers.

Step 3: Take Advantage of Transparent Searches

Rather than treating search terms like darts in a darkroom, take advantage of the “transparency” of searching capabilities that allow you to actually “see” what words (and permutations thereof) exist in your database. Validate your assumptions. Expand or refine your scope of keyword terms. Search with confidence.

Step 4: Prioritize and Organize your Documents

Instead of simply breaking up the collection into random batches, use what you have learned from the previous steps to prioritize and organize your documents. Group similar documents, set potentially irrelevant documents aside for the time being and quarantine potentially privileged documents for special treatment. Most importantly, get your team started on priority documents first so that you can find that "needle in the haystack" and be better prepared for the next steps on your case.

Step 5: Quality Control your Production

Consider using analytics to quality control your production before it goes out the door. This is particularly useful on large, complex cases with short timelines. Validate the work of reviewers, cross-check that nothing has been missed and confirm that nothing sensitive is going out the door.  And to make review of the other side’s documents more effective, use analytics to get a quick summary of the collection or compare it against your own to leverage all of your issue coding and categorization to date.

Simple enough, right?

So, why are we not all jumping on the technology band-wagon like we have with Google, Netflix and iTunes? Perhaps the biggest hurdle is convincing the subject matter experts on your case that analytics simply reinforces their critical contribution to the case. In fact, once they have tried it, they will wonder why they are limiting themselves to increasingly arduous and menial tasks and not harnessing their tacit knowledge to drive immediate return and results.

Topics: Jennifer Johnson, eDiscovery Solutions

   

Tell Us What You Think