Building Ediscovery Search that People Actually Use

Back to Blog Posts

#DISCOmagic is a blog series that takes you behind the scenes to show you some examples of all the hard work that goes into making this magic happen. This installment comes from Principal Product Manager Jordan Janes.

Our users run well over 10 million searches in DISCO every year, so it’s important that they are as effective as possible. We use the data from these searches to make our search performance even better and our searches easier. 

There are three themes hidden in plain sight when it comes to searching in DISCO: search is available where you need it, we always sweat the details, and good is never good enough when it comes to search speed. 

Available where you need it, how you want it

The document list, search bar, and search filters are immediately available when you open DISCO Ediscovery. There’s no need to leave the document list, so you can quickly narrow or broaden your search as you scroll through the individual documents. Search filters are the quickest and easiest way to run a search — it’s as simple as a single mouse click! But even better, each filterable value has a document count so you can rapidly understand and explore your data. Each of these elements are designed to work together, so you can efficiently distill the signal from the noise. 

But search isn’t limited to a single page in the product. Search syntax and search builder are also used to create search term reports, quality control samples, and review stages. Unlike some products, we don’t force users to search in a particular way or limit filterable data to the columns in the document list. Search filters, search visualizer, and the search bar can get you to any document you’re looking for, from any set of data you choose to use, through any of our search tools, all without having to clutter the document list with unneeded columns.

We give you the right search tools wherever you need them, so you can work quickly and efficiently.

Sweating the details

We’ve learned the smallest details matter when providing a search experience that feels effortless. This not only extends to what you use to search (what tool you have to use) but also to how you interact with the tool you’ve chosen to use for your particular search. For example,  we recently expanded the number of items you can filter by in our standard filters pane. In doing so, we took the opportunity to redesign the interface to help you visualize your data and make searching as easy as shopping online. 

Take for example DISCO’s new histogram search filters, you can simultaneously explore and search your data in the way that fits you best. Quickly drag the thumbs to narrow your search criteria, select a specific range, or enter the values directly — whichever you choose. And importantly, we use descriptive date formats like Oct 01, 2020 so globally distributed teams don’t have to pause to remember if the date is formatted as m/d/yy or d/m/yy. 

We know there can be thousands and thousands of searchable values for some document attributes like email recipients. That’s why DISCO makes it easy to navigate into a focused view for any individual filter attribute where you can sort the values by count or alphabetically. 

We provide so many types of information in our filters, we even came up with a way to help you locate the filter you’re looking to use by searching for filter values that match your specified criteria. That’s right — you can rapidly search through all the searchable values!

The documents you work with contain so much more than just recipients or senders. Understanding this, we don’t provide only simple filters like a list of email recipients. Ingest location is an example of a powerful filter that helps you explore where documents were collected from a custodian’s hard drive. We represent every folder path within a custodian’s ingest, so you can understand where and how their data was organized. That makes conversations with your clients or key custodians easier since you can speak about the data locations in the same language. 

We mentioned there are millions of searches run in DISCO every year. Almost all of our users run their own searches — litigation support, first-year associates, more experienced associates, and even partners and corporate clients! We sweat the details to make DISCO search so effortless that there’s no need to ask a colleague (or vendor) to run searches for you. 

When it comes to performance, good is never good enough

We measure search performance in milliseconds even when searching across millions of documents. But we don’t only measure the speed of returning documents in the search results. Our commitment to clients is there’s no speed that is “fast enough,” and that is consistent for every critical interaction within the product. 

We recently improved the speed at which document counts are updated as users work in search filters by more than 50%. After every search update, we immediately update the document counts included for every available search filter. This helps you rapidly iterate through searches until you find the documents you’re looking for. 

We use many tactics to constantly improve performance. Besides using the latest cloud-native technology, we intelligently prefetch and update data using each click and scroll as a signal of your next action. This allows you to seamlessly explore, discover, and search quickly and easily. 

Improving your legal outcomes with search

At DISCO, we have a perpetual focus on making ediscovery easier, faster, and more accurate. We’ve learned that well-designed, incredibly fast search is one critical factor in helping our clients improve their legal outcomes. Want to see DISCO’s search in action? Schedule a demo with us.

Subscribe to the blog
Jordan Janes