Search is Easy
“We find things where we look for them.”
― Robert Brault
Google has shown us that search is easy.
Just type the first couple of letters and Google will instantly know exactly what you want to find. Not only that, Google will put the best possible answers right there at the top of the list instantly. You are either desperate or bored if you need to look past the first page of results.
It’s that good. It’s that simple.
Why can’t eDiscovery search be that simple?
Searching eDiscovery can get much closer to Google’s ease of use but we have a long way to go.
There is endless and exciting talk about analytics and algorithms, predictive coding, computer assistance, machine learning, and artificial intelligence for eDiscovery. Oddly, I’ve noticed that we don’t really mention or think of those technologies when searching with Google.
No doubt there are incredibly sophisticated technologies behind Google’s results but we don’t need to know about them to recognize when we’ve found what we want.
We see it and we know it.
Likewise, we will know these advanced technologies have arrived to eDiscovery search when it disappears from the conversation and our awareness altogether.
“The epitome of useful, technical, and advanced software is the simplification of analysis and research within the UX while harnessing all the technological marvels behind the scenes.”― Aaron Vick
Search is Hard
“Is that it?”
“No. That’s a wall.”
“It could be disguised.”
“You’re not very good at looking for things, are you?”
“I’m good at looking for walls. Look, I found another one.”
― Derek Landy, Kingdom of the Wicked
Searching by the general public is vastly different than searching eDiscovery. The web consumer is almost always looking for the “best” result.
Would it matter if a few million legitimate results were omitted if the first few results provide adequate results to satisfy our query? Would it matter if the actual “best” results weren’t retrieved at all?
Do we Googlers even know (or care!) if we are truly seeing the best results? We consumers only care that we get a good enough answer.
What would “good enough” mean for a legal professional responding to a request?
Consumer web search is often intimately related to the user’s personal and public behaviors as well as the activities of other users much like them.
Google can take advantage of these bits of information to retrieve better results. The users don’t even realize they’ve been telling Google about what they want to find long before they begin typing in their queries. These valuable clues have no bearing on a typical discovery search.
The context and motivation for the eDiscovery searches are completely unrelated to the person performing the search so these factors must be included in the search criteria itself.
You eDiscovery practitioners, on the other hand, are interested in ALL of the results regardless of which “page” the results appear. You want to retrieve only the items that are needed but not one item less than is required.
Sure, Google is searching vastly more data than we are in a single eDiscovery matter but even our meager piles of data overwhelm the capabilities of the voluminous, run-of-the-mill review tools.
It’s not as easy as ranking and filtering the best results to page one.
Multiple adverse parties often must negotiate and agree to the specific methods, techniques, and criteria for executing the searches. Later it may be necessary to demonstrate that the agreed upon methods were properly executed. Sometimes, it is just as important to know why something didn’t make it into the search results as it does to understand why other things did.
In Part 2 we will discuss why we search and the challenges peculiar to eDiscovery searching. Be sure to subscribe to our blog here so you won’t miss the rest of the series.