This is a live blog post from ILTACon, the 2016 conference of the International Legal Technology Association. This session is the Finding a Needle in a Haystack with 21st Century Expertise Systems with panelists with Kate Cain of Sidley Austin, Marybeth Corbett of WilmerHale, Julie Bozzell of Hogan Lovells, and Joshua Fireman of Fireman & Company.


A session description appears at the end of this post. I wrote this post while the session took place and published as it ended. So please forgive any typos or mistakes of understanding.

As a live post, please forgive any typos or misunderstandings of meaning.

Panel Discussion

We find ourselves using 19th century processes with 20th century technology to solve 21st century problems.

Terminology: expertise, experience, skills… is there a difference? One point of view: Expertise is people focused; it’s what they know and their skills. Expertise is what the organization knows and relates to matters. Skills go more to defined, universally recognized skills.

Is there anything objective here? One POV: combination of objective and subjective. Sidley uses Foundation to capture its experience. To determine expertise requires making some inferences, for example, finding who has experience in a particular jurisdiction. Doing that requires knowing how many matters and hours lawyers have worked on matters. But number of hours is not always the right indicator because the small number of partner hours may count for more than many hours of an associate. So expertise location means finding the right lawyers to tackle a client problem.

Drivers for each firm…

Hogan Lovells: Previously, firm had several ways for lawyers to self-identify experience, and primarily in the US. With lawyers coming and going, finding the expertise, especially on a global basis, was very hard. The firm wanted to find a more systematic way to locate expertise. Firm management to ask for a better solution.

Joshua: Hogan Lovells is obvioulsy large so the problem is clear. But we have observed that even smaller firms, say 150 lawyers, can have this problem.

WilmerHale: The firm has worked on this area for a long time. The firm reached a tipping point when it realized it needed more info on matters and what they were about. Responding to pitches and having to price matters made the need for expertise critical. So many threads came together to address expertise location more systematically and on a firm-wide basis.

Joshua: In the past, expertise location was about finding colleagues to do work. Now, it’s a business driver.

Sidley: When I started about 5 years ago, the firm thought of that as “marketing database” and a marketing problem. And from my own experience, I know that marketing and BD has much to gain from expertise location. But it won’t work as marketing only – it must touch every step of the entire client lifecycle. That means from prospect, to winning the business (including pricing), to staffing the matter, to finding lawyers in the future, to league tables. What really got management attention was staffing the matter as a big driver.

Joshua: so this is not just about selecting a product and rolling out. There are a lot of requirements, data prep, and adoption issues. So let’s look at use cases to help understand this in more detail.

WilmerHale: You have to understand the problem to solve before looking for solutions. Everyone at firm will say “I need to know about our work”. But you have to understand what that means – how much detail do stakeholders need. The answer will drive vendor evaluation and selection.

Sidley: When setting up a system, don’t ask lawyer what fields they want. You will end up with way too many requests. So you have to focus on scenarios that really occur. You can’t just say “let’s search”. You need to know how you will use data – that will drive what you will collect. It has to be balanced though by cost of data acquisition. Key advice: you have to start somewhere. By mapping out final destination, which may be four years out, you can decide where best to start.

Joshua: Defining use cases and scenarios means knowing inputs, outputs, and actors. What are touch points in the firm that affect information flows. Where can you tap those flow? And information you collect from the process may require different presentation lawyers for different purposes, for example, addressing pricing versus staffing matters.

Hogan Lovells: addressing issues Joshua raises meant, for us, to define a global taxonomy. This turned out be a very big project, over 18 months, to align on language and structure. Among other problems were dealing with US v UK spelling differences. Another example, drones have many other terms such UAV. If you rely on searching for words, you have to get the words right. The firm continues to work on the taxonomy. Term-based self-identification is more of a US approach. Rest of world viewed this as a bit of boastful approach. They might say simply “I’m an IP lawyer and do trademarks”.  But that’s not enough detail for most use cases.

Another challenge in taxonomies is getting the right number of skills. Some practices came back with lists too short, others too long. Current limit is 12 but firm likely will grow that number to capture more granularity.

David Hobbie asks if Hogan Lovells also went after industries as well legal areas. Julie replies that industry experience is a separate initiative – with its own taxonomy. The cross referencing of expertise and industries is coming eventually. We will likely require a threshold number of hours worked in an industry to declare expertise in that industry.

Audience Q: How do you maintain the data? How much do you have to track and maintain.

WilmerHale: This is the biggest challenge. You have to consider it upfront. You need to track the lifecycle of the matter at a granular level. This can also implicate allocation of tasks across staff departments. One approach is to break down the work to the most granular and manageable element and then find the most appropriate player in the lifecycle to capture it. Typically, the work is already being done in one way or another. But it’s hard to make sure you can capture it. But key message; break up the work across the lifecycle and functions

Sidley: be as opportunistic as possible; that is, where information already exists such as in NBI or accounting, pull from those sources. You also have to think about who enters information. At Sidley, central team captures matter type to ensure accuracy because decentralized is less accurate and reliable.

Joshua: 20 years ago, we learned that manually collecting all the information was failing. But just pulling information from production systems was not enough because much data requires vetting. Implication: many elements of expertise location takes work. I can see that allocation across staff departments is harder than it sounds given competing demands for time.

WilmerHale: it’s a challenge to get allocation and work balance across departments. We at least have everyone attending meetings. It’s getting easier because almost all stakeholders now feel pain when they don’t have good expertise data. So former resistors are becoming more willing to contribute. But you can’t come across as off-loading all work to someone else. Not quite in these words, but you have to answer the “what’s in it for me” question.

Sidley: Find people when they have a problem and then help them solve that problem. That can win them over in contributing to the effort to collect the appropriate data.

Joshua – let’s talk about the software and solutions you are using. Once you go through requirements analysis and use case development, it can point to different solutions. So what, software did you decide on and why?

Hogan Lovells: Neudesic Firm Directory replaces Intranet people directory. We are brining data form other systems – and this requires a fair bit of clean up. Compared to our prior product, we now have much easier to populate system and much better search and drill down capability. Now working on integrating expertise location with a search system.

Both WilmerHale and Sidley use Foundation software. Sidley pulls data from multiple sources into Foundation. Will tap additional systems over time and also push data out to Interaction CRM. We wanted to find information based on multiple, intersecting fields such as industry, office, and experience type. WilmerHale comes from a somewhat different position. We had purchased the Interaction Matters module. Our Foundation deployment is informed by that experience. Agree that it’s key to understand how you want to get data out of the system.


Session Description

Expertise location systems are ubiquitous at law firms of all sizes and are key solutions that help with everything from responding to client proposals to finding the right attorney to help with a particular matter. An effective expertise location tool can be a differentiator for law firms, yet they are tricky because of the need to pull together multiple sources of information while providing clear answers. People from the trenches will share their experiences of implementing various solutions and tips and tricks to keep in mind when you’re evaluating a new solution.