New Filevine Feature Uses AI to Summarize Incoming Leads and Predict the Best Matches for A Firm

From LawSites:

The law practice management platform Filevine today launched an enhancement to its Lead Docket lead intake tool that uses artificial intelligence to generate summaries of the key details of incoming leads and that help predict which leads are good matches for a law firm and which should be referred elsewhere. 

The new feature, LeadsAI, generates a summary of all the information provided in a lead intake form, including lead details, message correspondence, and notes, Filevine says, enabling attorneys to get a quick and accurate summarization of the details important to the firm. 

According to Filevine, features of LeadsAI include:

  • Lead Summarization: This feature uses AI to synthesize relevant data points from a lead, so a firm can quickly review the major details. For a personal injury lead, for example, a summary will include the incident, injury, contact information, pain points, and current state of the case.
  • Message Summarization: LeadsAI can use sentiment analysis to summarize the tone of communications between the intake professional and the potential client. In this way, Filevine says, firms can better understand engagement patterns and identify opportunities to refine their outreach. 
  • Notes Summarization: LeadsAI can generate consolidated timelines of all activities related to a lead, saving firms from having to sort through emails, calendars, and status reports. 

LeadAI also includes a predictive analysis tool that can help firms evaluate potential cases to determine whether or not representation is advised, Filevine says. Filevine says it uses proprietary data to predict whether a new lead or case should be signed, referred out, or rejected.

As it is released today, the tool is trained only on motor vehicle accident cases, but Filevine says it will eventually expand its scope to be useful for firms that handle large volumes of cases in personal injury, mass tort or class action litigation. 

Link to the rest at Lawsites

PG’s contacts within the larger legal community have atrophied to some extent since he took down his shingle. That said, he wonders if the OP describes a solution searching for a problem.

At least in the United States, clients usually come to lawyers in two ways:

They contact an attorney’s office and either:

  • 1. make an appointment to see an attorney (90%) or
  • 2. have a short phone discussion with an attorney so the attorney can determine if this is something he/she handles and respond accordingly (10%).

For PG, the face-to-face meeting was best, because, in addition to hearing about the prospective client’s legal concerns and deciding if it was a matter he could/would handle, it allowed PG to determine if the prospective client was a crazy person.

To be clear, the very large majority of those who contacted PG to talk about representation were perfectly normal individuals who needed some legal help. In PG’s experience, this was the case for other attorneys as well. The general populace does not produce a large number of crazy people. (Reasonable minds may differ on this topic.)

That said, PG could sometimes be fooled. “Fools can be so ingenious.”

On those few occasions when a crazy person did slip through PG’s vetting process, more than a few ended up (with identities fully protected and when PG was at least 300 miles from home) as PG’s most popular war stories shared with other attorneys only.

PG’s cats and copperheads case was one of his most popular war stories.

4 thoughts on “New Filevine Feature Uses AI to Summarize Incoming Leads and Predict the Best Matches for A Firm”

  1. Let’s not forget the two, by-far-most-prominent, means of obtaining clients:

    (1) Being on retainer for, or working for a firm that is captive to, a liability insurer.

    (2) Moving from a government legal position (however minor) to a firm (however small) practicing in the same area.

    Neither of which undermines PG’s point at all, and probably reinforces it: The described software program is a solution in search of a problem (and, of course, the client’s wallet… or, these days, recurring ACH transfer).

  2. I work in personal injury. Like PG, I wondered what problem this is intended to solve. My guess is that this is coming from a referral service. These guys advertise widely for injured parties, then farm them out to practices that pay a fee for referrals. We tried one out for a time. The problem is that the referral services were not attorneys, much less attorneys licensed in every jurisdiction, so they could not tell a caller that there was no case. The call got forwarded to us for that, resulting in a lot of time wasted on junk calls. We got some real cases out of it, but not enough to justify the time and expense, so we dropped it.

    So I wonder if what we have here is a way for the referral service to have the conversation with the prospective client, summarize it, and pass this on to the local attorney to review without having to talk to anyone. I wonder what is the follow-up with the non-cases. Does someone call them back and say they don’t have an lawyer who wants to take the case, or are they simply ghosted?

    Not that anything about this requires “AI.” That is simply hype.

    • Richard, my experience in creators’-side IP/entertainment practice bears that out. There was no (possible) referral service — not even by “volunteering” for arts organizations, because so much of that was orthogonal to my expertise and skill set (leaving aside the ethical quandaries raised by most members of those panels, and even governing bodies, working either in unrelated areas of law or for the Other Guys at the day job).

      I turned away about 65%† of inquiries during the initial consultation, either by phone or e-mail, because there was nothing I could do to assist them. Even in a multijurisdictional practice, certain kinds of disputes are clearly “local” and need to go to local counsel. Even more frequently, the creator had done something (or done nothing for too long!) that made the claim not just “economically nonviable,” but impossible to prevail upon.

      About a third of what remained got politely “terminated” after further investigation. That’s consistent with my general experience in both this and other areas. I suspect it’s similar for everyone except those actually working for insurers or in-house…

      † I kept records and ran analyses once there was a statistically acceptable baseline, and then repeated them periodically thereafter. It was pretty constant.

Comments are closed.