L

Legal Aid Chicago

Legal Aid Chicago cuts out-of-priority intake rate from 46% to 21% with AI voice screening pilot

Curated & reviewed by Peter Korpak, Founder & Chief Analyst, 100SignalsHow we verify
46% → 21%Out-of-Priority Screening Rate
417 calls (22+ hours)Calls Handled in 3-Week Pilot
4.1/5 avg (55% rated 5/5)Caller Satisfaction Rating

Vendor-reported figures — source: www.justicebench.org

The Challenge

Legal Aid Chicago receives more than 78,000 calls per year — roughly 6,500 per month — for family law help, far more than intake staff can screen, qualify, and refer. Callers routinely hear that the queue is closed, and people who need help don't get through. Under traditional phone triage, 46% of family law callers were ultimately rejected as out-of-priority only after staff had already spent time gathering information.

The Solution

Legal Aid Chicago ran a three-week pilot of Haven, a commercial voice-powered AI intake assistant, on its family law line starting August 11, 2025. Calls were routed to the AI agent, which conducted conversational screening using attorney-reviewed scripts, assessed eligibility against Legal Aid Chicago's criteria, and either flagged callers for staff follow-up or provided referrals, with a staff dashboard for reviewing determinations, transcripts, and overriding AI decisions. No phone infrastructure changes were required, and the team went through 2-3 iteration cycles with Haven, including a mid-pilot redesign of the intake process itself.

Results

Over the three-week pilot, the AI handled 417 calls totaling 22+ hours, averaging 3.24 minutes each. The out-of-priority screening rate dropped from 46% under traditional triage to 21%. Of the 417 calls, 281 were screened out with referrals, 136 were screened in for further review, 44 were flagged as domestic-violence priority intakes, and 20 received specialized handling for urgent cases with court dates within a week — including one caller whose stalking and retaliation risk the AI surfaced for a same-day callback. Among callers who reached a staff follow-up, satisfaction averaged 4.1 out of 5, with 55% rating the experience 5/5.

Key Takeaways

  • Structured AI questioning surfaced domestic-violence risk (e.g., stalking and retaliation) that callers hadn't disclosed through phone-menu options, and callers were more comfortable sharing sensitive details with the AI than staff expected.
  • The pilot exposed pre-existing flaws in the human intake workflow — screening too many callers as eligible — prompting Legal Aid Chicago to redesign its intake questions and process mid-pilot.
  • Deployment required no phone infrastructure changes but took 2-3 iteration cycles with the vendor to tune conversational flow and eligibility criteria.

Share:

Details

Industry
Law Firms
Company Size
MidMarket
Quality
Curated
Last verified
Jul 28, 2026

Have a similar implementation?

Share your customer's AI results and link it to your vendor profile.

Submit a case study →