Predict for solo & small plaintiff firms.
1–10 attorneys, paralegal-supported, taking 30–80 intake calls a week. The case-valuation model is the senior partner's head — and the senior partner is on a call. Predict puts the same number on the paralegal's desk, the new associate's screen, and the founder's phone.
“There is a need for a predictive tool to be able to say what case is worth taking. A tool like this would save weeks off my productivity.”
Cases — this week's pipeline
| Case | Jurisdiction | Predicted | Band | Status | Owner | Activity |
|---|---|---|---|---|---|---|
| #2026-0419 Henderson v. Acme Transport MVA · rear-end | TX · Harris | $485K | ± $42K | Above threshold | M. Williams | 2h ago |
| #2026-0418 Vargas v. Stratford Plaza PL · slip & fall | FL · Miami-Dade | $72K | ± $34K | Partner review | D. Chen | 4h ago |
| #2026-0417 Liu v. Crestview Apartments PL · premises | CA · Alameda | $148K | ± $24K | Above threshold | M. Williams | yesterday |
| #2026-0416 Okafor v. Beltline Logistics MVA · intersection | NC · Mecklenburg | $58K | ± $11K | Below — refuse | Paralegal | yesterday |
| #2026-0415 Rivera v. Westside Diner PL · slip & fall | TX · Bexar | $94K | ± $48K | Wide band · review | D. Chen | 2d ago |
| #2026-0414 Park v. North Star Freight MVA · rear-end | TX · Harris | $312K | ± $38K | Above threshold | M. Williams | 3d ago |
In a small firm, the senior is the model — and the bottleneck.
A 5-year-out solo PI attorney has settled enough rear-end MVAs to value one in 60 seconds. They've also settled three commercial premises cases — not enough to value a fourth with confidence. The institutional memory works for the cases inside it and silently fails on the cases outside it.
Valuation lives in the senior's head
The paralegal pre-screens, the associate drafts, the senior partner makes the call — every time. The number doesn't scale past the senior's calendar, and nobody else can defend it.
Same number, every desk
Paralegal runs the intake; Predict returns a confidence-banded value in 60 seconds; associate sees the same band; senior reviews methodology, not math. The firm runs the model, not the partner's recall.
Bad cases displace good ones
An $18K-med-specials case that turns into a $14K settlement consumed 80 hours of paralegal time. That's the case-management capacity the firm could have spent on the $180K rear-end the partner turned away because the caller "sounded difficult."
Triage before you take the case
Predicted band under the firm's case-value threshold → respectful "we can't take this one." Same paralegal hours, materially higher revenue per signed case. The triage is the leverage.
From 50 calls a week to 3 better cases a week.
Status quo. The volume is real; the triage is the problem.
Per case. Inside the intake call, before you take the case.
Same intake volume. Same paralegal hours. The case mix shifts.
The lever is case-selection, not volume. Refuse one $14K case that would have consumed 80 paralegal hours, sign one $180K case you would have screened out as "sounds difficult," and the average value of a signed case moves without a single extra intake call. We don't quote a single headline lift number, because the size of that shift depends on your jurisdictions and your current threshold. During the trial you run a real intake week through Predict and see your own shift, with the same confidence band we put on every prediction.
Built for the firm where everyone wears two hats.
A solo or small firm doesn't have a valuation team. The same person who runs intake also drafts demand letters; the same paralegal who calls the carrier also pulls comps. Predict consolidates the valuation step into a 60-second model output — and frees the people doing the work to do the parts a model can't.
The same valuation follows the case across its whole lifecycle: a confidence band on the intake screen, a full demand letter exported in your firm's own style and formatting (drawing on police reports and medical records as they arrive), and an audit trail per case. Same number on every desk, every time. It is a gross case value — before attorney fees, case costs, and medical liens.
The number is the headline; the band is the methodology. On a held-out test set the model lands within 90–92% median accuracy (MdAPE) on higher-value cases, and every prediction is shown with the 90% confidence band you see here, not a bare point estimate. That is the reason to trust it over recall plus a raw comp search. See the accuracy results →
See how it works →Priced for the firm that's paying for it out of practice income.
Triage one week of intake with Predict.
Predict your first case free in 60 seconds, no signup. The 14-day trial unlocks unlimited predictions, a full demand letter exported in your firm's style, and the full comp set behind every number, long enough to put it next to a real intake week. $0 today — card on file, cancel anytime; nothing is charged until day 15, and the 14 days cost nothing.