Portfolio Control
How do we spot patterns across case outcomes portfolio-wide?
Updated July 2026
You compile your own outcomes into structured data, then read them by dimension. Group closed matters by defense firm, venue, case type, and settlement timing, and the patterns surface: which firms win hard cases, which venues cost the most, where you settle late and pay for it. The prerequisite is structure. Patterns only appear when the facts stop living in separate reports.
What patterns can you actually see across a portfolio?
The ones that repeat across cases but hide inside single files. Defense firm performance, which firms win the hard matters and which run up cost. Venue behavior, where awards and cycle times run high. Case-mix, which types drive your exposure. Settlement timing, whether you resolve early or pay a late premium. Each is invisible one case at a time and obvious once the outcomes sit in one structured view.
| Dimension | The question it answers | The decision it drives |
|---|---|---|
| Defense firm performance | Who wins the hard cases and who runs up cost? | Where to assign your strongest counsel |
| Venue behavior | Which venues drive the largest awards and longest cycles? | How to weigh venue in strategy and settlement |
| Case-mix | Which case types concentrate your exposure? | Where to focus early intervention |
| Settlement timing | Where do you settle late and pay a premium? | Which matters to resolve before the demand climbs |
Why is structured data the prerequisite?
Because a pattern is a comparison, and you cannot compare prose. The facts that define an outcome, venue, opposing firm, exposure, timing, and result, live as narrative inside separate status reports. Until those facts are extracted into fields, every question requires re-reading files. Structure turns hundreds of individual reports into one dataset you can group, sort, and filter. No structure, no pattern, only anecdotes.
- A pattern is a comparison across many matters, and prose cannot be compared at scale.
- The facts that define an outcome sit as narrative in separate reports.
- Extracting them into fields turns the reports into one queryable dataset.
- Without that structure, you have anecdotes, not a portfolio-wide read.
How does CaseGlide build the dataset?
It structures the status reports your defense counsel already file. Case Clerk AI reads those reports, not your email or work product, and extracts venue, opposing firm, posture, exposure, key dates, and outcome into fields. Repeated across every matter, that produces a portfolio-wide dataset. Then you can group by any dimension and read the pattern instead of reading the files.
- Defense counsel file status reports as they already do.
- Case Clerk AI reads those reports only and extracts the outcome facts into fields.
- Every matter contributes the same structured fields, building one dataset.
- You group by firm, venue, case type, or timing and read the pattern directly.
What do you do with the patterns once you see them?
You act on them, case by case, going forward. Assign your strongest counsel to the firms and venues where the tail is fattest. Resolve the case types you consistently lose late before the demand climbs. Reprice or renegotiate with firms whose results do not justify their spend. The pattern is not a report to file. It is a set of decisions about the open book.
- Match your best counsel to the venues and firms where exposure concentrates.
- Resolve the case types you tend to lose late before the demand hardens.
- Renegotiate with firms whose results do not justify their cost.
- Turn the pattern into decisions about open matters, not a report that gets filed.
Common questions
Is spotting patterns the same as predicting outcomes?
No, and the distinction matters. CaseGlide does not predict verdicts or forecast what a given case will do. Spotting patterns is descriptive: it reads your closed outcomes and shows, from history you already own, which firms won hard cases, which venues cost the most, and where you tend to settle late. That is a read on the past to inform the present, not a forecast of any single matter. The judgment about each open case stays with your team. What the patterns give you is a factual basis for that judgment, drawn from your own record rather than instinct or a borrowed benchmark from someone else's book.
Which plaintiff firms drive exposure→What data do I need before patterns are reliable?
Enough closed matters, tagged consistently on the dimensions you want to compare. A pattern in ten cases is an anecdote; a pattern across hundreds, grouped by venue, firm, and case type, is a signal. The dimensions that matter most are opposing or defense firm, venue, case type, exposure, and settlement timing, because those are the levers you can actually pull. The common failure is not too little data but data that was never structured, so the same fact is captured three different ways or not at all. Consistent structure across the book is what makes a pattern trustworthy rather than a story you talked yourself into.
How much litigation data is trapped in emails→Which pattern usually pays off first?
Settlement timing, for most programs. The single most common and most expensive pattern is settling late: a case that could have resolved early instead runs for months, accrues defense cost, and settles for more once the plaintiff has leverage. Seeing that pattern across the portfolio, which case types and which firms consistently drive late resolution, points straight at money you are leaving on the table. Venue and firm patterns matter too, but they usually inform strategy over a longer horizon. Late-settlement patterns tend to surface savings you can act on inside the current book of open matters, which is why they often show a return first.
Why litigated claims settle late→Can I compare my defense firms fairly this way?
Yes, if you compare like with like. A firm that handles your hardest cases in your toughest venues will show worse raw numbers than one handed routine matters, so a fair read groups by case type and venue before ranking on result and cost. Done that way, the pattern shows which firms actually outperform on comparable work and which do not. That is the basis for assigning your strongest counsel where exposure is real and renegotiating where results lag. The structured dataset is what makes the comparison fair, because it holds the context, the venue and difficulty, not just the headline outcome.
Measure defense attorney performance→CaseGlide is the litigation intelligence platform for Fortune 500 legal departments and insurance claims organizations. It structures live litigation data from defense counsel into executive decisions: reducing defense spend, settling the right cases sooner, and shrinking litigated claim volume.
Next step · See it on your docket
See what your litigation portfolio is telling you
A 30 minute walkthrough on your own docket. No slides, no committee.