Research
Patterns are counts, not conclusions.
A future research layer would summarize anonymized structure across many discovery files: which issue categories recur, in which states, in which years, and where contract language is materially similar. It would never expose a narrative, a document or an identity.
Pattern panels
Illustrative pattern intelligence across demo records.
Every figure on this screen is fabricated for demonstration. No real carrier, agency, or platform statistics are represented.
Issue categories
Files by year of events
Geography
- Texas22 files
- Florida18 files
- Ohio13 files
- Arizona11 files
- Georgia9 files
Anonymized cohorts
Retroactive compensation adjustment · 2023–2025
cohort size 34
High contract-language similarity across 21 of 34 files
17 resolved internally · 9 in counsel review · 8 no reported outcome
Termination notice shorter than agreement interval
cohort size 19
Shared clause phrasing detected in 12 agreements
8 resolved internally · 6 in counsel review · 5 no reported outcome
Deferred compensation forfeiture at exit
cohort size 26
Comparable forfeiture language in 15 agreements
11 resolved internally · 10 in counsel review · 5 no reported outcome
What these panels cannot tell you
- Whether any organization acted improperly
- How common an experience is across the real industry
- The outcome of any individual matter
- Anything about a contributor's identity or narrative
Method
What would make a pattern publishable.
Minimum cohort size
No pattern is shown below a threshold that could re-identify a contributor.
Consent-scoped inclusion
Only files explicitly marked aggregate-eligible are counted, and removal is retroactive.
Stated limitations
Every figure would carry its denominator, its date range, and what it cannot support.