Glossary
Terms I actually
use, defined plainly.
No jargon for its own sake. If a word shows up more than once across this site, its definition is here.
- Matter
- The single unit everything else in a practice attaches to: one case, one file, one deal. If you cannot open a matter cold and understand its state in a few minutes, the matter has not been built correctly yet.
- System
- What I call a matter setup once it has a fixed sequence of stages, a defined document set, and at least one computed number a lawyer actually checks. Most practices run on habit instead of a system, which is why the habit breaks the moment the lawyer is out sick.
- Computed module
- A single number or chart derived automatically from the underlying facts of a matter, for example, claimed damages measured against evidence-supported damages in a litigation file. The point of a computed module is that nobody has to remember to calculate it.
- Practice pack
- A pre-built system for one practice area, tuned to how that area actually moves. Twenty-two of these live on this site under Practice Systems, each with its own stages, vitals, and checklist.
- Vitals
- The two or three facts about a matter that a lawyer in that practice checks constantly, such as a statute of limitations countdown or a discovery cutoff. Vitals are the reason a dashboard beats a folder.
- The MATTER Method
- My own doctrine for building any of this: name the primitives, build the system, then let AI carry the part of the work that does not need judgment. I wrote the full argument in Writing, and both MatterOS and LexOS run on it.
- Audit, Build, Install, Measure
- The four phases of a custom build, whether it runs through me directly or through adnah as a structured engagement: find what is actually broken, build the fix, put it into daily use, then check that it worked.
- AI doing the heavy lifting
- How I describe the role AI plays once a system exists: drafting the first pass, flagging the exception, doing the repetitive read, while a person keeps the judgment call. AI without a system underneath it is just a faster way to produce unreliable work.
- Evidence exhibit
- In litigation-technology work, the structured, source-anchored body of digital or electronic evidence built for a US court, an SEC matter, or an HNW divorce case. Different discipline from matter management: this is evidentiary craft, not workflow design.
- Digital footprint review
- The first stage of most litigation-technology engagements: mapping what electronic evidence actually exists (devices, accounts, communications, financial records) before deciding what is worth structuring into an exhibit.
- MatterOS
- The agentic matter-management product I built and sell directly at thematteros.com, configurable by practice pack. Adnah's own transformations run on the same product, but the product itself is mine.
- LexOS
- A Notion-based practice-operating system I built and sell as a one-time purchase at lexos.pro, with a lighter free companion at Notion for Lawyers. Independent of adnah's own portfolio, though adnah maintains the Portal shown in its case studies.
- Custom build
- A system built privately for one firm or team rather than shipped as a configurable product. This is the work that also runs through adnah as a structured engagement for firms that want the same transformation applied to their own practice.
- Adnah
- The studio I founded so firms and legal teams that are not me can get the same transformation, run as a proper engagement rather than a favor. Read more at adnahlegal.com.
- Fixed-sequence build
- My term for refusing to let a client buy software before the underlying process has been named. Skipping this step is the single most common reason a legal AI purchase gets abandoned within a year.
- Litigation technology
- The second, separate practice I run: structuring and presenting digital evidence for courts, distinct from the matter-management systems work. See Services for how this engagement is scoped.
- Expert testimony
- Serving as a testifying or consulting expert on the digital and electronic evidence side of a case, once the exhibit has been built. Not every litigation-technology engagement needs this, but some do.
- Onboarding checklist
- The fixed opening sequence for a new matter inside any practice pack, from conflicts check through the first calendared deadline. A checklist exists so the first week of a matter does not depend on memory.
- AI-native practice
- A practice built around AI from the ground up rather than one that bolted a tool onto an old process. I write about this at length under The AI-Native Practice, and the short version is: the advantage belongs to whoever has the least to unlearn.
- AI readiness
- Where a practice sits between doing everything manually and running fully AI-native. The quiz on this site scores a practice against four stages, from manual to AI-native, in about ten questions.
- Hallucination, structurally addressed
- My working answer to AI hallucination is not a better model, it is a rule: nothing gets asserted in a work product without a checkable source attached. That rule lives inside every system I build, not in a disclaimer.
- Systematise before you automate
- The middle step of how I work, and the one people most often skip. A process has to be named and written down before it is worth handing to AI, otherwise you are just automating the ambiguity.
- Independent builder
- How I describe my own working arrangement: no account manager, no vendor team between the client and the person doing the work. You talk to me from the first call through delivery.