The MATTER Method manifesto.

This sounds obvious and is not. Most practices are organised around something else entirely. A solo lawyer is often organised around their inbox: the work exists wherever the latest email is. A small firm is organised around people: that case belongs to a specific colleague, and its entire state lives inside their head. Document-heavy practices are organised around files: the matter is wherever the folder is, and the folder is wherever someone last saved it.

The MATTER position is that the matter itself must be the container, and everything else, every action, deadline, person, document and piece of research, attaches to it. One matter, one place, one state. If someone asks where a matter stands, the answer should exist as a fact you can look at, not a reconstruction someone performs from memory, inbox archaeology and three phone calls.

Here is what years of practice tend to obscure: within any given matter type, the actions repeat. Two disputes in different cities are not the same case, but they are overwhelmingly the same sequence of actions with different facts poured in. Practitioners experience this repetition as expertise, but expertise is simply an action library stored in one person, retrievable only by that person, and lost when they leave.

The MATTER position is to extract the library. Every matter type a practice handles has a describable action sequence, and that sequence should exist outside anyone's head, as an explicit, ordered, improvable playbook. Not so people become script-followers, but for the opposite reason: once the routine sequence is externalised, attention is finally free for the parts that are genuinely not routine.

The billable hour turned time into revenue and, in doing so, trained an entire profession to record time rather than manage it. Meanwhile the two forms of time that actually decide outcomes go largely unmanaged: deadlines, which are absolute and unforgiving, and durations, how long each action actually takes, which nobody measures because nobody named the actions in the first place.

The MATTER position treats time as a property of the matter, not a product sold to the client. Every matter carries its dates as structured facts that surface themselves before they become urgent, and every named action accumulates duration data across every matter, permanently. A practice that knows its true durations can price fixed fees with confidence instead of fear, because pricing stops being a guess about time and becomes arithmetic on data.

Traditional practice answers this by hierarchy and habit. Work flows to whoever is available, review happens if someone senior has time, and responsibility is discovered retroactively, usually when something has already gone wrong.

The MATTER position is that roles attach to actions, explicitly. Every action has a doer and, where it matters, a reviewer. This sounds bureaucratic until you notice what it enables: once roles are explicit, team member stops being a category limited to humans. Most organisations cannot safely delegate to AI because they never built a delegation structure for humans either, no defined doer, no defined reviewer, no explicit handoff. Who does, who reviews, who decides is the entire governance model of safe AI-assisted work.

Everything asserted in matter-based work stands on something: a document, a testimony, a precedent, a fact. Evidence is the primitive of grounding, and it is, in most practices, the most disorganised primitive of all. The documents exist, thousands of them, but the connections do not. The link between one clause and the page that supports it lives in the drafting person's memory, to be rebuilt live, under pressure, months later.

The MATTER position is that the connections are the asset. Every claim in every draft should be anchored to its source, page and paragraph, as structure, not memory. A practice that does this owns something durable: its work product becomes self-verifying, and AI fabrication has a standing answer, a rule that no assertion enters work product without an anchor to a source.

The traditional model treats research as an event. A question arises, someone researches it, an answer comes back, and the work evaporates. The memo serves its matter and dies in its folder. Next year, the same question arises in a different matter, and someone researches it again, from zero.

The MATTER position is that every question answered should compound into a growing, searchable body of institutional knowledge, attached to the matters it served but living beyond them. A practice that runs research this way for a few years develops institutional memory no rival can buy. Machines are extraordinary at retrieval and synthesis, which makes a well-structured research stock exponentially more valuable, and unreliable at invention, which makes grounding research in verified sources non-negotiable.