Legal Team Roles: Who Moves the Work Forward
Summary
Your firm almost certainly has clear job titles. Most do. What far fewer have is a clear answer to a much more practical question:
Right now, on this task, who is responsible for moving it forward?
Titles describe what someone is qualified to do. They don't say who is accountable for progressing a specific piece of work today, and that gap is where deadlines quietly go missing. It's also where a surprising amount of daily anxiety lives, because when responsibility is vague, people spend energy wondering whether something is theirs instead of doing it.
This guide is about closing that gap: what clarity of responsibility actually buys you, why supervision and delegation are the same activity viewed from different ends, and why the single most useful decision you can make is to allocate work by role rather than by person when it comes to your workflows or standard processes.
Contents
- The question that actually matters
- What it costs when people have to wonder
- Dilution of responsibility, and what it does to a team
- Supervision, delegation and leverage are one thing
- The upskilling flywheel
- Allocate by role, not by person
- What role-based allocation unlocks
- Key takeaways
- Frequently asked questions
The question that actually matters
Legal teams know their structure. Everyone understands who the senior lawyers are, what a paralegal does, where the line sits between substantive and administrative work. That isn't the problem, and any guide telling a firm to define its roles is answering a question nobody asked.
The problem is narrower and much more practical. On any given matter, at any given moment, there are a dozen things that need to happen next. And for a good number of them, no one person is unambiguously responsible for making sure they happen.
Not because responsibility was never discussed. Because it was discussed in general terms, about categories of work, and then never attached to the actual task sitting in front of the team this Tuesday.
Key point: Knowing who does what kind of work is not the same as knowing who is progressing this task. Most firms have the first and assume it delivers the second.
What it costs when people have to wonder
The obvious cost is work that stalls. The less obvious one is the mental overhead paid by everyone around it.
When responsibility isn't explicit, people carry a running background question: is that mine? They check. They ask a colleague, who also isn't sure. They decide to leave it because someone more senior is probably across it, then worry about it again that evening. Multiply that by a team and a caseload and it amounts to a significant tax on attention, paid by conscientious people precisely because they are conscientious.
The ones who feel it worst are usually your best staff, because they're the ones who notice the gaps in the first place.
Explicit ownership removes the question entirely. Not by making anyone work harder, but by giving them one less thing to hold in mind.
Dilution of responsibility, and what it does to a team
When something does get missed, the instinct is to widen the net: make sure everyone is watching for it next time. That instinct makes the problem worse.
"Clarity beats shared responsibility every time. If a job belongs to everyone, it belongs to no one, and you find out which on the day it's missed."
— Ashley Kelso, Hivelight
Everyone has seen the aftermath. A deadline passes and four capable people each assumed one of the other three had it. Nobody was careless. The task simply had no name attached, and diffuse responsibility feels like coverage right up until it isn't.
What happens next is the part firms don't discuss. A missed deadline creates an emergency, the emergency creates blame, and the blame lands unevenly, usually on whoever is least able to push back. Staff who were never told the work was theirs end up defending themselves for not doing it. That is corrosive in a way the missed deadline itself isn't, and it happens most in the firms with the least visibility, because when nobody can see what actually occurred, the argument is decided by seniority rather than facts.
Accountability and visibility together prevent both halves. If every task has one owner and the state of the work is visible, deadlines are seen coming instead of discovered late, and on the rare occasions something does slip, the record shows what happened rather than leaving people to litigate it.
Key point: Vague responsibility doesn't just cause missed deadlines. It causes the argument afterwards, which does more lasting damage than the deadline did.
Supervision, delegation and leverage are one thing
These get treated as three separate topics. They're the same activity seen from different ends.
Delegation is pushing work down to the level that can competently handle it. Supervision is the guardrail that makes doing so safe. Leverage is the result: a firm that produces more than its senior people could produce alone.
Take away any one and the other two stop working. Delegation without supervision is how mistakes reach clients. Supervision without delegation is just senior staff watching people do work they'd have been faster doing themselves. And leverage isn't a separate initiative at all, it's what you get when the first two are running properly.
The practical form this takes is unglamorous: every task has someone doing it and, where it matters, someone reviewing it. Both are decided when the work is planned, not when it lands. (The economics of this are covered in delegation in a law firm.)
The upskilling flywheel
There's a compounding effect here that firms rarely make explicit, and it's worth doing so because it changes how you think about pushing work down.
Work slightly beyond someone's current range is how they grow. Someone doing the same tasks they did two years ago is not becoming more valuable. Someone regularly handed work at the edge of their competence is. Guardrails are what make that safe to do on purpose: you can stretch people deliberately when you know a mistake will be caught before it reaches a client or a court.
And the economics of that deserve naming. Capability moves continuously. Salary moves in steps, once or twice a year. So a staff member who is being stretched is producing more work, and more valuable work, across the months between reviews. The firm gets the benefit of that growth as it happens. The staff member arrives at their review with a demonstrable case rather than a feeling, and a stronger one than they'd have had doing the same work all year.
That's a better deal for both sides than the alternative, which is a firm where people do exactly what their title describes until somebody leaves.
Key point: Stretching people is only safe when the safety net is real. Build the guardrails first and the upskilling takes care of itself.
Allocate by role, not by person
Here is the decision that determines whether any of this survives contact with a growing firm.
A workflow that lists the tasks and stages of a matter type but says nothing about who does each task and who reviews it is half a process. It tells you what needs doing and leaves the most error-prone part, allocation, to be improvised every time.
So firms fill the gap by writing names in. And that works, briefly, until:
- Someone is promoted, and half your templates now allocate work below their level
- Someone leaves, and every template mentioning them needs editing
- You hire, and nothing routes to the new person until someone updates the templates
- You start a second team on the same matter type, and discover your process is hard-wired to the first team's names
That last one catches firms out most. You build a good process around the team running family law, it works, so you grow and put a second team on family law. The process should transfer intact. Instead it needs rebuilding, because it encoded people rather than positions.
Specifying by role fixes it permanently. The workflow says this task goes to a paralegal-level person and is reviewed by a senior lawyer. Who that resolves to is worked out per matter, from whoever is on it. Promotions, departures, new hires and new teams all just work, with no template maintenance at all.
Key point: Allocate by person and your processes need maintaining forever. Allocate by role and they keep working through promotions, departures and new teams.
What role-based allocation unlocks
Once work is allocated by role rather than name, a set of things become possible that simply aren't otherwise.
Work routes and re-routes itself. When someone comes off a matter, goes on leave, or leaves the firm, their tasks travel to the next appropriate person by role rather than sitting assigned to somebody who isn't there.
This one is worth separating from the rest, because it's the only item on this list you can't achieve with discipline alone. Everything else here is a decision you can make on paper: write the owner against the task, specify roles instead of names in your templates, agree who reviews what. All of that works in a spreadsheet if you're rigorous.
Re-routing doesn't. For work to move when a person moves, something has to know the roles, know who is on the matter, and act on both without being asked. That's software, and it has to be software that understands seniority rather than just storing a name in a field. Most legal tools don't: they'll leave a task assigned to someone who left in March and consider that correct behaviour, because as far as the system is concerned the field is populated.
Hivelight does this automatically, and it's the main reason role-based allocation is worth setting up properly rather than approximately. (Mechanics are in the help centre.)
Senior staff stop being the allocation department. Manual reallocation is a genuinely expensive habit, because it consumes exactly the hours you can least spare. Work that routes itself removes that job.
Work reaches suitable people who have capacity. Allocating by role turns "who could do this?" into a question the system can answer, rather than a judgement someone makes from memory about who looks busy.
AI gets usable context. Any AI you bring in needs to know what a task is, what standard it's held to, and where it sits in the matter. Work that's specified by role and task type carries that context already. Work allocated ad hoc to whoever was free carries none of it. (More in support staff, specialists and AI.)
None of that requires a re-organisation. It requires writing down, once per matter type, which level of person does each task and who checks it.
Key takeaways
- The useful question isn't who does what job. It's who is responsible for progressing this task, right now.
- Vague responsibility taxes your best people, who spend attention wondering whether things are theirs.
- Shared responsibility isn't coverage. One owner per task, or you get the missed deadline and the argument that follows it.
- Delegation, supervision and leverage are one activity. Remove any one and the other two stop working.
- Guardrails let you stretch people on purpose, and capability grows faster than salary steps, which benefits the firm and gives the staff member a real case at review.
- A workflow that doesn't say who does and reviews each task is half a process.
- Allocate by role, never by name. Names need maintaining through every promotion, departure and new team. Roles don't.
- Role-based allocation is what enables auto-routing, capacity-aware assignment, and usable context for AI.
Want to see work that allocates itself? Take a look at Hivelight.
Frequently asked questions
Who should be responsible for what in a law firm's support team?
The more useful question is not what each person is responsible for in general, but who is responsible for progressing each specific task right now. Firms usually have the first and assume it delivers the second. It doesn't. Put one name against every task at the point the work is planned, and decide who reviews it at the same time.
What's the difference between a Matter Owner and a Matter Lead?
The Matter Owner is ultimately responsible for the matter. The Matter Lead is responsible for its day-to-day management. One of each per matter, sometimes the same person on a small file and usually different people on a large one. The distinction matters because it separates accountability for the outcome from accountability for keeping things moving, which are different jobs and often belong to different people.
Should we build a responsibility matrix?
Usually not, or at least not first. A full matrix takes a fortnight to build and is out of date within two months, at which point it's worse than nothing because people trust it and it's wrong. Start with one owner per task and a named owner and lead on every matter. Add structure only where ambiguity keeps causing real problems.
How do we allocate work when more than one team runs the same matter type?
This is exactly why work should be allocated by role rather than by name. A process built around the names in your first family law team has to be rebuilt when you stand up a second one. A process that says "paralegal-level person does this, senior lawyer reviews it" transfers intact, and keeps working through promotions, departures and new hires without anyone editing a template.
Doesn't pushing work down risk mistakes reaching clients?
It does, unless the review step is designed in at the same time. That's the point of treating delegation and supervision as one activity rather than two. Guardrails are what make it safe to stretch people deliberately, which is also how they grow into more valuable work.