Legal Project Management Fundamentals for Firms That Never Did It

Most firms that have never done legal project management assume it is a corporate import, something a consultant sells to a two hundred lawyer outfit with a pricing committee and a dedicated LPM director. It is not. Legal project management is four questions asked before the work starts and answered again at fixed points while the work runs. What exactly are we doing. In what order. Who owns each part. When do we stop and check. That is the whole discipline. Everything else sold under the name is packaging.

The reason small and mid-size firms need it more than large ones is that they have less slack to absorb a matter that goes sideways. A firm of six can lose a quarter to one commercial dispute that was quoted at forty hours and consumed a hundred and ninety, and nobody noticed until the write-off hit the collections report. Nobody was negligent. There was simply never a moment where a specific person was responsible for saying the work had drifted past what the client agreed to pay for.

What follows is how to apply scope, phases, owners, and checkpoints to a real matter, starting from a firm that currently plans nothing and runs on the instinct of whoever happens to be senior on the file.

Scope is a decision, not a paragraph in the engagement letter

Scope in most firms is whatever the engagement letter says, which is usually broad enough to be unfalsifiable. "Represent the client in connection with the dispute" tells you nothing about whether an appeal is in, whether a parallel regulatory response is in, or whether three rounds of document review were contemplated when the fee was quoted. A scope statement that cannot be violated is not a scope statement. It is a liability shield, and it does a fine job of that, but it will never help an associate decide on a Tuesday afternoon whether the thing the client just asked for is inside the deal.

Write scope twice. Once for the client in the engagement letter, in the register your regulator expects, and once internally in language your team can actually apply. The internal version names what is excluded as clearly as what is included. Appeals excluded. Third party claims excluded. One round of substantive revisions on the draft, further rounds billed separately. Costs disclosure and fee variation rules differ meaningfully across jurisdictions, and the duty to update an estimate when it materially changes is explicit in some places and softer in others, so confirm what your own bar or regulator requires before you rely on the internal note as a client-facing document. The internal note is for your team. The client-facing scope still has to satisfy local rules.

Phases are how you make a matter estimable

A matter estimated as one number is a guess. A matter estimated as five phases is five smaller guesses, and the errors do not compound the way people fear, they partially cancel. More importantly, phasing gives you a place to put a checkpoint. You cannot check in on "the litigation." You can check in at the end of investigation, before discovery opens, when the actual cost of the next phase is knowable in a way it was not at intake.

Phases should follow the real sequence of the practice area, not a generic project template. A transactional matter phases as engagement, diligence, drafting, negotiation, closing. A defence matter phases as intake, investigation, pleadings, discovery, motions, resolution. An immigration or estate matter phases differently again. The test is simple. If your team already uses the phase name out loud in a corridor conversation, it is a real phase. If you invented it to make the chart look balanced, delete it.

Every phase needs one owner, and one means one

The single most common failure in firms new to this is assigning a phase to "litigation" or "the team." Shared ownership is the absence of ownership. When discovery has three names on it, the person who notices a deadline drifting assumes one of the other two is handling it, and that assumption is completely reasonable, which is exactly why it is dangerous. One named owner per phase does not mean one person does the work. It means one person is accountable for the phase reaching its checkpoint on time and inside budget.

The owner does not have to be the most senior lawyer on the matter. For document-heavy phases the right owner is often a senior paralegal who genuinely knows where everything is, with the supervising attorney reviewing at the checkpoint. That distinction matters for cost as much as for clarity, because a phase owned by a partner tends to get partner-rate attention on tasks that never needed it. Assign ownership to the person who will actually chase the phase, then make sure the system reflects it so nobody has to ask who is running discovery this month.

Checkpoints are the point where someone is allowed to stop

A checkpoint is not a status meeting. A status meeting reports what happened. A checkpoint is a decision gate where a specific person confirms three things: the phase that just ended is genuinely complete, the budget consumed matches the budget planned, and the next phase is still the right thing to do. The third one is what firms skip, and it is the only one that ever saves a matter. Matters run over because nobody is ever formally permitted to say the plan changed.

Put the checkpoint on the calendar as a date, not as an event that happens when the phase finishes. If discovery is planned to close on the fifteenth and it is the twenty-second with no checkpoint held, that gap is the signal, and it is more useful than any report you could run. In Casely the deadline diary attaches deadlines to the matter itself with next-date auto-tracking, so a checkpoint sits in the same place as the court dates rather than in someone's personal calendar where nobody else can see it slip.

Why matters run over, and it is almost never the obvious reason

Ask a partner why a matter blew its estimate and you will hear that the other side was difficult, or the client kept changing their mind, or the documents were worse than expected. Those are all real, and none of them are the reason. Difficult opponents and messy documents are the normal condition of legal work, and a plan that only survives easy matters is not a plan. The reason matters run over is that the overrun was invisible until the invoice, which means nobody could act on it while acting was still cheap.

There are four mechanisms, and they are boringly consistent. Work was done that was never in scope and nobody flagged it. A phase quietly absorbed the next phase's budget because the boundary was never defined. An owner changed by drift rather than by decision, and the new owner inherited no context. And time was recorded late, so the budget picture the team was working from was days or weeks behind reality. Every one of those is a visibility failure, not a competence failure, which is genuinely good news, because visibility is the cheapest thing to fix.

Scope creep does not announce itself, it arrives as a favour

Nobody sends an email titled "expanding the engagement." They ask whether you could take a quick look at the supplier contract while you are in there, and the associate says yes, because saying yes is what good service feels like. Ten of those across a matter is a phase of unbilled work that never appeared in any estimate. The problem is not that the answer was yes. In many cases yes is commercially correct. The problem is that the yes was never recorded, never priced, and never surfaced to whoever owns the matter's economics.

The fix is procedural and it is small. Any request that falls outside the written exclusions gets logged against the matter with a one-line note before the work starts, and the phase owner decides whether it is absorbed or billed. In Casely, connected matters exist precisely for the version of this where the request is genuinely a different matter, linking the related file with the reason stated so the relationship survives after everyone involved has forgotten the conversation. Small favours that become their own file should look like their own file.

Budget belongs to the phase, not to the matter

A matter-level budget tells you that you are eighty percent through the money. It does not tell you whether that is fine or catastrophic, because it says nothing about how much work remains. A phase-level budget answers that immediately. Sixty percent of the money consumed with drafting and negotiation still ahead is a problem you can name on the day you notice it, and naming it on that day is what makes the client conversation possible.

This is also where billing model and project structure meet. Flat-fee and contingency matters need phase budgets more than hourly ones, not less, because the firm carries the entire overrun risk. Casely handles hourly, flat-fee, contingency and blended billing natively, and one-click invoicing converts every unbilled hour on a matter into a single itemised draft, which matters here for a reason that is not obvious: the fastest way to know whether a phase is on budget is for the time to already be captured against the matter rather than sitting in someone's notebook until Friday.

Making the plan visible is the entire trick

A plan that lives in the partner's head is not a plan, it is a preference. A plan that lives in a spreadsheet on a shared drive is marginally better and still fails, because the spreadsheet and the matter drift apart within about three weeks and nobody trusts it after that. The plan has to live on the matter, in the same place the documents and the deadlines and the time entries live, or it will be abandoned quietly and everyone will go back to running on instinct.

This is what a configurable stage tracker is actually for. Casely's matter stage tracker is a clickable stepper on the matter itself, configurable per firm and per practice area, so the phases on screen are the phases your firm genuinely uses rather than a vendor's default pipeline. When an associate opens the file they see the current phase without asking anyone. When a partner scans the caseload they see which matters have been sitting in the same phase past their checkpoint. The abstract plan becomes a thing the whole team can look at, and that visibility is doing almost all of the work.

Handoffs are where legal project management usually dies

Phases end. People move between them, go on holiday, leave the firm, get pulled onto a trial. Every one of those transitions is a moment where context either transfers or evaporates, and in most firms it evaporates, because the context was never written down anywhere except in the outgoing person's memory of the last four months. The new owner rebuilds it by reading the file from the start, which is real work nobody budgeted and nobody bills.

Make the handoff a checkpoint of its own with a written statement of where the phase stands, what is outstanding, and what the incoming owner should not assume. Document-level context helps enormously here, and in Casely every document carries a comment field recording what changed and why, so the incoming owner reads the reasoning rather than reverse-engineering it from three similarly named drafts. Ethical walls matter at handoffs too, since they are enforced at the server and data-access layer, meaning a walled user genuinely cannot reach a restricted matter through search, the calendar, or a forwarded link, and a handoff never accidentally becomes an access problem.

Checkpoint the client, not just the team

Half the value of running a matter as a project is external. Clients do not usually object to a matter taking longer than expected. They object to finding out late. A checkpoint that produces a two-sentence client update at the close of each phase converts the most common source of relationship damage into a routine touchpoint, and it does it before the invoice arrives rather than after, which is the entire difference between a conversation and a dispute.

The mechanics should not cost you an hour each time. Casely's client portal is real-time and privilege-filtered automatically per document, works on mobile, and includes e-signature inside the same login with no separate account to create, so a client checking where things stand does not generate a phone call to a paralegal. When the phase change is visible to the client the moment the stepper moves, the update stops being a task somebody has to remember and becomes a property of doing the work.

Closing is a phase, and skipping it costs you the next matter

Most firms have no closing phase. The work finishes, the final invoice goes out, and the file goes quiet with loose ends still in it: a trust balance not returned, documents never sent to the client, a conflict-relevant party never labelled, no record of what the matter actually cost against what it was quoted. Every one of those becomes somebody's problem later, and the person who inherits it has no context at all.

Treat closing as a phase with an owner and a checkpoint like any other. Return trust funds and confirm the ledger is at zero, which Casely enforces structurally since any disbursement exceeding a matter's actual trust balance is blocked at the database transaction level rather than by a warning dialog, and corrections are voided and remain visible rather than deleted. Label every party with the role they played so conflict checking, which searches the full contact and matter history across every role including closed matters, can find them years from now. Then record the actual versus estimated cost per phase, because that single number is what makes your next estimate on similar work better than a guess.

Where to start if you have never done any of this

Do not roll this out across the whole firm. Pick one practice area and one matter type that you run often enough to learn from, define its phases in the words your team already uses, and assign one owner per phase on the next three matters that come in. That is a week of thinking and roughly an hour of setup. Resist the urge to design a firm-wide methodology first, because methodology written before any matter has run through it is always wrong in ways you cannot predict from a conference room.

Configure the stage tracker to match those phases so the plan is visible on the file rather than in a document nobody opens, put the checkpoint dates in the deadline diary alongside the court dates, and hold the checkpoints even when the matter is obviously fine. The discipline is built by holding the boring checkpoints, not the alarming ones. After three or four matters you will have real numbers for how long each phase actually takes in your firm, which is when estimating stops being a performance and starts being arithmetic. If you want to see how the phases, owners, deadlines and budget sit together on a single file, our matter management software page walks through the structure, and legal billing software covers the phase-level budget side in more detail.

None of this requires a new job title or a certification. It requires that somebody, by name, is accountable for each phase of each matter, and that there is a date on the calendar where that person has to say out loud whether the plan still holds. Firms that do this stop discovering overruns at the invoice. Firms that do not will keep having the same conversation about the same write-off, once a quarter, forever.