When I started researching legal technology, the first thing everyone said was "missed deadlines." It came up in every conversation — from advocates, from vendors, from the people writing about the field. The story was always the same. Lawyers miss things. They have too much going on. They forget. And someone, somewhere, was always building the thing that would help them remember.
Then I sat with a single advocate for an afternoon and looked at her diary. She had nineteen active matters. Three of them had hearings the same week, in three different counties. One had a registry filing sitting unprocessed for ten days. Another had opposing counsel who had just dropped a 47-page submission with three working days to respond. She was supposed to draft a sale agreement that evening. Her pupil had a question she hadn't gotten to.
She had not forgotten anything. She knew exactly what was on her plate. The problem was that "her plate" was structurally larger than what one human could carry, and the calendar in front of her — populated by the courts, the registries, opposing counsel, and her own commitments — guaranteed that something would slip. The only open question was which thing.
That afternoon changed how I think about legal tech. Because if she didn't fail because she forgot, then no notification, no reminder, no smarter calendar was going to fix what was actually broken. A large and well-funded part of our industry has been building memory aids for what is fundamentally a planning problem. We have been solving the wrong thing.
Missed deadlines are not always a result of forgetting
The "missed deadline = memory failure" framing is intuitive, and wrong. It's intuitive because the visible failure mode is forgetting. The deadline passes, the lawyer says "I lost track," the file gets fixed at cost. The remediation seems obvious — more reminders, sharper alerts, smarter calendars. An entire product category has been built on this logic.
But the framing collapses on contact with how lawyers actually work. A lawyer with twenty active matters does not have twenty deadlines. She has hundreds. The cognitive question isn't "did I remember this?" It's "which of these forty things due in the next ten days do I touch first, given that I can realistically touch maybe twenty-five of them?"
That's not a memory question. That's a planning, prioritisation, and capacity question — and it's the question reminders cannot answer. A reminder fires when something is due. It does not tell you whether you have the bandwidth to do it. It does not tell you what to drop if you don't. It does not weigh the matter you are about to neglect against the one you'd be picking up. It just makes noise. And noise, to a saturated practitioner, is not help. It is one more thing to manage.
The deeper problem is that the failures aren't random. They're produced by conditions the lawyer doesn't control. Court listings get assigned without consultation. Registries process at their own pace. Adjournments cascade through diaries. Opposing counsel files at 4:55pm on a Friday. Counsel-on-record concentrates liability on one person who cannot physically be in two places at once. Asking her to "manage her time better" is asking her to solve a problem someone else is creating.
This is why the memory framing has staying power despite being wrong: it locates the failure inside the lawyer, where it can be fixed by a product. Naming it as structural is harder, because the structure isn't going anywhere.
But naming it correctly is the precondition for building anything that actually helps.
Planning works differently when the environment is unpredictable
In stable environments, planning is decomposition, sequencing, allocation, tracking. Legal practice doesn't work that way. The operating environment is the dominant source of variance. So planning has to be a discipline of working with stochasticity, not against it. A few things shift when you adopt this frame.
Planning becomes capacity-first rather than deadline-first. The traditional view starts with what's due and works backward. The capacity-first view starts with what time is available — given court attendance, travel, the realistic limits of focused drafting — and asks how much load can fit inside it. Most overcommitted lawyers have never done this calculation. They accept matters because the work is there, and only discover the overcommitment once it's already breaking.
Triage becomes a discipline, not a panic response. When everything cannot be done, the question is which things matter most — and the answer is usually not "whichever is loudest." A planning practice involves explicit, advance triage rules: when X and Y collide, Y wins. Most lawyers triage in the moment, under pressure, and they triage badly because the cost of the wrong call isn't visible until weeks later.
Foresight horizons stretch. The default planning horizon for most practitioners is the week. The 60- and 90-day views, where strategic decisions actually live, are almost universally absent. This isn't a failure of discipline — it's a failure of tooling. Calendar applications don't show you the load curve sixty days out. They certainly don't show you that taking on a new matter today will collide with three things in November.
Slack stops being laziness and becomes infrastructure. The most counterintuitive lesson from operating in a stochastic system is that planning to 100% capacity is planning to fail. The lawyers who run sustainable practices are the ones who refuse to fill their diaries. This is hard to internalise in a profession that equates busyness with productivity.
What I'm describing isn't a methodology. It's closer to a posture. Planning, in this environment, is the discipline of seeing your own practice clearly enough to make load-bearing decisions before they become emergencies. Most legal tech does not help with this. It helps you remember things you already knew you had to do.
Why most planning systems fall apart
Here is where every article on planning usually ends — with the prescription. As if naming the practice were sufficient.
It isn't. The harder problem is that planning systems decay. Lawyers try them. They build elaborate Notion workspaces, adopt Trello boards, configure Clio's task module. And six weeks later, the system is in partial abandonment: half the matters logged, the others forgotten; the dashboard showing data from three weeks ago. This is so universal it's almost a law of legal-tech adoption.
Every planning system depends on input, and the cost of capture is paid up front, in small daily increments, by a practitioner who is already overloaded. The benefits are diffuse and arrive later. The asymmetry is brutal, and the lawyer-rational response is to skip capture during the bad weeks — which are precisely the weeks when capture matters most.
Then there is the bad-week problem. Every planning system gets one chaotic week early in its life. A trial runs over. A client emergency consumes three days. The system goes stale. Now logging in feels like archaeological work, and the cost of catching up is higher than the cost of starting over. So she starts over. And the next chaotic week, the same thing happens.
If we don't take these failure modes seriously, every planning intervention is just another well-intentioned thing that will be abandoned. The question isn't "how do we get lawyers to plan?" It's "how do we build planning practices that can survive the conditions lawyers actually work in?"
What actually survives a chaotic week
Some principles, drawn from watching what actually holds up.
Co-locate capture with the work itself. The systems that get used are the ones where logging happens during the work, not after. Anything that requires "I'll update the system later" loses, because there is no later.
Make capacity visible, not just commitments. A diary showing what's booked is a memory aid. A diary showing how many committable hours remain in the next sixty days is a planning aid. Almost no legal tech product does the second well, which is a category-level opportunity sitting in plain sight.
Plan at the right horizon. The week is too short. The year is too long. Strategic decisions in legal practice live at 30–60–90 days, which is roughly where you can still refuse a matter, brief it out, or restructure your team. Planning practices that anchor at this horizon outperform daily and weekly rituals because they match the time scale at which the actual decisions get made.
Build for repair, not perfection. A system that assumes you'll always log in real time will fail. A system with explicit catch-up affordances — bulk import, "matter bankruptcy," flag-for-review — is one a real practitioner can live with. Design for the worst week, not the best one, and the system has a chance.
Run the bad-week test. Any habit you adopt should be evaluated by one question: does this survive a bad week? If it requires twenty minutes of daily curation to remain useful, it will die the first week you don't have twenty minutes.
What this changes for legal tech and KM
For those of us building, advising, or studying this space, the planning frame opens territory the memory frame cannot reach.
The product category that should exist — and mostly doesn't — is decision-support tooling for practice operations. Not "what's due today" but "where is the system going to break next month, and what should you do about it now." The leading indicators of failure in a practice are visible weeks before the failure itself, but only if someone is looking. Most tools aren't built to surface them.
The metric that should replace "deadlines met" is something closer to "matters accepted versus capacity to deliver" — measured before the matter is taken on, not after it's botched. We don't currently measure either, which is why neither shape gets rewarded.
And the role of paralegals deserves a complete reframe. In the planning view, paralegals are not task-doers under direction — they are the firm's operational nervous system, the people closest to the registries, the courts, the clients. A paralegal who can flag "you have three things landing on the 14th" is more valuable than one who only does what she's told. In African and Asian small-firm contexts where paralegals are operationally central, this is one of the highest-leverage areas in the field.
A more demanding mandate
I went into legal tech research expecting to find a memory problem. I came out with something harder and more interesting: a planning problem, dressed up as a memory problem because the memory framing was easier to sell.
The advocate I sat with that afternoon eventually closed her laptop and said something I've thought about often since. She said: "I'm not behind. I'm overcommitted. There's a difference."
There is a difference, and the difference is the whole field. A lawyer who is behind needs to remember more, work harder, catch up. A lawyer who is overcommitted needs to refuse more, plan further out, build slack, allocate differently. The interventions are not just different — they're opposites. And we have spent two decades building tools for the first lawyer when most of them are the second.
The work in front of us — for those of us who build, advise, write, and teach in this space — is to stop selling memory to people whose memory is fine. It is to take seriously that the practitioners we serve are not failing inside their work. The work is failing them. And it is in our power, slowly and with care, to build something that meets them where they actually are.
That is a more demanding mandate than we have been operating under. It is also a more hopeful one.
Comments (0)