Consider this: it’s March 2020, and the COVID-19 pandemic has just begun. Not exactly a favorite memory, we know. But hang in there - we promise this is going somewhere worthwhile. Clients needed urgent advice on force majeure clauses, business interruption insurance, employment law changes and cross-border regulatory obligations. All at once, across multiple jurisdictions.
Some firms had answers within hours. Others were still scrambling weeks later. The difference? Easy access to the lessons learned during the H1N1 outbreak a decade earlier. The firms that could tap into that knowledge started advising immediately. Those who could not were forced to start from scratch.
That experience highlights a challenge many law firms still face today. Most are sitting on more knowledge than they realize, but almost none of it is easy to find. It lives in closed matter files, in email threads and in the heads of lawyers who don’t work there any longer. In this article, we explore why that happens, what it costs and how to fix it. Let’s get into it.
Legal knowledge is an unusual asset. Every matter generates more of it: precedents, regulatory expertise, client intelligence and hard-won judgment. The problem is that it is deeply tied to the people and matters that created it, which makes it surprisingly hard to hold on to.
Three challenges come up consistently:
Without a proper system, pulling regulatory guidance across multiple jurisdictions can take two to three days. With a connected, queryable one, the same task takes seconds and with cited sources, you can verify.
It is easy to talk about knowledge loss in the abstract. But what does it actually look like in a law firm?
Think about how many difficult questions your firm has answered over the years. Every matter adds another piece of analysis, another negotiated position, another lesson learned. Multiply that across hundreds or thousands of matters, and you start to see the scale of what firms are sitting on.
This one is even less visible because it frequently doesn’t get written down at all. Ask a senior partner what position the firm takes on indemnity caps in SaaS contracts and they will tell you without hesitation. Finding the written version can take considerably longer.
The one that really stings when someone leaves. Much of the client knowledge never makes it into the matter file: who makes decisions, what matters to the client and the context behind each conversation. That intelligence, built up over years, disappears with the person who held it.
In compliance-heavy practice areas such as GDPR, employment law, financial services regulation and tax, regulatory tracking knowledge is often built from scratch. Without a way to capture and reuse knowledge in a reliable way, lawyers end up researching issues their colleagues have already answered.
Junior lawyers often learn by asking the same questions their colleagues answered years ago. Without access to previous work and firm expertise, development becomes slower and more dependent on who happens to be available to help.
Once you understand what is being lost, the next question is: why is it so difficult to solve? These are the five obstacles that law firms usually encounter when they try to do something about it.
This is where off-the-shelf tools fall short. Law firms need permission controls at the matter, client and practice-group level. And they need to work in practice, not just on paper. Any vendor should be able to demonstrate this in a realistic scenario.
The corporate team and the employment team might share a client, but they rarely share knowledge. Expertise ends up scattered across systems, folders and teams, making it difficult for the wider firm to benefit from it.
Some of the best legal knowledge never makes it beyond the inbox. Partner feedback, regulatory analysis and negotiation strategy often remain buried in email threads, making that expertise difficult to find and reuse.
The logical moment to capture knowledge is when a matter closes. Unfortunately, billing pressure makes this the least favorable moment to ask lawyers to do anything non-billable. They have already moved on mentally to the next matter. Knowledge capture feels like overhead.
Measuring the return on knowledge management is not always straightforward. Many firms focus on technology first, but without a clear knowledge strategy, even the best tools can struggle to deliver measurable results.
The legal industry has been thinking about this problem for a long time. What has actually changed recently is the technology available to address it. Here is how that evolution has played out:
Large firms have long employed PSLs (professional support lawyers) to maintain precedent banks and curate practice guides. It works well, up to a point. But it is expensive, hard to scale and dependent on the PSL knowing what they need to know. The best knowledge of the matter rarely makes it to them because the fee earners generating it are already on to the next thing.
iManage, NetDocuments and similar systems have become standard at most mid-size firms. They are good at storing documents and retrieving them by metadata. What they cannot do is answer a question. Type in "what is our standard position on limitation of liability in technology contracts?" and you get a list of documents, not an answer. You still have to open them all and figure it out yourself.
The real shift is not from documents to AI. It is from storing knowledge to making it accessible.
This is where things are moving now. LexisNexis research shows that 61% of UK lawyers already use AI tools in some capacity as part of their daily work. The meaningful shift in 2026 is using AI not just as a drafting aid but as the interface for knowledge itself. Ask a question in plain language, get an answer with cited sources you can verify.
The critical word there is cited. An AI tool that gives you a confident answer with no source trail is not suitable for legal use. You need to be able to verify every piece of knowledge surfaced, trace it back to the document it came from and be confident that what you are reading is current. That is a higher bar than most general AI tools meet.
One more thing worth noting: the enterprise tools you see getting the most press coverage (Harvey, Legora, Luminance) are designed for Am Law 100 firms with dedicated KM teams and six-figure implementation budgets. Most mid-sized firms need something more practical: the same confidentiality and citation standards, without requiring a specialist team to run it.
There is a lot of software in this space, and most of it was not built with legal practice in mind. If you want a deeper look at the options, our guide to legal knowledge management software is a good place to start. In the meantime, here are the five things that genuinely matter and the questions worth asking any vendor before you sign.
That last point deserves a moment. Data protection requirements should be part of any knowledge management evaluation. Depending on the jurisdictions you operate in and the clients you serve, that may include EU GDPR, UK GDPR or both. Your vendor should be able to explain exactly how their platform supports those requirements, without hesitation.
The firms that fail at knowledge management usually fail in the same way: they try to do everything at once, the project becomes overwhelming and it slowly dies.
Here is a more practical approach:
Start with the practice group where knowledge loss is costing you most visibly: usually the highest-revenue area, or the one that keeps rebuilding the same regulatory research. Get that working well before you even think about expanding. A small, well-used knowledge system is worth more than a grand one that nobody trusts.
The fastest win is not asking lawyers to produce new documentation, but making the knowledge that already exists findable. Connect your document management system, email and communication tools to a searchable layer first. That alone will immediately cut duplicated research.
A structured 30-minute exercise at matter close (key decisions, negotiated positions, documents worth preserving, lessons learned) accumulates into something significant over time. The format needs to be simple and consistent.
The associate who needs to understand the firm's standard position on a clause type, or who is researching a regulatory question that three seniors have already answered, is the person who will get the most value from your knowledge system. Design for their questions and their workflow.
Knowledge that belongs to everyone tends to be maintained by nobody. Assign named ownership of knowledge areas to practiced group leaders. The firms making the most progress right now are the ones that appointed a senior knowledge leader to build a strategy before they invested in any technology.
The next regulatory shock will not announce itself in advance. Whether it arrives as a new AI regulation, a major data protection reform or an unexpected geopolitical event, firms will face the same question they faced in March 2020: do we already know more than we think we do?
The firms that answer fastest will not necessarily be the ones with the most lawyers or the newest technology. They will be the ones who can find, trust and reuse the knowledge they already have.
That is the difference between knowing something and being able to find it when it matters. AI may be the interface, but the real advantage comes from the quality of the knowledge underneath it. If you are ready to build a knowledge base your lawyers can actually use, see how we help legal and compliance teams do exactly that.
Three reasons: matter knowledge is rarely captured when matters close, lawyer mobility means expertise leaves with people, and confidentiality requirements make knowledge harder to share than in most organizations.
Client relationship knowledge is often the most painful to lose. Who makes decisions? What does the client care about? What is the context behind past advice? Firms also lose undocumented precedent knowledge, regulatory expertise and institutional know-how.
At a minimum: matter-level permission controls, source citation, audit trails, integration with existing document management systems and support for relevant data protection requirements.
Large firms often have dedicated KM teams and enterprise budgets. Boutique and mid-sized firms need the same standards for confidentiality and citation, but without the cost, complexity or need for specialist resources.
Your hub for research, frameworks, and practical guidance on workforce transition, organizational knowledge, AI adoption and more.
