By Special Correspondent, Will Seaton at Draftwise.
Missed day one at ILTACon? Across sessions on change management, taxonomy, AI, and document management, a common theme emerged: before firms add more technology, there’s some groundwork to do.
What problem are you solving? Who needs to be involved? Is your firm’s information structured well enough to support what you want to do? And is the tool you’re evaluating actually built for the workflow you have in mind? Here are the key takeaways from day one.
1. Talk openly about the problem you’re solving
In a session on change management, Harriet Joubert-Vaklyes of Michael Best & Friedrich LLP put it simply: ‘You need to start with questions long before you decide that a new tool is necessary.’
A live audience poll found that peer champions were seen as the most effective way to drive adoption, ahead of top-down mandates or training alone. At the same time, many firms acknowledged that change management planning still happens shortly before lawyers are expected to start using a new tool.
The takeaway: Start the adoption work early. Understand the problem, involve the people who will use the technology, and give them a role in shaping the decision.
2. Get your own language straight
Heath Harris of NetDocuments made the case that well-structured data can make AI more efficient, including by reducing the amount of context an LLM needs to consume.
His practical advice was straightforward: agree on what things mean. Is it a stock purchase agreement or a share purchase agreement? Who decides? When designing taxonomies for AI, Harris cautioned that ‘more is not better’.
Ivy Grey of iManage pointed to librarians as an existing source of expertise inside firms: ‘They understand how information is supposed to work together.’
Despite representing different parts of the legal technology ecosystem, the speakers showed notable alignment around the importance of data structure, governance, and reuse. Taxonomy is becoming a much more central part of the AI conversation.
The takeaway: If AI results are inconsistent, it’s worth looking at the taxonomy and governance underneath them.
3. Match the tool to the workflow, not the other way around
Two other day-one sessions pushed back on the idea of picking a single AI platform for the whole firm. ‘Platform Wars: Which AI Tools Do Lawyers Really Want?’ looked at how tool evaluation needs to happen practice group by practice group, since different teams want different things from AI, and a single firm-wide pick rarely satisfies everyone. A companion session, ‘Whose Legal AI Workflow Is It Anyway?’, dug into a related question: which parts of a given workflow are actually suited to agentic AI, and where a person still needs to be in the loop.
The takeaway: Before standardizing on a platform, map out the specific workflows and practice groups it needs to serve. A tool that wins in litigation support may be the wrong fit for transactional work, and treating “which AI tool should we buy” as a single firm-wide question can obscure that.
The bottom line
Day one made a strong case for doing the groundwork early.
Define the problem. Involve the people who will use the technology. Agree on how your firm describes and organizes its information. And match the tool to the actual workflow it needs to support, rather than the other way around.
Those decisions will shape how much value firms ultimately get from the technology they choose.
–
Many thanks to Will – the Chief Customer Officer at Draftwise – and team for providing some great on-the-ground insights into what’s going down at ILTACon this year. Look out for more on Day Two.
—
[ This is a special reportage by Will for AL, who kindly offered to share what’s been happening in Nashville. ]
Discover more from Artificial Lawyer
Subscribe to get the latest posts sent to your email.