Context beats the model: StructureFlow on where legal AI is heading
StructureFlow's Tim Follett and Neal Chatrath with Tim Kenney at ILTACON 2026.
"There had to be a better way to do this than using PowerPoint."
That frustration is where StructureFlow began. At ILTACON 2026, Tim Kenney sat down with founder and CEO Tim Follett and president Neal Chatrath to talk about structural intelligence, legal data and the company's growth in North America.
The big shift they described is one every legal technology leader should be watching: the AI conversation is moving away from the models and towards the data you feed them.
1. Build from a problem you've lived
Before StructureFlow, Tim was a corporate lawyer working on private equity fund structuring, building structure charts in PowerPoint. The product grew straight out of that work. It's an AI-enabled platform that takes linked data about entities, documents, assets and people, and generates structure charts, flow charts and timelines automatically.
The lesson: the most useful legal tech is often built by someone who has done the work and felt exactly where it breaks.
2. Hire for your client's world, not just the law firm's
Neal's background is unusual at a legal tech conference. He comes from fintech, working with hedge funds, fund service providers and investment banks. He joined StructureFlow as president in August, its first senior hire in North America, based in New York, where most of the company's revenue already comes from.
His brief is to deepen relationships with law firms and connect with their clients. A month in, he is already seeing those clients put a pointed question to their lawyers: "How are you doing this?"
The lesson: clients are starting to ask about method, not just outcome. Vendors who understand the client's side can help firms answer that question well.
3. Context is overtaking the model
The headline takeaway from the conversation: the AI debate is shifting from which model is best to what context you give it. "Ontology" was a word Tim kept hearing at ILTACON. Structured, mapped data is something both people and machines can navigate, and it gives AI the context it needs to do valuable work.
The lesson: before reaching for a bigger model, check whether your data is organised well enough for any model to use.
4. Diagramming was the start, not the destination
StructureFlow's next step is a graph-based data layer that can connect to document repositories and systems of record. The diagrams are what lawyers see first. The mapped data underneath is where the company is heading.
What legal leaders should take from this
Expect clients to ask how work gets done, and be ready to show them.
Treat structured data as the foundation for AI, not an afterthought.
Back tools built by people who have done the work themselves.
Watch the data layer. The next round of legal AI value sits underneath the interface.
Watch the full conversation
People in Legal is where legal leaders and legal technology meet. Want your leaders in front of our audience at the next event? Partner with us or email team@peopleinlegal.com.
Quotes lightly edited for clarity.
Comments