About The Contract Signal

This newsletter focuses on post-executed enterprise contract intelligence: how organizations turn already-signed contracts into structured data, searchable knowledge, useful reports, trusted AI workflows, and better legal and business decisions.

The emphasis is practical - enterprise contracts are complex and messy. The buyer is often not the user. Contract data is often fragmented or unreliable.

The lens I bring

I’m not a practicing attorney.

But I spent 10 years focused on post-executed enterprise contract intelligence at Axiom and Knowable (a LexisNexis company), working across product, user research, taxonomy design, solution sales, contract analysis and data creation, traditional machine learning, and generative AI.

That gave me a cross-functional view of the market that is hard to get from the outside.

I’ve done direct end-user research with the people who actually work with contracts — not just buyer conversations that feed sales decks, but granular conversations about workflows, friction, mental models, and what happens when a product promise meets real day-to-day work.

I’ve helped build contract AI products, from earlier machine learning systems involving annotation, training data, and evaluation workflows, to generative AI features integrated into production software. That experience shaped one of my core beliefs: the model is rarely the whole product. The hard part is the surrounding system — the data strategy, evaluation framework, workflow design, infrastructure, and UX that make AI output reliable and usable.

I’ve also experienced the commercial side: proposals, RFPs, change orders, customer commitments, demos, proof-of-concepts, and production deployments. That matters because the gap between what is pitched and what is delivered is often where the real story lives.

And I’ve gone through contracts workflows myself: organizing repositories, finding agreements, extracting clause and data point information, testing different approaches, and seeing what worked and what didn’t. Not as in-house counsel, and not formally as legal ops, but close enough to understand the practical texture of the problem.

Before all of that, I earned a JD from Emory, focused on transactional law taught by real practitioners. I don’t practice law, but that training gave me the legal vocabulary and analytical grounding to recognize when legaltech is solving a real problem — and when it is missing the point.

Why that matters

Most analysis of contracts AI comes from one lens at a time.

Some writers understand legal practice but haven’t built AI products. Some technologists understand models but not contract workflows. Some investors and founders understand the market but have incentives that shape what they emphasize. Some analysts can summarize the landscape but haven’t lived through the product, delivery, and user-research realities underneath it.

Each lens is useful. None is complete.

The Contract Signal is my attempt to write from the overlap: legal domain understanding, contract workflow exposure, product-building scar tissue, AI implementation experience, and an independent view of the market.

I have no vendor to sell, no portfolio to protect, and no employer to promote.

What I’ve seen

Over the last decade, I’ve heard smart, well-resourced people repeatedly underestimate the complexity of enterprise contracts.

In 2020, credible people at a major technology company predicted they could capture the enterprise contracts market within months. In 2021, a client’s internal data science team believed they could automate much of what our business did in a matter of weeks. Neither prediction came close.

That history is not a reason to be pessimistic about AI. AI can already do a lot in contracts, and the capabilities are improving quickly.

But it is a reason to be precise.

Enterprise contracts are messy. The data is fragmented. The workflows are often organization-specific. The buyer is often not the user. The demo is not the deployment. And a model capability is not the same thing as a reliable product.

That is the gap this newsletter is focused on.

What to expect

The goal is tight, non-technical, and high-signal: clear explanations of what contracts AI tools claim, what they actually appear to do, where the market is moving, and what matters for people working with contracts in practice.

The analysis will prove itself over time, or it won’t. That’s the right bar.

If you work with contracts, buy or evaluate AI tools for legal workflows, build in this market, or want an independent read on where contracts AI is going, subscribe to follow along.

No ads. No sponsored content. No vendor relationships to protect.