Most organizations have critical relationships—vendor contracts, partnership agreements, investment portfolios—buried in PDFs, spreadsheets, and tribal knowledge. The intelligence that should drive decisions is hidden, and nobody has time to extract, normalize, and compare it against actual business outcomes. That's the problem we built against. At Domo, we're engineering agentic workflows that begin where the data actually lives: unstructured documents, scanned PDFs, and disparate systems. The agent reads the document, extracts what matters, normalizes it against a common schema, and performs the analysis your analysts would do by hand in seconds, across every record simultaneously. The result is a living intelligence layer that benchmarks every investment against the portfolio, surfaces underperformers before the window to act closes, and routes the decision to the right person with the rationale already written. Featured Session: The Agent That Turns Any Complex Agreement Into a Managed Business Decision Join Domo CMO Mark Boothe and Paul McCusker, Forward Deployed Engineer at Domo, as they walk through a purpose-built agentic application for a document-heavy investment management use case and explain the architecture that applies across industries. The demo uses athlete licensing contracts for a global sports merchandise leader, but the architecture works for vendor contracts, real estate leases, insurance policies, or any domain where decisions live inside documents your systems can't read. What You’ll See: • Document Intelligence at scale: The agent reads PDFs, scans, and handwritten terms, extracts what matters, normalizes into a common structure, and makes them immediately queryable. • Portfolio benchmarking: Every agreement is scored against every other—a ranked priority list of which relationships generate returns above cost, which are underwater, and which are approaching a decision window your team is about to miss. • Action routing with rationale: The agent writes the business case: what the data shows, the recommended action, and what changes if you wait. Every decision arrives in an approval queue with the work already done. • Human-in-the-loop governance: No autonomous renegotiations, no autonomous commitments. The agent recommends; the decision-maker approves. Governance is built into the architecture. • A conversation with your portfolio: Executives and analysts can ask natural-language questions grounded in actual contract and performance data. These are grounded answers with the numbers behind them. Paul will show you what it looks like when an agent handles extraction, normalization, benchmarking, and triage, and your team focuses entirely on the decisions.

 0      1