A logistics lead at a global shipping firm recently discovered that an autonomous procurement system had renegotiated three supplier contracts before the legal department even realized the renewals were due. The system didn’t just flag the dates; it cross-referenced global shipping rates against the supplier's SLA performance and executed the most favorable terms available in the market. While the lawyers were looking for the folders, the data was generating margin.
Most executives still treat contracts as a necessary legal burden—a "filing cabinet" problem. This is a mistake of economic logic. True to the definition of an asset, a contract is a resource with potential future economic benefit. Yet, in most organizations, these assets remain buried in SharePoint folders or 100-page PDFs, creating what is perhaps better understood as "data debt".
The $3 Trillion Black Hole
The scale of this neglected topography is staggering. McKinsey estimates that $3 trillion is wasted annually in B2B transactions because decision-makers lack access to contract analytics. This isn't a failure of organization; it's a failure of data liquidity.
Oliver Hart, the Harvard economist and Nobel laureate, has long argued that the economic boundaries of the firm are defined by the control of assets. In his view of "incomplete contracts," the inability to specify every future contingency makes the control of the underlying contract lifecycle management (CLM) data paramount. If an executive cannot see the terms, do they truly control the asset?
"We see $100 million contracts being managed in a spreadsheet," observes Brandon Card, CEO of Terzo, a platform focused on the financial externalization of contract data. "While legacy CLM platforms like DocuSign or Icertis are useful at a high level for e-signatures and templating, they do not show SKU-level pricing data that a CFO needs to drive cost optimization".
Mind the Value Leakage Gap
This disconnects results in what procurement experts call "contract value leakage"—the erosion of value that occurs between the time a deal is signed and when it is actually executed. It is the gap between negotiated value and realized value. When a company cannot track its obligations or its suppliers' deliverables in real time, the margin established during negotiations simply evaporates.
The ownership of the document creates a certain tension. Although the word "contract" evokes the legal department, business people negotiate a majority of the terms. These are financial documents, yet they are rarely treated with the same rigor as a P&L statement. By shifting the focus from legal risk to financial utility, organizations can begin to recapture this leaked capital.
“When people hear the word 'contract,’ they think about lawyers,” Card observed. “They should be thinking about dollars. These are financial assets, and they need to be treated like one.”
Human Remains
For years, companies relied on manual labor to bridge this gap. A massive organization might employ hundreds of people in an offshore center just to review contracts. When a company acquires at the pace of every six weeks, the resulting data debt is anthropogenic—a byproduct of siloed systems. By interrogating across these silos, one Fortune 100 tech company claims it saved $100 million in one year on a single category, proving that acquisition speed need not come at the expense of fiscal discipline.
"We don’t believe you need fourteen stakeholders involved in a renewal,” added Card. “By building agents atop a structured data layer, the renegotiation process can be automated almost entirely.”
Legacy competitors like SAP Ariba or Coupa offer broad source-to-pay automation, but can struggle with the in-situ extraction of complex metadata from existing legacy documents. They are built more so to manage the process than analyze the asset. Newer entrants like Sirion or Kira Systems focus heavily on legal redlining, which is useful for the general counsel but provides little utility for a CFO trying to correlate service-level agreement (SLA) data with historical transactions.
The Destination Problem
Traditional optical character recognition (OCR) tends to fail when faced with the idiosyncratic topography of global legal terms. This requires a shift toward small expert models. Unlike general-purpose LLMs, these small-compute models are trained specifically on financial and legal parameters. They don’t merely summarize; they structure the data.
At Home Depot, 600 employees log into a contract intelligence platform daily. And because the data is reusable and progenitive, the more it is used for renewals and supply chain audits, the more accurate the enterprise’s risk score becomes. As these systems reach agentic maturity, organizations will no longer need a $50 million engagements large accounting firms to perform a post-merger audit when an autonomous agent can do the same work for far less. And faster.
Liquidated Assets
As Charles Fox, author of Working with Contracts, observes, a contract is fundamentally a record of a party’s business, assets, and financial condition. However, Fox notes that materiality only becomes clear when there are reference points to receive new information. Without structured data, the "materiality" of a contract remains a ghost in the machine.
If we push the autonomous “self-driving” organization concept to its conclusion, we see a future where agents talk to agents. For example, a contract stipulates a discount if a supplier fails a delivery and the autonomous system detects the failure, calculates the credit, and adjusts the payment without human intervention.
This also requires a purge of data junk. Organizations often find that 20% or more of their document volume consists of duplicates of files accidentally uploaded to the cloud. Before a company can generate data value, it must first navigate the scavenger hunt. The era of the filing cabinet is all but dead. In this information economy, the organizations that access data fastest and use it most broadly are the ones that are thriving.
Full Article: https://www.forbes.com/sites/douglaslaney/2026/03/12/contracts-as-capital-recovering-3-trillion-in-data-debt/
