The call with Aaron Reiss came back in January, unsolicited.
"Your Q4 announcement," Reiss said, without preamble. He had read the brief press release Marcus had put out about the new features and the closed contracts. "The continuous monitoring capability."
"Yes."
"That's not a journalism feature."
"No. It's a compliance feature."
"If it can monitor government procurement data in real time for anomalies, it can monitor corporate supply chain data in real time for anomalies." A pause. "Or for changes."
"Changes are a different problem from anomalies," Marcus said. "Anomalies require a baseline model and a deviation threshold. Changes just require a delta. The delta problem is technically simpler."
"How much simpler?"
"Considerably."
"I want to revisit the conversation we had six months ago," Reiss said. "I have a more specific proposal."
"Go ahead."
"There's a category of equity strategy built around supply chain disruption risk. The idea is that companies with opaque supply chain relationships are exposed to disruption events that aren't priced by the market because the market can't see them. If you could identify, in advance, structural anomalies in corporate supply chain data that historically correlate with disruption events—"
"You're describing a predictive model," Marcus said. "Not a monitoring tool."
"A monitoring tool that feeds a predictive model."
"The predictive component requires a training set. Historical supply chain data correlated with disruption events. Do you have that data?"
"We have four years of it. Proprietary, acquired from a data vendor. Fifty thousand companies, quarterly updates."
Marcus thought about the architecture. A monitoring layer over real-time supply chain data, feeding into a predictive model trained on historical disruption correlations. The methodology was a natural extension of the Threadline approach. The data science was non-trivial but not beyond what Amir and Yuki could manage.
"Revenue share structure," Marcus said. "What are you proposing specifically?"
"Twelve percent of net revenue from affected strategies, uncapped. Plus a data licensing fee for the historical dataset — we'll provide it exclusively to you for twelve months."
"Fifteen percent, uncapped. Exclusive data access for eighteen months. And I want a right of first refusal on any subsequent data licensing in this space."
A brief pause. "Thirteen percent. Eighteen months, agreed. Right of first refusal, agreed."
"Fourteen," Marcus said. "And I need six months to build the monitoring component before we connect it to your predictive model."
"Six months is fine." Another pause. "Fourteen, agreed."
"Send a term sheet to Marsh."
"Her office will have it by end of week."
Marcus hung up and looked at the Threadline revenue model on his screen. He added a new line: *Supply Chain Intelligence — Year 2 projection.* He put a range in and looked at the number.
The number was large.
---
He brought it to the full team the next morning. He described the Reiss partnership without naming Reiss or his fund — those details would come later, when the term sheet was reviewed and signed. He laid out the technical requirements and the timeline.
Yuki was the first to speak after he finished.
"The supply chain data," she said. "Corporate supply chain relationships are commercially sensitive. If we're ingesting it and our infrastructure is compromised—"
"The security model for this product line will be separate from Threadline's current architecture," Marcus said. "Isolated data environment, stricter access controls, separate audit logging."
"You've already designed it."
"Broad strokes. I'll need your review before we build."
Yuki looked at him. "You knew I'd raise the security question."
"I knew you'd raise the right question," Marcus said. "That's why I want your review."
Something shifted in her expression — not quite a smile. She wrote something in her notebook.
Priya was looking at the revenue model. "This is a different customer," she said. "A hedge fund is not a newsroom and it's not a compliance team. The sales motion is different, the relationship is different, the risk profile is different."
"All true," Marcus said. "Thoughts?"
"I think we can manage it. But I want a separate product identity for the supply chain intelligence line. Different brand surface, different contract templates. We don't want the Monitor stories and the hedge fund on the same page."
"Agreed."
Jin was looking at the technical spec. "The monitoring component — what's the data ingestion source?"
"Multiple. Corporate procurement disclosures, supplier diversity filings, logistics network data from public and licensed sources." Marcus looked at Jin. "It's architecturally similar to the Threadline ingestion layer but with a higher volume profile. You'll need Amir."
"I'll talk to Amir."
Marcus looked around the table. The team was thinking, engaging, raising the right concerns without flinching from the opportunity. He thought about the Fourth Gate.
*The architecture of trust.*
He thought he was building it. He wasn't sure he was building it fast enough.