Most Privacy-Enhancing Technologies (PETs) projects don’t fail because the math is flawed. They fail because they get stuck answering a deceptively simple question from the business:
“So what?”
When the pitch for PETs begins and ends with “privacy,” it often fails to gain traction. This isn’t because privacy is unimportant, but because most organizations already have established methods for managing it: legal agreements, data minimization, de-identification, and trusted third-party intermediaries.
This creates a trap. PETs are perceived as a more complex version of an existing workaround, rather than what they truly are: a vehicle for unlocking business outcomes that are otherwise blocked by risk, regulation, or competitive friction.
To move PETs from a research curiosity to a production priority, the conversation must start where the business lives: value.
In a recent webinar on the use of PETs in real-world projects, Antigoni Polychroniadou of J.P.Morgan made a critical point: when teams already have a solution like a trusted third party, a pitch centered on “privacy” is insufficient to justify change. The business will naturally respond with a list of legitimate concerns:
She offered a perfect example: banks pooling “bad beneficiary” lists to combat fraud. In theory, PETs allow these banks to find overlaps without revealing their proprietary lists. In practice, many already use a third-party vendor for this function.
So, what is the compelling argument for PETs?
It isn’t “better privacy.” It’s better outcomes.
Polychroniadou’s point is that if banks don’t fully trust the intermediary, they may be holding back their most sensitive or recent data. With the mathematical guarantees of PETs, they can contribute more complete datasets with greater confidence. This improves the quality of the shared intelligence and, ultimately, leads to more effective fraud detection for everyone.
The pitch transforms from risk mitigation to value creation: PETs increase participation and improve data quality because trust is no longer the limiting factor.
A clear way to structure this value-first approach is to determine whether PETs will serve as an enabler for a new process or a displacer of an existing one.
Here, PETs act as an Enabler, making a new type of collaboration possible. The business case should quantify what new capability is being unlocked.
Key questions to answer:
A prime example is a secure auction or matching system where participants will not join if their sensitive intent is exposed. PETs enable the market to exist in the first place.
Here, PETs act as a Displacer, offering a superior alternative. The business case must demonstrate that the existing solution is leaving value, security, or efficiency on the table.
Key questions to answer:
In both scenarios, privacy is an essential feature, but the engine driving the project forward is tangible business value.
When PETs projects gain internal traction, they almost always map to one of these core value propositions:
Successful project proposals typically focus on one primary and one secondary value driver. Pitching all five at once can dilute the message and appear unfocused.
One of the most compelling PETs projects in financial services focuses on cross-bank credit risk modeling, an area where collaboration is valuable but traditional data sharing is effectively impossible.
Banks, lenders, and financial institutions each hold partial insight into credit exposure, default patterns, and early warning indicators. During periods of economic volatility, this fragmentation leads to delayed responses, conservative modeling assumptions, and inefficient capital allocation.
In theory, collaborative modeling would significantly improve outcomes. In practice, sharing loan-level or counterparty data across institutions introduces unacceptable regulatory, commercial, and reputational risk.
Today, most institutions address this challenge through:
While compliant, these approaches degrade value. Data is summarized, delayed, or selectively contributed, resulting in models that lack precision and timeliness.
From a business perspective, this is a classic example of a PETs project opportunity hidden inside an apparently “solved” problem.
With Privacy-Enhancing Technologies (PETs), financial institutions can jointly train credit risk or stress-testing models without exposing raw data, portfolios, or customer information.
Each participant computes locally on its own sensitive data. Only cryptographically protected outputs are shared for collaborative analysis. No centralized dataset is created, and no party gains visibility into another’s inputs.
The result is a fundamentally different trust model.
This PETs project delivers measurable business outcomes:
Because participants no longer need to rely on contractual trust alone, they are willing to contribute more complete and timely data. The quality of collaboration improves, and so do the results.
This use case highlights a recurring pattern across successful PETs projects:
They do not win by promising “better privacy.”
They win by enabling better decisions, faster collaboration, and higher-quality outcomes.
Privacy-Enhancing Technologies are a mechanism for unlocking business value that existing trust models suppress.
PETs projects fail when they are framed as a privacy upgrade, a security initiative, or a research experiment.
They succeed when they are championed as a revenue enabler, an ecosystem unlock, or a strategy for making new and valuable collaborations possible.
The fastest path to production isn’t teaching the organization about cryptography. It’s showing them what they stand to gain. Once the business value is undeniable, the organization will find the time, budget, and patience required for the technology.
That is how PETs projects move from theory to reality.
