Competing where it matters, Standardising where it doesn’t
[see also part 5 of this series]
The previous articles argued that the industry’s challenge is no longer document management. We have become remarkably good at creating, exchanging, storing, filing, and retrieving artefacts. Yet much of the effort associated with quality control, inspection readiness, and compliance still depends on human interpretation. Ultimately, reviewers are not interested in artefacts themselves. They are interested in what those artefacts prove.
If that observation is correct, an obvious question emerges. Is this a problem that every sponsor, CRO, site, and technology vendor should solve independently?
The banking industry faced a remarkably similar question after the introduction of SWIFT. By the late 1980s, the communication problem had largely been solved. Banks could exchange payment instructions electronically and reliably. Message standards existed. The network existed. Connectivity was no longer the primary challenge.
Yet substantial manual processing remained.
The reason was straightforward. Different institutions continued to interpret messages differently. Different institutions imposed different operational conventions. Different institutions defined different requirements for information quality and completeness. A payment instruction could travel electronically across the world and still require significant human intervention before it could be processed.
Connectivity had improved. Consistency had not.
During the 1990s, SWIFT launched the Straight-Through Processing initiative. Having participated in that effort, one observation still stands out. The biggest challenge was not technology. The biggest challenge was convincing participants that some problems should no longer be competitive.
Banks compete on customers, products, service quality, innovation, market reach, and operational excellence. They do not compete on the structure of a payment message. They do not compete on how an account number is represented. They do not compete on how a settlement date should be expressed. Those questions were gradually removed from the competitive arena because everybody benefited from having a common answer.
The result was not less competition. The result was better competition.
Banks could focus on serving customers and developing products rather than repeatedly solving the same infrastructure problems in slightly different ways. Standardisation did not weaken the industry. It strengthened it.
When I look at clinical research today, I sometimes wonder whether we are approaching a similar moment.
Sponsors define their own quality expectations. CROs establish their own review procedures. Technology vendors build proprietary compliance logic. Sites learn different conventions depending on which sponsor they happen to support. Some variation is inevitable and even desirable. Different organisations have different operating models, therapeutic focus areas, and risk appetites.
Yet I increasingly struggle to see where the industry derives value from maintaining different interpretations of common compliance questions.
Does a delegation log demonstrate appropriate authority? Does a training record demonstrate appropriate qualification? Does an informed consent form demonstrate valid consent? Does a temperature log demonstrate acceptable storage conditions?
These questions are asked thousands of times every day across sponsors, CROs, sites, auditors, and regulators. They are often answered by highly skilled professionals applying broadly similar reasoning to broadly similar evidence. The effort is repeated continuously across the ecosystem.
The question is not whether organisations should continue competing. They absolutely should.
The question is whether these particular problems belong in the competitive domain at all.
Over the last decades, the industry has invested heavily in systems of record, document repositories, workflow platforms, and quality processes. Those investments were necessary and remain important. However, they have not fundamentally changed the fact that different participants continue to evaluate similar evidence independently and often according to slightly different interpretations.
Straight-Through Compliance, if such a concept is ever to emerge, will not be created by technology alone. Technology can help extract evidence, maintain provenance, execute rules, and support human review. Those capabilities are increasingly available. The harder challenge is achieving sufficient agreement on how common evidence should be evaluated.
This is ultimately not a vendor question. It is not a platform question. It is not even a technology question.
It is an industry question.
The question is no longer whether Straight-Through Compliance is technically possible. The more interesting question is whether enough participants are willing to converge around common evaluation practices to make it meaningful.
If the answer is no, the industry will continue building proprietary review processes, proprietary compliance logic, and proprietary interpretations of largely common questions.
If the answer is yes, the industry has an opportunity to create something similar to what SWIFT created for banking: a shared layer of trust, consistency, and interoperability that benefits all participants while leaving competition exactly where it belongs.
Based on the patterns I have observed, I believe that possibility is worth exploring. Not because it benefits a particular sponsor, CRO, site, or technology vendor. But because some challenges are simply too fundamental to solve independently.
Perhaps the next stage of maturity for clinical research is not another repository, another workflow platform, or another generation of document management. Perhaps it is deciding, collectively, which compliance problems should remain competitive and which should become common infrastructure.
That is a conversation worth having.
