Nothing in this article constitutes legal advice. All compliance frameworks described are educational illustrations only. Firms should maintain qualified legal and compliance counsel for all expert network programs.
The intersection of expert networks and securities law is one of the most misunderstood areas of investment operations. Practitioners who work extensively with expert networks often develop a working sense of where the lines fall — but that intuition is rarely tested until something goes wrong.
Most participants understand the general prohibition on trading on material non-public information. Fewer understand exactly where the line falls in the context of an expert call program — particularly when it comes to the mosaic theory, pre-call screening obligations, and what happens when an expert starts disclosing more than they should.
What MNPI Is — and Isn't
MNPI stands for Material Non-Public Information. To be MNPI, information must satisfy two independent criteria simultaneously.
Material means that a reasonable investor would consider the information important in making an investment decision — or that the information would, if disclosed to the market, have a substantial impact on the price of a security. Materiality is not about certainty. A piece of information can be material even if it describes a possible future event, provided the probability and magnitude are sufficient.
Non-public means the information has not yet been disseminated to the general market and is not available from public sources. Regulatory filings, press releases, earnings calls, and widely-cited industry reports are generally considered public. Information shared in a private conversation — regardless of who shared it — is generally not.
The most important misunderstanding to correct: "I heard it from an expert" does not make information public. MNPI obtained through an expert call is still MNPI. The channel of transmission is irrelevant to the legal analysis. What matters is whether the underlying information was publicly available before the call.
What is NOT MNPI: expert opinions and analysis derived entirely from public information; industry knowledge that is widely known and freely discussed among market participants; general market observations that do not reference specific company-level disclosures; and historical context about how an industry operates that the expert could share with any journalist.
“The most common compliance failure I see is not malicious. It's an expert who starts with general industry commentary and, responding to follow-up questions, gradually discloses company-specific information they shouldn't be sharing. The analyst doesn't flag it because the conversation felt natural.”
This gradual drift from general to specific is the primary mechanism by which MNPI enters expert call programs. Understanding it is the first step toward preventing it.
The Mosaic Theory: What It Protects and What It Doesn't
Mosaic theory holds that an analyst may legitimately reach investment conclusions by combining multiple pieces of individually non-material or public information — even if the resulting investment thesis, if widely disclosed, would be material to the price of a security.
The protected activity looks like this: an analyst synthesizes publicly available financial filings, supply chain data gleaned from public sources, expert knowledge about industry operating dynamics, and primary research observations gathered from channel checks. Each tile in the mosaic is independently legitimate. The conclusion that emerges from combining them is the analyst's proprietary insight — not MNPI.
What the mosaic theory does NOT protect is substantially different. The mosaic theory cannot be used to launder MNPI. If one piece of information in the analyst's mosaic was obtained through a disclosure that should not have been made — company-specific, non-public, and material — the mosaic defense fails for that investment thesis.
The practical test: would each individual piece of information in the mosaic be appropriate to trade on independently? If any single tile would, standing alone, constitute a trade on MNPI, the mosaic defense fails. The number of legitimate tiles surrounding it does not dilute the tainted one.
“Mosaic theory is a research methodology, not a compliance shield. The moment one piece of information in your mosaic would be material if disclosed, you've left the protected territory.”
— Securities compliance attorney (illustrative)A further practical implication: the mosaic theory protects synthesis, not accumulation. An analyst who collects ten pieces of public information and one piece of MNPI has not created a mosaic. They have created an MNPI-tainted thesis with ten pieces of public information attached to it.
The mosaic theory was articulated by regulators precisely to protect legitimate fundamental research — the kind of painstaking, resource-intensive analysis that aggregates public data into investment conclusions. It was not designed to provide cover for programs that lack the compliance infrastructure to distinguish between legitimate research and MNPI exposure.
Building Compliance Infrastructure for Expert Programs
A defensible expert network compliance program is not a single policy document. It is a set of operational protocols that function at each stage of the expert engagement process. The following framework describes the minimum viable compliance infrastructure for institutional expert programs.
Pre-Call Screening
Every expert must be screened before the call occurs. Screening has three mandatory components. First, confirm that the expert is no longer in a position of active access to the target company's material information — this means verifying both their departure date and the nature of their last role. An expert who left a company three months ago may still be subject to confidentiality obligations and may still have access to systems.
Second, confirm that the expert's engagement terms include an explicit prohibition on disclosing MNPI. Reputable expert network platforms include this in their standard expert agreements, but the buy-side compliance function should verify this rather than assume it.
Third, review the expert's last role and departure date against any known material corporate events at the target company. An expert who departed a company two weeks before a significant undisclosed transaction should not be engaged on topics related to that company until the relevant window has passed.
Call Monitoring Protocols
All expert calls should either be recorded for compliance review or conducted with a compliance-trained moderator present on the call. These are not equivalent options — real-time moderation provides the ability to intervene before a disclosure is completed. Recording-only protocols can identify MNPI after the fact, but cannot prevent it.
Before each call begins, a "stop list" of prohibited topics should be defined and distributed to the moderator and the analyst conducting the call. The stop list should include: specific undisclosed transactions the firm is aware of, any topics identified during pre-call screening as elevated risk, and standing prohibitions on asking about specific company guidance, pricing, or contract terms.
The Stop Protocol
If an expert begins disclosing what appears to be MNPI, the call moderator must: immediately interrupt the disclosure using a prepared phrase (e.g., "I need to pause you there — we should not go further on that topic"), note the exact timestamp of the interruption, and escalate to the compliance function before any trading action is taken on the information disclosed up to that point.
The stop protocol must be a firm rule, not a judgment call made by the analyst in the moment. Analysts are motivated to continue gathering information. They are also, by training, pattern-recognition practitioners who may rationalize that the expert's disclosure was not actually material. Neither of these factors is compatible with sound MNPI stop-protocol execution.
Post-Call Documentation
Every expert call should produce a documented compliance sign-off within 24 hours of the call concluding. The compliance sign-off must include: confirmation that no MNPI was disclosed (or a flag that the call is under review), the name of the moderator or reviewer, the date and duration of the call, and the topics covered.
“The stop protocol is not optional. If you don't have a real-time mechanism for halting an MNPI disclosure, you are relying entirely on the expert's judgment. That is not a compliance program. That is hope.”
Post-call documentation serves two functions: it creates the compliance audit trail that regulators and internal auditors will examine during any review, and it forces a deliberate compliance evaluation of each call rather than allowing the information to flow into the investment process without a compliance checkpoint.
Common Compliance Failure Patterns
Expert network compliance failures tend to concentrate in a small number of recurring patterns. Identifying these patterns in advance is the most efficient way to design controls that prevent them.
Pattern 1: The gradual disclosure. The expert begins the call with general industry commentary — entirely appropriate and publicly grounded. Over the course of the call, responding to increasingly specific follow-up questions, the expert moves from industry dynamics to company-level observations to specific undisclosed information. No one interrupts the progression because each individual step felt like a reasonable continuation of the prior topic.
Pattern 2: The former insider who doesn't know the rules. An expert who departed a company relatively recently may not realize that knowledge they carry from their tenure is still material and non-public. They may believe that because they no longer work there, the information is somehow no longer sensitive. This is incorrect as a matter of law, and no expert screening process that relies solely on the expert's self-assessment of what is appropriate will catch it.
Pattern 3: The documentation gap. A call occurred. Something questionable was disclosed. Nobody logged it. The trading desk proceeded without a compliance checkpoint. A compliance audit months later reveals the gap — and by that point, reconstructing what was said and what was acted on is extremely difficult. Documentation gaps are often not the result of intentional concealment. They are the result of compliance processes that are treated as optional when calls feel routine.
Pattern 4: The platform-only assumption. Expert network platforms conduct their own screening of experts. This screening is a necessary component of a defensible compliance program — but it is not sufficient on its own. Platform screening is designed to catch systemic issues. It is not a substitute for firm-level compliance protocols that are calibrated to the firm's specific investment activities and the specific research questions being asked on each call.
Running a Legally Defensible Expert Program
Expert network programs are legally defensible when run with proper compliance infrastructure. The history of enforcement actions in this area is not a story of expert network programs being inherently dangerous — it is a story of specific program failures that were, in retrospect, entirely preventable.
The vast majority of expert calls produce no MNPI exposure whatsoever. Experts who understand their obligations, working with analysts who understand what they can and cannot ask, supervised by compliance infrastructure that creates real checkpoints — that combination produces a research program that is both legally sound and operationally effective.
The risk is concentrated in a small number of scenarios that are entirely preventable with the right protocols. The four failure patterns described above account for the vast majority of expert network compliance incidents. Each of them has a corresponding structural control: call monitoring for gradual disclosure, departure-date verification for the former insider problem, mandatory documentation for the documentation gap, and firm-level compliance protocols for the platform-only assumption.
Compliance infrastructure is not a constraint on research quality. It is the operational foundation that makes a research program durable. Programs that invest in compliance infrastructure can engage experts confidently, document their processes transparently, and defend their research methodology to any stakeholder who examines it. That confidence is not just a legal benefit — it is a competitive one.