The Cost of Untested Assumptions in Regulatory Approval Strategy

The problem with a strategy built on inference

Most regulatory approval strategies are built on second-hand intelligence: what your regulatory affairs team heard at an industry event, what your external counsel remembers from a similar case two years ago, what a former regulator now on your payroll thinks the current team will care about. None of this is worthless. All of it is inference.

The risk is not that your strategy will be wrong in some abstract sense. The risk is that you will discover it is wrong at the point of maximum cost: after submission, after public commitment, after your board has been told the approval is on track.

Where the assumption risk actually sits

The failure mode is rarely that a reviewer wakes up hostile. It is that the strategy solves for the wrong concern. You build a case optimised for capital adequacy questions when the reviewer is quietly worried about operational resilience. You prepare a governance narrative when the actual scepticism is about conduct history. You address the concerns the last case turned on, not the ones this case will turn on.

Untested strategies tend to fail in four specific ways:

1. Priority mismatch. Your submission emphasises what you think matters. The reviewer weights something else entirely.
2. Tone mismatch. The register is wrong: too commercial for a prudential reviewer, too defensive for a supervisor who wanted candour.
3. Sequencing mismatch. You lead with the argument that should have been third, burying the point the reviewer needed to see first.
4. Personnel mismatch. You assume continuity in the review team and miss that the individual whose views shaped your strategy has moved on.

Each of these is invisible until the strategy meets the actual decision-maker. By then, correction is expensive and sometimes impossible.

What testing actually means

Direct testing does not mean asking a regulator to pre-approve your strategy. It means creating structured, appropriate opportunities to surface how the actual reviewers think about the questions your submission will raise.

Good practice includes:

• Pre-application meetings used properly. Most firms use these to present. The value is in what you ask and how carefully you listen for what is not said.
• Structured third-party soundings. Advisers with current, direct relationships can test specific propositions without your name attached, provided the brief is precise.
• Peer intelligence. Firms that have recently been through the same review team can tell you what actually mattered, not what the guidance says should matter.
• Written questions ahead of submission. A short list of clarifying questions often reveals more about reviewer priorities than any meeting.

What most people get wrong is treating these channels as confirmation exercises. You are not looking for reassurance. You are looking for the gap between what you assumed and what is actually true.

The judgement call: how much testing is enough

There is a real trade-off. Over-testing signals uncertainty, consumes reviewer goodwill, and can slow the process. Under-testing leaves you exposed. The calibration depends on three factors:

• Novelty. The more unusual your application, the less prior case history you can rely on, and the more direct testing you need.
• Reviewer turnover. If the team has changed materially since the last comparable case, historical intelligence is degraded.
• Reversibility. If getting it wrong means resubmission, adjust accordingly. If it means withdrawal and reputational damage, test harder.

A useful discipline: for each major pillar of your strategy, ask what specifically you know about how the actual review team will respond, and how you know it. If the answer is "our adviser thinks" or "last time they said," you have an assumption, not a tested position.

What good looks like

A well-tested strategy has three characteristics. First, the team can articulate, for each significant argument, the specific evidence that this argument will resonate with these reviewers. Second, there is a documented list of assumptions that could not be tested, with an assessment of the exposure each represents. Third, the strategy has been revised at least once in response to what testing surfaced. Strategies that survive testing unchanged usually were not tested hard enough.

Your next decision

Before your next submission milestone, take an hour with your regulatory lead. For each pillar of the strategy, ask: what have we directly tested, what have we inferred, and what would we do differently if the inference is wrong. If that conversation produces surprises, you have your answer about whether to proceed on the current timeline.