Testing Regulator and Decision-Maker Support Before You Build

Start with the decision you actually need to de-risk

Before any outreach, write down the specific commitment you are about to make: the build, the vendor contract, the hiring plan. Then list the two or three regulatory judgements it depends on. Most validation exercises fail because teams test the general approach rather than the specific interpretations that will sink the project if they turn out to be wrong.

If you cannot name the interpretations, you are not ready to test them.

Step 1: Separate the audiences

Regulators, trade bodies, and industry decision-makers respond to different signals and will give you different kinds of validation. Split them out:

• Formal regulator contacts: supervisors, policy leads, technical specialists.
• Informal regulator signals: speeches, Dear CEO letters, enforcement patterns, published Q&As from the last 18 months.
• Industry decision-makers: peer firms, key clients, distribution partners, auditors, and the law firms who quietly shape regulator expectations.

Each group requires a different test. Treating them as one pool is where most validation goes wrong.

Step 2: Build a falsifiable hypothesis, not a pitch

Frame each interpretation as something that can be disproved. "We believe the PRA will accept a proportionality argument for X because of Y" is testable. "We think our approach is compliant" is not.

Good hypotheses are specific enough that a 20-minute conversation with the right person can move your confidence up or down materially. If a conversation cannot shift your view, you are asking the wrong question.

Step 3: Test the written record first

Before you burn a single relationship, exhaust what is already public. Read the last three years of relevant speeches, consultation responses from your regulator, and enforcement notices in adjacent areas. Pay particular attention to the language used when regulators describe firms that got it wrong. That vocabulary tells you where the sensitivity sits.

What most people miss: the gap between what regulators say in policy statements and what they emphasise in supervisory letters. The letters are where actual expectations live.

Step 4: Use proxies before principals

Go to the ex-regulators, the specialist counsel, the Big Four regulatory partners, and the trade body technical committees before you approach the supervisor directly. They will tell you, in plain terms, whether your interpretation is defensible, unusual, or actively risky. A single 45-minute call with the right former Head of Division is often worth more than a full month of internal debate.

Run these conversations in parallel, not in sequence. If three independent sources flag the same concern, that is your answer.

Step 5: Approach the regulator with a specific, not a general, question

When you go direct, do not ask "are we compliant?" You will get a non-answer. Ask a narrow technical question that reveals the interpretation without seeking approval. "We are considering treating X as Y for the purposes of Rule Z. Are there recent supervisory expectations we should be aware of?" gives supervisors permission to signal concern without formally opining.

Watch for what is not said. Silence on a specific point, or a redirect to another team, is usually meaningful.

Step 6: Test with the peer group carefully

Peer validation is useful but treacherous. Firms will not tell you their real position on a call. They will tell you in the margins of a trade body working group, in a bilateral over coffee, or through their external counsel. Structure the conversation so they can share directionally without exposing their strategy.

If every peer says "we're still working through it," that is a signal. Either the interpretation is harder than you think, or nobody wants to be first.

Step 7: Score the evidence and make the call

Before the exercise, define what would constitute enough support to proceed, enough concern to pause, and enough opposition to redesign. Do this in advance. Teams that set the threshold after gathering evidence almost always talk themselves into proceeding.

A workable rule: if two or more independent, credible sources raise the same substantive concern, treat the interpretation as unvalidated regardless of how many others were positive.

What good looks like

A validation exercise that takes three to five weeks, uses six to twelve conversations across the three audiences, produces a written record of what was tested and what came back, and ends with a clear go, adjust, or pause decision tied to the specific interpretations you started with.

Your next action

Before your next steering committee, write down the two or three interpretations your plan depends on and the specific evidence that would falsify each. If you cannot do that in an hour, your team is not yet ready to commit resources.