PowerUp Coaching — Living Knowledge Base

Insight: The Analytical Dead End — When Smart AI Answers Don't Land, the Blocker Is an Unexamined Core Belief

6 min read

“I was approaching it more from a psychological point of view. What is blocking me in order to making a choice?” — Dirk Ohlmeier

“It came out, yeah, you know, you hate executive search category because you believe it’s a broken system, so, hello! Nothing’s gonna sound right to you.” — Dirk Ohlmeier

“Usually when you have that resistance, you can’t… you don’t acknowledge to yourself that you have the resistance. Like, it’s one of those brain tricks that it says, I have no resistance, I’m completely open, but in the meantime, there’s a whole bunch of resistance at play.” — Lou D’Alo

Session context: 2026-03-12_Mastermind — Dirk shared a breakthrough about his GEO category naming problem. He had spent sessions asking Claude tactical questions (what should my category be?) and getting smart, well-organized answers that never felt right. After a workout, he reframed the question entirely — from “what should I do?” to “what’s blocking me from making this choice?” — and got a conversation about core beliefs that unlocked the actual problem.

Core Idea

AI is very good at giving analytical answers. When someone asks a tactical question — what category should I use for my business?, what’s the right positioning?, which option should I choose? — it will produce multiple well-reasoned options. The answers are logically sound. They are not wrong.

And yet the person keeps feeling like nothing fits. They dismiss option after option. They go back for more options. Still nothing lands. The conversation generates volume but no resolution.

Dirk’s discovery names the mechanism: when smart answers don’t land, the problem is not the answers. It is an unexamined belief underneath the question. Dirk’s belief was that executive search as a category is broken. With that belief operating as an invisible filter, no category name within executive search could ever feel right — because the category itself was pre-rejected at the belief level. AI was efficiently optimizing inside a frame that was itself the problem.

Lou adds the general principle: resistance of this kind is typically self-invisible. The brain does not present it as “I have a belief that’s blocking me.” It presents as “none of these feel right” or “I’m not sure any of these capture what I do” or simply a persistent flatness in response to logically sound suggestions. The person experiences themselves as open and searching. The resistance is operating below that.

The pivot that unlocked it: Dirk explicitly changed what he was asking AI to do. He stopped asking for solutions and started asking for the psychological layer: “What is blocking me in order to making a choice?” That reframe invited a different kind of conversation — one that excavated the belief rather than trying to work around it.

This is a practical AI coaching technique with a specific trigger condition: when a person has received multiple high-quality AI responses and still can’t commit to any of them, the tactical question is the wrong question. The right question is psychological.

Relationship to the Belief-Fear-Obstacle-Solution Arc: This is the AI-mediated version of the Arc’s level diagnosis problem. The Arc says that coaches tend to intervene at the obstacle level (level 3) when the actual issue is at the belief level (level 1). AI conversations create an exact parallel: the user keeps prompting for solution-level (level 4) help when the blocker is at level 1. The analytical dead end is the signal that you’re prompting at the wrong level.

Practical Application

The Analytical Dead End Signal — deploy when you or a client have received multiple well-reasoned AI answers to a question and none of them feel right:

  1. Recognize the pattern: More than 3 rounds of high-quality AI suggestions all dismissed or left as “I’ll think about it.” This is the trigger.

  2. Name it to the client (or yourself): “The AI is giving you good answers. That means the problem isn’t the answers. Let’s go looking for the thing that makes all of these feel wrong.”

  3. Pivot the AI question from tactical to psychological:

    • Instead of: “What should my positioning statement be?”
    • Ask: “I’ve been given multiple well-reasoned positioning options and none feel right. What core belief might I be holding about this domain that would make any option within it feel unacceptable?”

  4. Let AI excavate the belief: In this mode, AI shifts from option-generator to belief-surfacer. The conversation should start producing statements like “you seem to believe that…” or “an implicit assumption here might be…” — and those are what you evaluate, not the options.

  5. Once the belief is named: Decide whether to examine/challenge the belief or redesign around it. Dirk’s revelation was that he genuinely believed executive search was a broken system — which is legitimate. The question then became: do I want to position within that system or as an alternative to it?

Coaching adaptation: Use the same pivot in client sessions. When a client has received clear, actionable coaching recommendations across multiple sessions and is still not implementing any of them, stop generating better recommendations. Ask: “What would have to be true about you or your situation for any of these to feel like the right move?”

Evolution Across Sessions

This establishes the baseline for using psychological reframing as an AI prompting technique when tactical prompting produces high-quality answers that the person still can’t use. Dirk’s story is a particularly clean example because he had a specific, nameable belief (executive search is broken), a specific failed approach (tactical category questions), and a specific pivot that worked (psychological reframe). The directness of the example makes it highly extractable. Extracted via Mode B human-dimension-audit retroactive pass, 2026-06-16.

Graph View