The Org Chart is a Decoy

Every organization has two structures running at once. One is on the slide: reporting lines, titles, dotted-line relationships. The other is the one that actually decides things, and it rarely matches the first.

You can tell the difference by watching what happens before a meeting, not during it. The real decision is usually made in a hallway conversation, a text thread, or a five-minute call two days earlier. What happens in the room is often just the room finding out. If you’ve ever walked into a “discussion” that was clearly already settled, you’ve seen this firsthand. And the uncomfortable part is that nobody tells you when it happens. You have to notice.

Here’s what I look for now, and what I wish someone had handed me a decade earlier:

Who gets asked for their read before the decision, not after. Formal authority tells you who signs off. It doesn’t tell you whose objection actually stops something. Those are frequently different people, and the second group is the one worth mapping.

Who gets copied on the thread that matters, not the one that’s performative. Every initiative has a real distribution list and a decorative one. Learn to tell them apart quickly. It’s usually obvious once you know to look.

Whose “no” is expensive to override. Some people can be outvoted cheaply. Others can’t, regardless of title, because overriding them costs political capital nobody wants to spend. That’s real power, and it’s almost never printed on a business card.

None of this is about becoming cynical or transactional. It’s about accuracy. Reading the room correctly is a technical skill, the same as reading a balance sheet. Much like financial fluency, most of us were never actually taught it. We picked it up slowly, expensively, usually after getting it wrong at least once in a meeting that mattered.

This isn’t a place AI replaces your judgment. Reading a room is still a human skill. But it’s a genuinely useful way to pressure-test your read before you walk in, especially when you’re new to a room or the stakes are high enough that a blind spot would be expensive.

Try this prompt with whatever generative AI tool you use. It’s built around CLEAR, the framework used in Her Executive Ascent’s leadership programs: Context, Length (of response), Examples, Actions, Role.

Context: I’m walking into a meeting about [decision or initiative]. Here is a list of who’s in the room, by role and level only, no names: [list roles and levels, plus what you know about each one’s incentives, history with this topic, or recent friction].

Length: Three short sections, one per action above, no more than two sentences each.

Examples: A blocker might look like someone who wasn’t consulted and has quietly pushed back on similar initiatives before. A surprise objection might come from a function that seems unrelated but has exposure I’m not accounting for.

Actions: Identify who is likely to support this and why, who has enough informal influence to slow or block it even without formal authority, and what objection is most likely to come from a direction I’m not expecting.

Role: Act as an experienced organizational strategist helping me prepare for a meeting. Ask me clarifying questions, one at a time, until you have enough information to provide a practical response.

Every line does one job:

• Context hands over the situation without a single name attached,

• Length of response tells it exactly how much to hand back so you’re not wading through paragraphs to find the one useful line,

• Examples anchor what “blocker” and “surprise objection” actually mean instead of leaving the model to guess,

• Actions is the specific ask, and

• Role sets the lens it’s answering from.

The value isn’t the output on the first pass. It’s that a good model will surface something you hadn’t thought to check yourself. Use it to catch what your own position in the room might be hiding from you, not as a replacement for your own read. And role and level only is the right call regardless of the tool. There’s no version of this exercise that requires the AI to know who anyone actually is. Never share identifying, private, or confidential information with any AI tool. This is the exact kind of gap AI Fluent: Foundations for Executive Women was built to close.

P.S. It’s a self-paced course with lifetime access, and you’ll practice the framework directly with an AI Foundations Coach built for this course, not just read about it.