Most Consulting Is Theater

Most consulting is theater.

You pay for analysis, receive a beautifully formatted diagnosis, everyone nods dutifully in the readout, and ninety days later, nothing in the business actually runs differently except the invoice has cleared.

I spent over two decades learning operations, engineering, technical program management, and AI/software policy in some of the most profitable engineering companies at the bleeding edge of technology. Not talking about things or making recommendations without seeing the end result. Actually holding accountability for success or failure at the end and learning lessons the hard way. That world beat one habit into me: an analysis that doesn’t change what happens on Monday morning is a hobby, not a deliverable.

I set one rule for myself: nothing leaves my outbox unless something in the business runs differently within two weeks of receiving it. That applies to the way I worked at Google, the way I run projects for myself, and in how I provide value to my clients at Aetherr Advisory.

The cycle we’re all stuck in

The pattern I keep running into across every industry and business at every size and level of complexity is that somebody believes they can only grow by growing headcount. More leads, more coordinators, more clients, more assistants means more revenue but also more payroll. Payroll and revenue are somehow intrinsically linked, and the best we can hope for is some kind of non-linearity. AI was supposed to break that. Instead, most owners got a new hobby: copy-paste prompts, a ChatGPT tab, maybe a $200/month tool stack nobody fully adopted. The tools got cheaper. The bottlenecks didn’t move.

That’s not a technology problem. It’s an architecture problem. Nobody in the building owns the job of looking at how the business actually runs (not the org chart version, the real one) and translating that into systems.

What we actually do

Every Aetherr engagement runs the same loop: Decipher, Extrapolate, Create. No cute acronym for it, no “this is my framework”, just 3 steps to actually get some work done.

1. Decipher. Audit how the business actually runs. Not the pitch deck. Where time actually goes, where data actually lives, where work actually stalls. Which means we need to talk to the people on the ground, not the folks managing the roadmap.
2. Extrapolate. Strip away the guru hype and map which frictions AI genuinely removes, and which ones it simply shifted to be somebody else’s problem.
3. Create. Build the thing. Production-grade working demonstration, not a slide about a future state. Leveraging AI means that your explanation of how to do the thing probably took more time than just having the AI prototype it.

Proof, not promises

Last week we ran the same process for 5 clients that all look very different on the surface, but in reality all have the same core issue: how can I increase my online visibility without hiring a bunch of junior marketing or sales assistants?

• 3 independent real estate agents with very different online footprints went through full operational and search-visibility audits. Each walked away with an automated solution to the single most painful workflow in their pipeline. One they built for themselves after understanding how to use AI tools to drive scalability for things they already know how to do for themselves.
• One prop-tech platform got the same treatment at the company level: search architecture, cloud cost posture, and an implementation sequence their own AI coding assistant can execute.
• And for good measure, we turned the same flow back on each of these engagements to better understand what’s working, what’s not, and how we can serve more clients asking this same question. The result? We packaged the audit itself as a product: a self-serve diagnostic for real estate agents that runs the analysis, delivers the results, and hands them a path forward - no call required to start. It’s the same audit, minus the calendar friction.

The detail almost nobody does

Here’s the part I’m proudest of, and it costs any consulting firm exactly nothing to copy:

Every report we deliver ships in two forms. The human version: executive brief, prioritized fixes, plain language. And the machine version — clean Markdown, structured so you can paste the entire analysis straight into Cursor, Claude, Copilot, or your assistant of choice, and have it start implementing the same day.

Listen, it’s 2026 and the bottleneck is no longer generating analysis. Any model can produce a competent-sounding audit. The bottleneck is the translation layer between here’s what’s wrong and here’s what’s fixed. That layer is where consultants used to hide their value. We just make it executable. If you’re buying advisory work in 2026, that’s the new bar. Anything less is theater with better headshots.

Three tests for your next consultant

Whether you hire us or anyone else, run these:

1. Ask what runs differently in 14 days. If the answer is a workshop, keep walking.
2. Ask for the deliverable in a format your AI assistant can execute. If they can’t produce it, they’re selling analysis, not outcomes.
3. Ask what they’ve automated in their own business. If you get a dodge, they’re either admitting they don’t practice what they preach or trying to keep the secret sauce for themselves.

Give me your toughest problem, and let’s see what I can do with it.