The Waste-to-Wealth Methodology: Why Human-First AI Wins

Pull out your phone. Open ChatGPT. Type this exact prompt into a new chat: "Give me a random number between one and twenty-five. Don't tell me what it is."

I do this exercise in nearly every room I walk into now. And every single time, somewhere between sixty and ninety percent of the room gets the same number.

Seventeen.

Why? Because the algorithm has decided that seventeen is the most statistically satisfying random number between one and twenty-five. It is not random. It is pattern recognition wearing a costume. And that is the entire point.

AI Is Not Smart. It's a Pattern Recognition Game.

People talk about AI like it's smart. People talk about AI like it's going to leave us behind. The reality is, it's a pattern recognition game. It is only as good as what you ask it, what you feed it, and the guardrails you set in place around it.

Which means it is a tool. Not a save-all. Not a solution. No different than any other tool you have ever used. The question is not whether you have AI. The question is whether you and your team are learning to use the tool effectively, or whether you are just becoming a consumer who's stacking subscriptions and asking it to make your emails sound more professional.

I want to be really, really clear about this, because most of what I see right now in the executive rooms I sit in is one specific failure pattern. Not because the leaders are not smart. They're some of the sharpest people I've ever worked with. The pattern is structural.

Most businesses are layering a tool on top of an already broken system. And now they're moving faster in the wrong direction. They cannot understand why they're not getting returns. They cannot get ahead of it. They cannot pull the details down. Right?

Waste In, Waste Out Faster

Here's the reality. If you have waste in, you are going to get waste out faster. If you have clarity in, you can build some really cool things. We live in an era where if you can envision it and you can communicate it clearly, you can build it. That is the new ceiling. That is the new floor. The people who treat AI as a clarity multiplier are pulling away from the people who treat it as a replacement for the thinking they have not done yet.

I have been in the waste-to-wealth business since I was eleven. My first business was picking up dog waste in my neighborhood for profit. Literally turning trash into cash. I have spent the last twenty-five years doing variations of the same work, just at progressively bigger scales. With humans first. Now with humans and AI together.

What I have learned over those twenty-five years is this: anything you do that does not add incremental value is, by definition, waste. Neutral is not the game. Additive is the game. And the first leader inside any industry to find the margin wins. Margin is not in pricing. Margin is not in volume. Margin is in the waste you have not named yet.

The Three Categories Of Waste Every CEO Has

There are three categories of waste in every business. Most CEOs start with the wrong one.

Leadership waste. Decisions made through misalignment. Trust gaps. Conflict avoidance. Decisions made in a vacuum. Patterns nobody wants to name in the room. This is the biggest gap between where you want to be and where you are right now, and it almost always comes back to the people. If this part does not get serviced first, nothing else will.

Financial waste. A byproduct of leadership waste, every time. Are we spending in directions we do not need to? Are our hires effective? Are the investments we're making producing returns? Most companies have no real metrics around this, because the leadership work has not surfaced what's actually true.

Operational waste. This is where the majority of AI tools live. And this is where most leaders try to start. The problem? If you have not done the work in the first two categories, every operational tool you bolt on amplifies the dysfunction. You move faster, you don't move better.

Operational waste itself shows up in four specific shapes inside almost every business I work with. It's almost always disguised as productivity.

• ▸Repetition waste. Same task over and over. Same email structures. Same quoting process. Same approach to answering the same five questions. There is never enough data entry, and there is no business where data entry is the highest use of a person's energy.
• ▸Context-switching waste. Twenty-three minutes, on average, to refocus to the original task you were on after an interruption. Multiply that across your team and your week and ask yourself how much of your payroll is funding people swimming back to where they were.
• ▸Decision waste. The same micro-decisions, repeatedly, by the same people. Pricing thresholds. Approval limits. Routing logic. If you have to make the same call ten times, you do not have a decision. You have an undocumented system.
• ▸Human-in-workflow waste. Work that crosses categories. Transactional, end-to-end processes that require a human to forward, approve, retype, and rename, when the work itself is not where the human is most valuable.

If you can eliminate the waste first and simplify the process, then you automate what is left. That is the order. Every time. Reverse it and you are paying for tools to make broken processes faster.

Why Most AI Projects Fail

Before any of my teams build a thing for any organization, we test for three failure modes. If we cannot get clean answers in all three, we do not build.

Knowledge curation. You have to have all the voices, all the decision-makers, all the players in the room when you are building something for a specific application. If you cannot capture the collective voice, you are missing a piece. The reason we have been able to build what we have built for some of the organizations I work with is that almost every conversation has five or six people in it. We get every angle and every perspective that one person simply cannot cover.

Adoption. If nobody uses it, it is waste by definition. The greatest source of AI waste in the world right now is people building cool things that look fun and don't function. Not integratable. Not secure. Not built around the actual end user. I just finished a tool for a thought leader in his eighties. The build constraint was simple: if there is a way an eighty-year-old can get lost on a single web page with one button, they will. So we engineered the tool around that reality, not around what looked impressive in a demo. If your people see your AI tool as a threat, or as something that does not add value, they will never use it.

Governance. Stale knowledge bases. Bad data. The second you get a wrong answer out of the system, the second you build off bad inputs, you are off the rails. And almost nobody is paying attention to this part yet, which leads me to the part of this conversation that almost no CEO is having out loud.

The Eighteen-Month Sleeper Risk

I want to be really, really clear here. This is not fear-based. This is just a structural reality that most businesses are walking into without seeing it.

Not all LLMs and not all accounts are created equal. Many personal-tier accounts on the major platforms train the underlying models on whatever you feed them. Which means anything fed in becomes part of what the system has access to. Most companies have thirty, fifty, seventy-five people on their own personal accounts, with no policies, no guardrails, no ability to monitor what is being shared. Proprietary information. Client data. Intellectual property. Quietly going into systems the business has no visibility into.

Eighteen months from now, this becomes one of the bigger exposures sitting on a CEO's desk. Not because of a malicious actor. Because of forty employees who have been pasting in client emails, contracts, and pricing for the last year and a half because nobody told them not to.

You do not have to use AI broadly to put governance in place. You do have to put governance in place. The order matters.

What It Looks Like When You Do It Right

I want to walk you through a real arc. Names removed, but the texture is real.

Nine months ago, I started working with a logistics company. Family-built business. Second-generation CEO who had recently bought the firm from his family. Strong leadership team. Real people. Real work. They had a strategic-planning issue, which is what got me in the door. By the end of the second meeting, I was asking the only question that actually mattered: where can you have the biggest impact, and what does this company need to become to get there?

What surfaced: data mistrust. Half-baked ideas held by individuals but not aligned across the team. Decisions out of alignment. Lack of trust at the executive level. Unclear documentation. Weekly meetings where everyone walked out on different pages.

I want to be really, really clear. This is not a knock on this team. This is how most companies operate. The question is not whether the dysfunction exists. The question is whether the dysfunction still serves where you want to go.

The decision in the room was no.

So we started with the people. Owner accountability on every project. Better questions in meetings. A single source of truth (an AI recorder running every executive session, with weekly summaries). One member of the team owns the process. If there is ever a dispute, the transcript settles it.

Then a tech audit, owned by the executive who used to be the most resistant to technology. Now he's the one leading it. Why did we buy this originally? What did we intend it to do? What is it doing today? Where is the waste? They consolidated their tech stack into one platform and saved roughly thirty thousand dollars on a renewal that was three weeks away.

Then the verticals. Three of their bigger industries had never been positioned as expertise areas on their site. We sat in a room together, took the team's collective language, and built three vertical landing pages live, in front of them, in a single hour. Six iterations on the first one. Done. Same content was repurposed into a four-page sales handout, a one-pager, an event strategy for the year, and a LinkedIn content engine running across three voices on the leadership team.

Then a content engine. Five agents that handle research, post writing, article writing, newsletter, and blog production, with a sixth agent that does voice and policy validation against a master brand voice. The marketing team reviews and approves. Nothing externally facing publishes without a human signoff. Ever.

None of this required a custom multi-million-dollar SaaS spend. None of it required a six-month build. It required clarity first, sequence second, AI third.

And here's the part that matters most. The people work has not stopped. Nine months in, I'm still in one-on-ones with members of that leadership team. Working through hard conversations. Naming patterns. The systems and tools followed the people. They did not replace the people. Does that make sense?

Three Layers Of Leverage

If you take nothing else from this, take this. There are three layers of AI leverage available to you. Most businesses stop at layer one and wonder why nothing is compounding.

Layer one: prompts. Templates. Reusable language. Copy-paste assets. The email structure your team uses every day. Saves time. Fully repeatable.

Layer two: custom assistants. A custom GPT or assistant that holds your methodology, your voice, your context. You are not re-explaining yourself every conversation. The risk: many off-the-shelf assistants have no validation, no memory, and a lot of drift. Useful, but watch for hallucination.

Layer three: full systems. End-to-end builds. The content engine I described above. A booking calculator that replaces five back-and-forth phone calls. A sales-intelligence pipeline that constantly runs in the background and surfaces opportunities to the team with a trigger and a reason to act. You build it once, it runs ten thousand times.

Any highly repetitive task, any decision task, any context-switching task fits one of these three layers. Start with layer one this week. Move to layer two this quarter. Build toward layer three this year. The compounding is real, and it does not start until you stop treating AI like a chatbot.

Three Demands You Make Of AI Every Time

Three small shifts, and the quality of what AI gives back to you transforms. None of these require any new tool. They are how you use the one you already have.

One. Cast a vision before you ask for output. Hit the voice-record button on your AI app and talk for ninety seconds about what you want. Why you are doing this. What problem it solves. What the ideal output looks like. The more context you feed in up front, the more accurate the output, and the less you wrestle the result into place after the fact. Most people skip this and then spend an hour editing.

Two. Assign a role. You would not hire a generalist to do specialized work in your business. Stop letting your AI do it either. "Operate as the world's best COO with thirty-five years of operations experience inside the moving and logistics industry." "You are a specialized claims adjuster who has seen millions of claims." "You are a CFO with deep experience in private-equity-backed services businesses." Set the role first. The output level changes immediately.

Three. Demand questions back. I never let an AI start building until I understand and approve the plan. I plan first. Research first. Set context first. Then I always close with: "What questions do you need to ask me to better perform this task accurately?" You will be surprised what gets asked. And you will know exactly where the system was missing context that would have produced a worse answer.

Plan, role, questions back. That is it. Three demands. Better output every time.

Three Questions For Every CEO In This Window

I'd like to leave you with three questions. They take ten minutes. Ask them of yourself, then bring them to your executive team.

One. What are we protecting that no longer serves us? Every business has a legacy line of business, a sacred process, a "we've always done it this way" that has not been questioned in years. Tradition is not a strategy. Audit it.

Two. Where are we adding effort without adding value? Time. Energy. Resources. Spend that is not generating revenue. Spend that is not deepening client relationships. Spend that is not driving the business forward. We all have it. The faster you name it, the faster you can move it.

Three. If we weren't forced to change later, what would we choose to change now? The next eighteen to twenty-four months are going to force change on every business in this country, whether the leadership wants it or not. The leaders who choose the change ahead of the curve will own the margin. The leaders who wait will be reacting from a worse seat.

Ask those three questions and you will walk out with more clarity than ninety percent of your peers. You may even walk out with a roadmap.

Architects Will Never Be Replaced. Doers Will.

I want to close with this. AI is not coming for the architects. It is coming for the doing. The repetitive doing. The transactional doing. The same-task-fifty-times-a-week doing. That work is changing fast.

The architects, the people who can hold a vision, ask the right questions, set context, and design the system, will be more valuable than ever. AI is about making humans more human. About leveling people up so they can do higher-value work inside the business. The point is not to run with fewer people. The point is to have all of your people running at their actual capacity.

That requires you, as a leader, to do three things in this order: clarity first, sequence second, AI third. Lead with the people. Surface the waste. Then automate what's left.

One last thing. The biggest investment you can make right now is not artificial intelligence. It is awareness and attention. Without those, your AI will not work. Your leadership will not work. Your marketing will not work. Your sales will not work. Everything moves at the speed of clarity, and clarity is a function of how willing you are to ask better questions.

Pain becomes perspective. Perspective becomes movement. And moved people move people. That is how this all works.

Where To Go From Here

If any of this hit, the question is not whether you have a waste-to-wealth opportunity inside your business. The question is which one you start with first, and whether you have the right people in the room to surface it. That is the work I do.