Hi Friends,

A CEO I am working with asked me a question last month that I keep coming back to.

His media company has 10 people. He has been rolling out Claude across the team for a few weeks. Real adoption. Real templates being used. People are genuinely faster.

He asked me: "Now that my employees are getting more efficient, how do we actually impact the business?"

All we hear about are billion-dollar companies running armies of AI agents, generating millions in new revenue overnight.

Nobody is talking to the ten-person company trying to figure out the same thing.

Most companies are at Level 2 and do not know it. By the end of this you will know where you stand and what to do about it."

Every small and mid-size company I advise has the same pattern.

Three people are absolutely crushing it with AI. One in sales. One in ops. One in marketing. They have built their own workflows. Their own prompts. Their own systems. They are dramatically faster than they were a year ago.

And none of it is connected. It’s like having organs with no circulatory system.

This is the invisible ceiling of Level 2 on the AI maturity pyramid below. Individual velocity is real. Collective intelligence is zero. What sales learns on Monday does not show up in what ops decides on Wednesday. What marketing discovers about customers does not change how the sales team pitches on Friday. It’s an area I am thinking a lot about lately.

Here is the AI journey I see most people on:

I think about Blue Bell ice cream a lot. The best ice cream in the country. Texan born. Their Cookies and Cream flavor is the standard everything else is measured against.

Now imagine their sales rep notices Cookies and Cream spiking in two Houston zip codes. She sees it in her order numbers. Ops is still running production based on last quarter's forecast. Marketing is pushing a summer flavor because nobody told them anything changed.

Three talented people. Working hard. In separate rooms.

That is Level 2.

What Level 4 Actually Looks Like

Same company. Connected agents.

The moment Cookies and Cream starts spiking, the sales agent flags it into a shared company feed. The ops agent adjusts Friday's production run. The marketing agent shifts spend toward the zip codes where demand is building.

The CEO walks into Monday with a briefing. Three things the business learned last week. Two decisions that need their call.

They did not chase any of it. Their job is to look at the pattern and decide what to do next. Double down on the flavor. Expand to a new market. Call the wholesale account before a competitor does.

That judgment is still theirs. Every piece of legwork that used to eat their week is not.

That is the difference between a faster company and a smarter one.

Most advice on AI adoption tells you to do ten things. Here is what actually moves the needle based on every company I have studied and advised.

One: Fix your data before adding more AI. Every failed AI initiative I have seen skipped this step. Every successful one did it first. Your AI is only as smart as the information it can access. If your CRM, your ops data, and your financial data are sitting in separate systems that do not talk to each other, your agents cannot connect the dots no matter how sophisticated they are. Pick your one most important data source and make it clean and accessible. Start there.

Two: Pick one cross-functional workflow and connect it. Not five workflows. One. The companies that tried to transform everything got nothing. The ones that picked the single highest-pain handoff between two departments and built one agent to bridge it saw results within weeks. For most $25M companies that handoff is between sales and ops. What sales learns about demand never reaches the people making production or inventory decisions. That gap costs money every week. Close it first.

Three: The CEO has to own this out loud. This is the most overlooked lever at mid-size companies. Research across 51 enterprise AI deployments found that CEO-sponsored projects cleared bottlenecks three times faster than those delegated to a technical lead. Your team takes its cues from you. If AI adoption is an IT initiative, it stays at Level 2. If it is something you talk about in your weekly meeting, ask about in one-on-ones, and visibly use yourself, the company moves. This costs nothing. It just requires you to be deliberate about it.

Nobody has fully solved this yet. Not even close.

The companies I advise are somewhere between Level 2 and Level 3. The vision is clear: a management team waking up to new insights every week, the system acting on what it can, surfacing what needs a human decision.

That is the company worth building.

Who knows. Maybe I will start a company to help solve this.

'Til next time,

--Ali

P.S. Where does your company sit on the pyramid? Reply and tell me honestly. I read every response and the pattern across companies will be its own issue soon.

About Me: I am an AI-powered go-to-market strategist advising B2B founders and CEOs on AI adoption and growth strategy. Currently working with a handful of companies to move from individual AI speed to collective company intelligence. NBA playoffs start tonight. The best time of the year. Go Rockets.