Last month I rebuilt my entire website in roughly 24 hours.
Not because I'm a developer. Not because I have a technical team. Because I finally stopped fighting a tool that wasn't working — Webflow, which had quietly broken my mobile site behind my back after weeks of work — and made a fast call to move to Replit instead.
Three sites. DNS routed. Mobile optimized. $460 back in my pocket come September.
I'm telling you this not because website builds are interesting but because of what it represents. That decision — read the situation, cut the loss, execute without a committee — is exactly the gap that a new Inc. report is trying to explain. And most of the commentary around it is missing the point entirely.
The numbers everyone is quoting
EY surveyed 15,000 employees across 29 countries late last year. The headline finding: 88% of workers use AI tools daily. Only 5% qualify as advanced users who extract real value. That 5% is gaining a day and a half of productivity every single week.
OpenAI's State of Enterprise AI report confirmed the pattern from a different angle. Workers in the 95th percentile send six times more prompts to AI tools than their median colleagues. For coding tasks, that number explodes to 17 times more.
Same tools. Universal access. Wildly different results.
The question everyone is asking is: what are the 5% doing differently? The article's answer is reasonable — they experiment more, they match tools to tasks, they think in discrete steps rather than asking AI to do everything at once.
That's all true. But it's framing the gap as a skills problem. I think it's a mindset problem. And that distinction matters enormously if you're a solo operator trying to figure out where you actually stand.
What the organizational framing misses
The Inc. piece is written for founders thinking about their teams. How do you close the AI gap across 50 or 100 or 500 employees? How do you train the 95% who are using AI for spell-check and document summarization when they should be using it to restructure a client proposal?
That's a legitimate problem. It's just not your problem.
If you're a solo fractional operator, a founder running lean, or a specialist who competes on depth rather than headcount — the gap isn't a threat to you. It's the whole game.
Here's the math that nobody is saying out loud: a 100-person company where 60 employees barely use AI is losing 40–60 hours of productivity every single day. Over a year, that's the equivalent of five full-time employees' worth of work, paid for and not delivered. Meanwhile, their competitors' 5% are compounding that advantage daily.
You are, by definition, in the 5%. You have to be — there's no team to carry you if you're not. And that means you are already outperforming the median employee at companies that are 10, 50, 100 times your size.
Not eventually. Right now.
The barbell nobody talks about
The report identifies something it calls a "barbell strategy" — the pattern that distinguishes power users from everyone else. Deep mastery of one or two primary tools, plus five to eight specialized applications rotated by task. Not chasing every new launch. Not dogmatically loyal to one platform either. Developed intuition about which tool is best for what.
This is not a revelation if you've been paying attention. But it's worth naming explicitly because most productivity advice goes in the opposite direction — pick one tool, go deep, ignore the rest.
The barbell works because different tools have genuinely different strengths. The power users in the study use ChatGPT for brainstorming, Claude for analysis, Midjourney for visuals. They're not forcing one tool to handle everything and blaming AI when it underperforms on tasks it wasn't built for.
My version of this: I know which problems I bring here, which ones I take to Perplexity, which ones I let build without giving it too many instructions. That rotation isn't random. It's pattern recognition built from enough reps to know what each tool does well and where it starts to hallucinate, hedge, or just get boring.
That knowledge is not transferable via a lunch-and-learn. It comes from the kind of daily, iterative, sometimes feral use that the EY survey is calling "advanced." It also comes from being willing to make a fast call — like abandoning a tool that's costing you more than it's giving you — without needing consensus from a room full of people.
The uncomfortable implication
BCG found that only one-third of employees say they've been properly trained on AI tools. That's usually framed as an organizational failure — leadership not investing in training, change management gaps, adoption stalling at the front line.
It's also an opportunity.
The training gap means your clients — the founders, the operators, the teams you work with — are almost certainly in the 95%. They have access to the same tools you do. They are not using them the way you are. And that gap is not closing as fast as the hype cycle suggests, because the hype cycle is selling tool access and the real differentiator is judgment about how to use it.
That's what you're selling, whether or not you've named it that way.
The 5% gap isn't closing. And for the solo operator who's already on the right side of it, that's not a problem.
That's the whole business.