The Window Is Closing…Don’t Get Left Behind

There's a famous experiment most people have heard of...

A frog dropped into boiling water jumps out immediately. A frog placed in cool water that slowly heats never jumps at all. The change is gradual enough to feel manageable at every moment.

Whether the science holds up doesn't matter. The pattern does. Because that pattern is exactly what's happening to most SMB owners right now.

The water is warm. Business is fine. Revenue is there. The team is functioning.

But, this isn't a warning about the future. The divergence has already started. What's still open is which side of it you end up on.

Why This Matters Now

The story most people tell about AI adoption goes something like this:

“Big companies are using it, small ones will get there eventually, and the gap isn't that serious yet….right?”

That story is wrong. And the numbers show why.

Salesforce tracked the difference between growing SMBs and declining ones at the end of 2024. Growing businesses planned to increase their AI investment at a rate of 78%. Declining businesses came in at 55%. That gap doesn't look dramatic in isolation. Compounded month over month, it becomes structural.

The JPMorgan Chase Institute showed that in January 2023, the AI adoption gap between employer and non-employer firms stood at 5.6 percentage points. By December 2025, it had grown to 10.8 points. The gap isn't closing. It's widening.

And here's the one that stops people: Goldman Sachs found that only 14% of small businesses are fully integrating AI into their core operations. The other 86% are experimenting, dabbling, or not engaging at all. The 14% are compounding exponentially. Much of the 86% are falling further behind every quarter.

TIME put it plainly: most AI initiatives never get out of the pilot phase and every quarter companies stay there…the gap between compounds.

Said another way… Pilots don't create compounding. Scaled operations do.

The gap isn't coming. It's already forming.

The Kodak Mirror

Kodak didn't fail because they didn't see transformation coming.

In 1975, a Kodak engineer named Steve Sasson built the world's first digital camera. Inside Kodak's own labs. They had the technology before anyone else.

In 1981, Kodak's head of market intelligence completed a major internal study at the CEO's request. The conclusion was clear: digital will eventually replace film. They knew.

By 1991, Kodak's executive team had internal projections showing digital would overtake film somewhere between 1996 and 1999. They were roughly right. It happened by 2003 to 2005. They knew the timing. They built models around it. They watched it come.

And yet…Kodak filed for bankruptcy in 2012.

So what happened? Four things. And every one of them is playing out in small businesses right now.

🔹 The margin trap. Film carried roughly 70% gross margins, and Kodak held 90% of the U.S. film market in 1976. Cutting into that revenue to fund an unprofitable digital division was financially irrational, quarter by quarter.

🔹 Organizational identity. Film people made it to the top at Kodak. The people who built the film business ran the company and decided what success looked like. Digital initiatives got funding.

🔹 They misread the competition. Kodak saw digital as a worse version of film: lower resolution, no physical prints, inferior quality. Digital wasn't trying to beat film on film's terms. It was solving a completely different job: instant sharing, portability, zero cost per shot.

🔹 They bet on the output of the old model. When digital became undeniable, Kodak invested heavily in photo printers. Pictures are eventually printed, aren't they?

Kodak didn't fail because they were blind. They failed because their existing model made it financially and organizationally DIFFICULT to act on what they could clearly see. That's not a vision problem. It's a structural one.

And it has played out again and again...

The U.S. travel agency industry had roughly 124,000 locations in 1995. Expedia launched in 1996. Priceline in 1998. By the mid-2000s, more than half were gone. Not bad businesses. Many had loyal clients, had been operating for 15 to 20 years, and were profitable. They just didn't adapt the operating model fast enough and subsequently and watched their client base leave.

The American Booksellers Association had around 12,000 member stores in 1991. Amazon launched in 1995. By 2009, fewer than 2,000 remained. The owners who sold in 1997 to 1999 got fair value. The ones who waited until 2003 sold at a fraction of that… or just closed.

The truth most people don't say out loud: it wasn't bankruptcy that got most of them. It was a slow, quiet slide. Watching competitors quote lower, move faster, serve more clients, until the business they'd built for 20 years was worth half what it should have been. The only real option was to sell it to someone who had figured out what they hadn't.

The frog stayed in the pot. The water kept heating. And by the time it felt urgent, the window had already closed.

What "Integration" Actually Means

Most of the conversation I see on this are getting it wrong.

When people talk about AI adoption…they mean adding a tool. A new app. An AI assistant that summarizes emails. Maybe a chatbot on the website. And they call it integration.

It isn't.

Real integration means every team, every workflow, and every decision point in your business has access to the same context about the business, and AI can act on it. Not one person's assistant. Not one department's experiment. The whole operating layer.

We just met with a boutique agency who is struggling to innovate. Client projects in one tool. Billing in another. Files in Google Drive. Team communication in Slack. Reports rebuilt in a spreadsheet every Friday.

When a client asks where their campaign stands, answering that question requires a few hours and touching four systems!

McKinsey found that workers spend roughly 1.8 hours every day searching for and gathering information. Nine hours a week. Per person!

The answer to that problem isn't another app. It's a connected operating layer that makes the right information available to the right person when it’s needed.

What we're talking about is a CompanyOS: the core operating engine that stores your business's context, your processes, your client history, your team's institutional knowledge, and makes it usable across every team.

AI then becomes the layer that accelerates it. Not bolted on. Built in.

The result isn't just faster responses.

It's faster decisions.

Fewer bottlenecks.

More time for strategy.

More time spent building relationships.

Here’s The Real Problem…

This shift is uncomfortable for most business owners to make.

The same way it was uncomfortable for travel agencies when booking moved online.

The same way it was uncomfortable for Kodak's executives who could see exactly what was coming and still couldn't get the organization to move.

The existing model still works well enough. The team knows how things run. The margin is still there. Disrupting that feels like unnecessary risk.

But the water is warming…

Final Thoughts

The door doesn't slam shut. That's what makes it dangerous.

It closes quietly, while business is still fine…while the team is still functioning…while the revenue is still there.

The frog doesn't notice the water heating up. Until it’s too late.

The businesses taking action now aren't doing it because they love change.

They're doing it in order to be on the growing end…not the closing end.

And, the distinction isn't between businesses that are trying AI and businesses that aren't.

It's between businesses that are experimenting and businesses that are transforming. And right now, transformation comes from one thing: a connected operational engine where AI doesn't assist the business…it powers it.

Time to close? It will vary by industry…but two to three years. That's what I’m reading, based on where we are in the AI adoption curve right now.

The ones who move won't just survive the next wave of disruption. They'll be the ones making it harder for everyone else to keep up.

See you next week,

Damon & Mark

Co-Founders, Modern Operators

This is Issue 55 of Modern Operators. We help founder-led businesses install the operating infrastructure that lets their team run without the founder in the middle of everything.