The AI hype cycle, and what's actually real for mid-size businesses

There is a new “AI-powered” version of nearly every category of software right now, and a constant stream of demos showing agents doing genuinely impressive things. It's very easy, as a business owner, to feel like you're falling behind by not adopting any of it.

In practice, most of what's trending solves problems that large tech companies have and mid-size businesses don't. A model that can pass a bar exam is not obviously useful to a business whose actual bottleneck is that quotes take four days to turn around because they route through three people's inboxes.

The businesses we see get real value from AI are the ones who resist starting with the tool. They start with a specific, named problem — slow quoting, manual data entry, a support inbox nobody can keep on top of — and then ask whether AI or automation is genuinely the best fix for that problem, not whether AI can be involved somehow.

So our honest advice is usually the least exciting thing you'll read about AI this year: ignore most of the noise, write down the two or three things that actually slow your team down, and only then ask what technology fixes them. Some of those fixes will involve AI. Some won't, and that's a fine outcome too.