5 min read
Seventy-seven percent of small and midsize businesses now say they use AI regularly. That number, from the 2026 Intuit QuickBooks AI Impact Report, was 48 percent just eighteen months earlier. Read it slowly, because it changes the question you should be asking. For two years the smart-sounding stance was to wait and see. As of this year, if you are not using AI in some real way, you are no longer being cautious. You are in the shrinking minority, and the gap is widening every quarter.
Why This Particular Number Is Worth Trusting
Plenty of AI statistics are marketing dressed as research, so it is fair to be skeptical. This one holds up better than most. The report draws on more than 34,000 survey responses collected across seven quarterly waves, combined with anonymized data from over 5 million businesses on the QuickBooks platform, and the same finding is echoed in the accountant-facing version of the report. That is a large, repeated sample from businesses that look like yours, not a survey of tech enthusiasts. When a number moves from 48 to 77 percent in a year and a half across that many businesses, it is describing a real shift, not a hype cycle.
The Honest Counter-Argument
Here is the pushback, stated fairly: “regularly use AI” is a soft phrase. Someone who asks a chatbot to reword an email once a week counts. So does someone running their whole customer service on it. The headline number lumps the dabbler in with the power user, and adoption is not the same as advantage. A business can adopt AI and see nothing change on the bottom line, and many have.
That objection is correct, and it is exactly why the number should worry the holdouts rather than reassure them. The risk was never that everyone else would suddenly become an expert. The risk is quieter: while you wait for certainty, your competitors are climbing a learning curve you have not started. The dabbler who began a year ago is now the power user. Adoption is not the finish line. It is the entrance to the track, and most of your peers are already on it.
So What Should You Actually Do
If you are in the 23 percent, the goal is not to catch up on everything. It is to start one real thing and let competence compound. Resist the urge to buy a stack of tools, because as we have argued before, you do not need another AI tool, you need to finish setting up the one you have. Pick the single task that eats the most of your week and point one general assistant at it.
The stack that works for a solo owner is deliberately small: one general assistant for writing, planning, and analysis, one research tool when you need cited answers, and at most one specialist tool for the job that costs you the most time. That is it. A realistic monthly spend sits in the range of a few tens of dollars, and the 20-dollar subscription remains the best deal you will get. Above that, the math only works if the tools are directly producing revenue, so add slowly and cut anything that is not paying for itself.
Three Places Almost Any Solo Owner Can Start
If pointing an assistant at one task still feels abstract, here are three that pay off quickly for nearly any one-person business. First, your repetitive writing: the quotes, follow-ups, and replies you type slightly differently every time. Hand the assistant your rough notes and let it produce the polished version you edit, rather than starting from a blank screen. Second, turning one thing into many: a single blog post, testimonial, or recorded call can become a week of social posts and an email, which is far faster than creating each from scratch. Third, making sense of your own documents: drop a contract, a long email thread, or a report in and ask for the summary and the three things you need to act on.
Notice what these have in common. None require a new tool, a subscription beyond the one you already pay for, or a technical skill. They are ordinary tasks you already do, handed to an assistant that does the first draft while you keep the judgment. That is the whole game at the start, and it is enough.
Where the Curve Is Heading Next
The reason to move this quarter rather than next year is that the tools are shifting from answering to doing. For most of AI’s mainstream life, it responded when you asked. Now it increasingly acts on its own across your apps, a shift we covered in our primer on AI agents for solo owners. That matters because the skill you build now, writing clear instructions and checking the output, is the exact skill those more capable tools will demand. The businesses that started early are not just ahead on tools. They are ahead on the habit of working alongside a capable assistant, and that habit is the part that actually compounds.
The One Move for the Next 90 Days
Choose the task you most dread, the one you put off every week. Spend one focused hour handing it to a single AI assistant and refining the result until it is genuinely useful. Do that same task with the tool for two weeks. By the end you will either have reclaimed real hours or learned that this particular task is not a fit, and both outcomes are worth far more than another month of watching from the 23 percent. The window where being early is an advantage is closing. Ninety days from now, 77 will be a larger number, and the only question that will matter is which side of it you are on.
Related reading
- You Do Not Need Another AI Tool. You Need to Finish Setting Up the One You Have
- AI Agents, Explained for Solo Owners: What They Are and Where to Start
- Your AI Tools Can Finally Do the Work, Not Just Answer
Are you in the 77 percent or the 23? And if you are holding out, what is the honest reason? I would like to hear it in the comments.
