New Federal Reserve research found that nearly 40% of small businesses surveyed were already using or planning to use artificial intelligence. Businesses reported using it for tasks ranging from marketing and customer service to analytics and operational work.
Yet owners also pointed to familiar obstacles: software costs, staff training, system upgrades, and uncertainty over how to implement the technology effectively.
Other 2026 research suggests adoption is accelerating. QuickBooks found that more than three in four U.S. small and midsize businesses regularly use AI, while Clutch reported in August that many AI-using small businesses still have gaps in data readiness, governance, and technology strategy.
For Nick Kurkov, owner of digital agency and software company iTishniki, that creates a problem businesses can easily overlook.
“AI can expose weaknesses in a business that were already there,” Kurkov says. “You add a new tool expecting it to save time, but then someone is copying information between systems, the website does not connect properly with the CRM, customer data is sitting in different places, or nobody knows which version of something is correct. Now you have a faster tool working on top of a messy process.”
He says businesses should examine four areas before adding another AI product.
1. Where does the data actually live?
An AI system is only as useful as the information it can reliably access.
Businesses often have customer information spread across email accounts, spreadsheets, ecommerce platforms, websites and separate software subscriptions.
“If your team already struggles to find the right information, adding AI does not automatically organize it,” Kurkov says. “First figure out what system should be the source of truth.”
2. Are employees repeating work between tools?
One warning sign is employees manually copying the same information from one platform into another.
That can indicate an integration problem rather than a staffing problem.
Kurkov says businesses should look for repetitive steps around leads, invoices, customer requests, reporting and website inquiries before deciding what to automate.
3. Is the underlying software still fit for the business?
A system that worked for a five-person company may become a bottleneck at 20 employees.
Growing businesses often accumulate plugins, subscriptions and temporary fixes because replacing the original setup feels disruptive.
“The expensive part is not always buying software,” Kurkov says. “It is keeping five different tools alive because nobody wants to fix the process underneath them.”
4. Can the business measure whether AI is helping?
AI spending is rising faster than many companies’ ability to measure its return.
Recent Gartner research reported by The Wall Street Journal found that 85% of surveyed functional leaders planned to increase AI spending in 2026, while 23% did not know what return their AI investments were generating.
Kurkov recommends measuring specific outcomes such as time saved, fewer manual steps, lead response time, conversion improvements, or reduced errors.
“Start with the business problem,” he says. “If you cannot explain what should get faster, cheaper or easier after adding the technology, you probably are not ready to buy another tool.”
Sign in to get access to this free resource, and all of our whitepapers and reports.
Download this content today!
Register Now Already registered? Click here to Log In
