A Record 73% of Small Businesses Say Costs Keep Rising — While AI Adoption Jumped to 80%

Flat-vector illustration of a shop owner figure behind a storefront counter with a tall stack of rising cost blocks pressing down on one side, while a friendly robot assistant lifts routine-task icons — a chat bubble, phone, and document — off the counter on the other, on a dark navy background with electric-blue accents

Two numbers came out of the same small-business survey this month, and they belong in the same sentence. In its Q3 2026 Voice of Main Street poll — released September 16 by Small Business Majority, surveying 222 small business owners between August 6 and September 8 — a record 73% reported that their business expenses rose over the past three months, the highest share since the survey began in early 2025 and six points above the previous high. And in the same poll, the share of small business owners not using any AI tools fell to 20%, down from 36% just six months earlier. Costs have never squeezed harder, and AI adoption has never been higher. Those facts are related, and the relationship is the whole story.

This post breaks down what the squeeze actually looks like line by line, why the AI surge is happening under duress rather than enthusiasm, the concerns owners carry into it (73% worry about accuracy, 72% about data privacy), and the one finding most owners have not thought about yet: 85% are concerned about AI being used to decide their loan applications — a process many cannot see or opt out of. Then the practical part: what a squeezed business should actually do this quarter.

The Squeeze: Every Cost Line Rising at Once

The cost numbers are close to unanimous. Among the businesses reporting each expense line, 92% saw raw goods and materials go up this year, 89% saw utilities rise, 89% saw healthcare premiums climb, 69% commercial rent, and 63% wages. There is no line item where a majority of affected businesses saw costs hold steady. This is not one input spiking while others absorb it — it is pressure from every direction simultaneously.

Revenue is not coming to the rescue. Over the past three months, 40% of owners watched revenue decline and 35% held flat; only one-quarter grew. Reduced consumer spending is a challenge for 83% of businesses, 65% have seen profit margins shrink, and 46% have watched customer growth slow. Difficulty making payroll or covering business expenses now touches 73% of respondents, and 63% are struggling to access capital or financing — the highest-stress combination this survey has recorded.

The consequences are already priced in: 68% of owners have raised their own prices this year, 45% have delayed business investments, 21% froze hiring, and 15% cut employee wages or benefits. Healthcare pushed people to more drastic moves — 22% changed their own health plan to save money, 7% dropped their own coverage entirely, and among businesses offering employee coverage, 11% dropped it. Notably, mass layoffs are not the story: only 8% cut workers, and headcount held steady for 56%. Owners are cutting investments and their own compensation before they cut people.

One more number worth sitting with: 72% say finding the right marketing channels is difficult, and 85% say attracting new customers is hard — right as 83% report their customers are spending less. Even optimism is rationed: 46% are optimistic about the next six months versus 34% pessimistic.

The Quiet AI Surge: 64% to 80% in Six Months

Against that backdrop, AI adoption didn't plateau — it jumped. Only 20% of owners now say they use no AI at all, versus 36% in the Q1 2026 edition of the same poll. Flip it: roughly 80% of these small businesses are now using AI in some form, up from about 64% two quarters ago. What they use is mostly off-the-shelf tools from the big vendors — 65% use tools from companies like OpenAI and Anthropic, 33% use AI features built into software they already pay for, 16% use a mix, and 10% built something in-house.

Independent data points the same direction, with an instructive gap. Homebase's 2026 Main Street AI Gap Report (750 decision-makers at hourly-team businesses, published September 22) found 74% now using or piloting AI, with adopters saving an estimated 7.7 hours and $343 a month when AI handles scheduling and payroll. But the U.S. Census Bureau's Business Trends and Outreach Survey — which asks employer businesses what they actually did in a two-week reference window, not whether they "use AI" — put measured AI use at 23.2% nationally for August 10–23, 2026 (California: 25.1%), with 27.3% expecting to adopt within six months. The honest read: "80% have touched AI" and "23% used AI in a given fortnight" are both true, because most small-business AI use is still occasional — a chatbot session here, a built-in feature there — rather than an installed workflow. The market has tried AI almost everywhere; it has industrialized it almost nowhere. That gap is where the remaining advantage lives.

The Voice of Main Street poll itself hints at why adoption accelerated in a cost crisis rather than a boom: when 73% of owners struggle to cover expenses, a tool that answers emails, drafts quotes, or takes phone calls at night stops being a curiosity and becomes payroll relief. The JPMorganChase Institute's bank-transaction data we covered earlier this month showed the same pattern in hard payments — adoption accelerating fastest exactly when margins compress.

What Owners Are Worried About — and the Rules They Want

Nobody is adopting blindly. The concern list is led by accuracy or reliability of AI outputs (73%), data privacy (72%), protection of intellectual property (63%), and security breaches (63%) — then compliance risk (43%), cost or unpredictable pricing (39%), lack of control over vendor updates (39%), staff expertise (27%), and integration difficulty (21%). Notice what is not at the top: cost.Owners' hesitation is trust-shaped, not budget-shaped — the same finding Homebase's report made independently.

Those concerns translate directly into policy preferences. 96% of owners say a national AI law should include data-privacy protections limiting how AI developers collect, store, and use sensitive data — 86% call it very important, and in this survey not a single respondent called privacy unimportant. On structure, 59% prefer a federal floor of minimum protections that states can build on, versus 18% who want one national standard that replaces state AI laws. For a small business, the practical translation is simpler than the politics: assume nothing about how a vendor handles your data, and get it in writing.

The 85% Number: AI May Already Be Judging Your Loan Application

The finding that deserves more attention than it will get: some lenders now use AI to evaluate credit risk and process small business loan applications, and 85% of surveyed owners are concerned about it — more than half (54%) very concerned. This matters double in a quarter where 63% of businesses report difficulty accessing capital. The system that decides whether your business survives the squeeze is itself automating, and applicants have zero visibility into it.

You cannot opt out of algorithmic lending, but you can influence its inputs. Lenders' models read the same public and institutional data everything else does: business information consistency across your Google Business Profile, directories, and website; revenue signals; and your banking record. Contradictory names, addresses, hours, and service areas across listings are not just a marketing problem — they are data-quality noise in every automated evaluation of your business, financial or otherwise. The consistency cleanup we recommend for AI search visibility is, functionally, the same cleanup that keeps your file clean everywhere else.

What a Squeezed Business Should Actually Do

Five moves, sized for a quarter where budget and attention are both scarce:

Pick one workflow, not a platform. The failure mode under pressure is buying three subscriptions and learning none. Choose the single highest-volume repetitive task you personally still do — quote drafting, appointment booking, after-hours phone coverage — and move only that. Homebase's adopters saved 7.7 hours a month on scheduling and payroll alone; one workflow done properly beats a stack of tools sampled shallowly.
Put AI on the expense review it deserves. Cost is 39% of owners' concern list for good reason: subscriptions and token-based pricing add up. Our guide to the hidden AI bill small businesses accumulate covers how to audit that line item so the tool meant to relieve the squeeze doesn't add to it.
Clean your public file like a lender is reading it — because one might be. Name, address, hours, services, and service area identical across your Google Business Profile, directories, and website. Our Google Business Profile checklist for the Ask AI era is the working list; the lending angle is simply one more reason it pays.
Vet vendors with four questions. Where is my data stored? Is it used to train your models? What happens when you change the product? Can a human override every customer-facing action? Those four map straight onto the 73/72/63/39 concern stack. A vendor who can't answer in writing has answered.
Respect bandwidth as the real constraint. The bank-transaction data showed solo operators adopt AI at barely half the rate of employer firms with identical revenue — not for lack of money but lack of time to implement. If that's you, the answer isn't more research; it's delegating the setup. One afternoon of help beats a saved bookmark folder.

The Temecula and Murrieta Angle

Local context sharpens all of it. In a market where 92%-of-respondents-style materials inflation hits contractors, landscapers, and trades first, and where visitor spending cycles seasonally, the businesses holding margin are the ones that converted fixed owner hours into automated coverage — a site that quotes and books at 11 p.m., a phone line that answers during a job. The adoption jump in this poll is Main Street under duress doing exactly that, mostly with $40–150 a month in tools. The owners losing ground are not the ones who chose the wrong AI; they are the ones still pricing, quoting, and answering messages by hand while their costs rose without permission.

The Bottom Line

The Q3 2026 Voice of Main Street poll describes a quarter where a record 73% of small businesses saw costs rise, 40% watched revenue fall — and the share using AI climbed to roughly 80%, with trust concerns (73% accuracy, 72% privacy) and lending anxiety (85%) riding shotgun. The pattern is clear: small businesses are turning to AI as cost relief, and the ones doing it deliberately — one workflow, audited spend, a clean public file, vetted vendors — are getting the savings without the new risks. If you want that scoped for your business, book the free 15-minute call. PepeWebTech builds AI-powered sites and automations for Temecula and Murrieta small businesses, month-to-month with no long-term contracts — cancel anytime — and every plan's pricing is published on our pricing page. Our blog library has over a hundred more guides on getting AI to actually work in a small business.