73% of restaurant brands are investing in AI. Only 9% say it's delivering meaningful results.

Qu surveyed 168 quick-service and fast-casual restaurant brands, about 94,000 locations in all. 73% are investing in AI now or will by the end of 2026. Only 9% say it's delivering meaningful or transformational results so far.

Another 33% say value is emerging, and 43% report limited value. So the biggest group has bought something and is getting limited value so far.

A few cautions on the source. Qu sells restaurant software, the survey covers restaurants only, and the answers are self-reported. I read it as a snapshot of what operators say, which is useful, but nobody has checked it against financials.

AI spend by category

Qu's own breakdown says AI spend concentrates in marketing, CRM and personalization (53%), predictive analytics (40%) and voice ordering (39%). Kitchen automation and AI ordering agents trail at 23% each. My read is that most of the money goes to tools that read data, and less to tools that change what happens in the kitchen.

Top barriers

Qu's report also asks about the guest experience in general, not AI alone. The top barriers are operations at 55% and fragmented systems and data at 37%. I take that as a warning for AI, because an AI tool reads whatever data you hand it. If location 12 counts labor hours one way and location 40 counts them another way, the tool gives you a confident answer built on 2 definitions. Data doesn't lie. Look at the source first, because it's usually someone's definition of a term.

One job, one number

Before buying anything, pick one measurable job: labor scheduling, food waste or ordering. Then pick one number to watch for that job, such as labor as a percent of sales or pounds of waste per week. Write down where the number stands in 3 locations today. Turn the tool on in those 3, leave the rest alone, and compare after 60 days.

If the number moved, roll it out and write down how those 3 locations used it, so location 4 does it the same way. If it didn't, you spent 60 days and a pilot budget instead of a chain-wide contract.

Vendor questions

• Which operators with as many locations as I have are using this, and for how long?
• What number moved for them, and can I talk to one of them?
• What data does it need, and who at my company owns that data today?
• What does it cost per location if the pilot works?

If a vendor can't answer the second question, I'd wait.

If you run multiple locations and want a second set of eyes on where AI would help first, grab a time with me.

Source: Qu, "2026 State of Digital & Beyond" restaurant technology benchmark, March 2026. Figures as reported by Restaurant Revenue Incubator, Aug 24, 2026.