Why most operations teams are still stuck experimenting with AI
Over the past year, we kept hearing the same thing: operations teams everywhere are using AI. The buzz was undeniable. But something felt off.
When we talked to operations leaders, we suspected they weren't using AI in ways that actually drove value. So we decided to dig deeper. We surveyed hundreds of operations leaders across industries to understand what "using AI" really meant.
The results were striking. While 98% of operations teams are using or experimenting with AI, when we looked at how they're actually using it, a surprising pattern emerged: most teams are stuck in what I call "pilot purgatory." Only 10% are using AI extensively enough to see real results.
After watching AI transform engineering teams over the last 12 months, operations is next. But there's a massive gap between the teams experimenting and the teams actually winning.
The 10 PM reconciliation that breaks everything
Picture this: It's 10 PM, and your inventory numbers don't match. Again. The ERP says you have 1,247 units of that bestselling SKU. Your 3PL reports 1,089. Shopify shows 1,156 available for sale.
Your choice? Ship orders based on wrong data and deal with angry customers tomorrow, or spend the night manually reconciling three systems that should talk to each other but don't. You choose the all-nighter. Again.
This isn't a horror story. It's Tuesday.
Our research confirms what you already know: 41% of operations teams said data entry and cleaning would deliver the highest ROI from AI—because that's where you're bleeding hours every single day. The spreadsheet juggling, the endless exports, the pivot tables that crash right before the board meeting, the VLOOKUPs that return #N/A when you need them most.
You spend 80% of your time just getting to the starting line. The remaining 20%? That's when you actually get to do the job you were hired for: making decisions that move the business forward.
But what if you could flip that ratio?
The experiment-to-value gap
Nearly half of operations teams are stuck in the middle. They've moved beyond "should we try AI?" but haven't reached "AI is transforming how we work." Only 21% have bridged what I call the experiment-to-value gap.
And here's what should keep you up at night: 68% of operations leaders expect AI to become standard in their industry within 1-3 years. If you're still stuck experimenting while your competitors are implementing, you won't just fall behind. You'll become irrelevant.
The clock is ticking loud: 37% think AI will be standard within the next 12 months.
Why smart teams stay stuck
MIT's Media Lab recently found that 95% of AI pilot projects fail to deliver measurable financial impact. The technology wasn't broken. The approach was. Companies pointed AI at marketing experiments while ignoring back-office operations—where the real savings live.
Our research reveals the real reason operations teams can't escape pilot purgatory, and it's not what you think:
Only 9% cite cost as their biggest barrier. The money is there—71% are increasing their AI budgets this year, with over 20% planning significant increases.
The real blockers are deeper:
- 41% don't trust AI to get it right (accuracy and reliability concerns)
- 36% don't know how to use it effectively (lack of expertise or training)
- 9% worry about budget (cost limitations)
- 4% can't see clear value (unclear ROI)
- 3% can't get leadership support (buy-in issues)
This tells a story: Teams want to use AI. They have the budget for AI. They can see the potential value of AI. But they're terrified it won't work when it matters most—like during that 10 PM inventory crisis.
The barrier isn't skepticism. It's confidence.
The teams that cracked the code
While most operations teams are stuck experimenting, some have moved far beyond that phase. Teams at On Running, Skims, Flexport, WHOOP, and hundreds more represent what happens when you bridge the experiment-to-value gap.
They're not just trying AI—they're living in the future the rest of us are still dreaming about:
Take DeJuan Hall at Seed Health. His team spent two hours every morning verifying reports and cleaning data. Invoice audits meant reconciling 40,000-line vendor files line by line in spreadsheets. Now? Those two-hour audits take five to ten minutes, with discrepancies flagged automatically.
Or Joshua Levy at Magic Spoon, who watched his team get buried under the complexity of scaling from DTC to major retail. Month-end reconciliations that stretched 10 hours now take one or two. The company hosted a "Parabola Day" where 50% of their workforce learned automation, turning it from an IT project into a company-wide skill.
These aren't isolated wins. They represent a fundamental shift:
- 500+ hours reclaimed through automated reconciliation
- Days of manual reporting compressed into minutes
- Problems caught and fixed before they spiral into customer issues
The result? That 80/20 problem finally gets solved. Less time lost to data drudgery, more time spent on strategic decisions that actually grow the business.
They're connecting data sources without CSV chaos, reconciling inventory across systems in minutes instead of days, spotting invoice mismatches before they cascade into errors.
The difference isn't better technology. It's better execution.
The window is closing
The transformation is happening whether you're ready or not. While 98% of teams have started the journey, only 21% have reached the destination where AI becomes a real competitive advantage. The window for gaining advantage is closing fast.
When 37% of your peers expect AI to be standard within 12 months, every quarter you spend stuck in pilot purgatory is a quarter your competitors are building unbeatable advantages.
The future of operations isn't endless spreadsheets and 10 PM reconciliation marathons. It's clarity, speed, and confidence. It's operators working alongside AI to solve the problems that used to steal their nights and weekends.
The question isn't whether your team should try AI—you probably already are. The question is how quickly you can escape pilot purgatory and join the 21% who are actually winning.