Most of the time, trends come and go. They’re fleeting. But sometimes a trend catches fire and burns a bridge to the future.
Sometimes it’s tempting to dismiss a trend as just hot air. Other times, it’s tempting to see if following a trend will light a fire in your own business. Getting it wrong can be both a lesson in restraint or in taking chances.
We asked our Fast Company Executive Board members what’s one trend they were wrong about. Below, 29 of them reveal the trend they got wrong and what they learned from it.
1. NEURAL NETWORKS
I never thought training deep neural networks on language modeling would result in the powerful emergent capabilities we clearly see in large language models today. I missed that due to my skepticism of neural networks. — Babak Hodjat, Cognizant
2. PESSIMISM ABOUT OUR GROWTH
I was too pessimistic about our own growth. A few years back, our addressable market shrank by about 50% because the number of vehicles coming off lease dropped sharply, a delayed ripple effect from COVID three years earlier. We braced for it, figuring we’d have to go into survival mode for a year just to weather the drop. Instead, we grew 50% to 70% that year, exactly like we had every year before, and we’ve done it again since. I was wrong to doubt it. It pretty cool to watch the business prove me wrong. — Zander Cook, Lease End
3. AI MAKING EVERYTHING GOOD, FAST, AND CHEAP
Early in 2024, I thought artificial intelligence broke the “good, fast, cheap, pick two” rule. I thought that we could finally have all three. Time proved me wrong. Every team is still making that tradeoff, whether they’re choosing it consciously or drifting into it. — Jen Jones, Siteimprove
4. TECHNOLOGY’S ABILITY TO SOLVE ORGANIZATIONAL PROBLEMS
One trend I was wrong about was believing technology could solve more organizational problems than it actually can. Technology can accelerate work, improve visibility, and expand capability, but it cannot repair an unhealthy process, overcome weak governance, or create trust. People and process must come first. Transformation begins with clear decisions, sound operating practices, and a workforce prepared to adopt change. Technology should enable that transformation, not be expected to create it. When organizations lead with the tool rather than the human system, they often digitize existing problems instead of solving them. — Britton Bloch, Navy Federal Credit Union
5. THE OPTIMIZATION PLAYBOOK
I bought the optimization playbook for years—best hour to post, the hook formulas, the cadence, whatever format the platform was rewarding that quarter. I ran it well. The numbers moved and none of it was ours. Reach fills a room. It does not tell you who stayed. Now the test is whether a sentence could have come from anyone else, and if it could, it does not go out. Slower, and the numbers are smaller. The messages that arrive at 2 a.m. are from people who read the whole thing. — Natanya Wachtel, New Solutions Network
6. THE ABILITY TO FILTER OUT FAKE REVIEWS AND AI SPAM
I thought platforms and regulators would be further along by now on filtering out fake reviews and artificial intelligence spam. When the Federal Trade Commission finalized its fake review rule in 2024 and the UK’s Competition and Markets Authority pushed Google into tougher policies, I expected the volume to drop. It didn’t. Enforcement reaches the occasional business that gets caught, not the people running review farms and generated content at scale. It’s still on the platforms to catch it, and that’s a hard position to win from. Google took down 170 million fake reviews in 2023, and the problem hasn’t gotten any smaller. — Travis Schreiber, Erase.com
7. PEOPLE WANTING SPEED FROM AI
I used to think speed was what people wanted most from artificial intelligence tools. Less time on execution, more time on everything else. Talking to actual customers changed my mind. What they cared about was style, the specific way they do things that clients hire them for, not just any acceptable result. Speed without that just meant faster, more generic work. If you are evaluating AI for your team, do not just ask how much time it saves. Ask if it protects the thing people are paying you for in the first place. — Yotam Gil, Imagen
8. MOBILE PHONES
I was spectacularly wrong about mobile phones. In 1995, during the SprintPCS network build, Lucent Technologies’ engineers told me everyone would someday carry a cellphone. I held up my pager, pointed to a bank of payphones, and asked why anyone would need one when phones were everywhere. I’ve worked in mobile ever since. It taught me an important lesson: Consumer behavior is the best indicator of where the market is going. Your own opinion is not. — Mack McKelvey, SalientMG & The Credentialed
9. AI AS A SOFTWARE ONLY PROBLEM
I initially thought artificial intelligence would primarily be a software problem, but the biggest constraint is access to high-quality, real-world operational data. Generative AI is only one chapter of AI. Companies creating the most value will be the ones who own or have privileged access to the richest, continuously updated operational datasets. The value of AI is in its ability to optimize systems to reduce energy consumption, lower operating costs, extend equipment life, and improve occupant experience. Ultimately, winners will be companies whose AI continuously learns from real-world operating conditions and improves over time. — Jennifer Nuckles, R-Zero
10. THE INVICIBILITY OF SCALE
One trend I misjudged was the invincibility of scale. During COVID, global production hubs were cut off from consumer markets. Brands closer to customers—with deeper insight and more responsive supply chains—fared far better. In the post‑pandemic world, and amid new tariff regimes, the lesson is clear: Nimble, distributed supply chains have become the stronger strategic asset. — Christopher Williams, Custament Partners
11. AI AS WORK AUTOMATION
I thought artificial intelligence would primarily automate work. Instead, it’s exposing leadership. As AI handles more execution, the remaining work becomes increasingly human: judgment, prioritization, trust, coaching, creativity, and decision-making under ambiguity. Technology isn’t replacing leadership. It’s increasing the premium on great leadership. — Dennis Richter, Fullscript
12. THE AI TAKEOVER OF CONTENT WRITING
In May 2023 I wrote in my private journal that artificial intelligence would be “a clean sweep of writers.” I walked away from content writing because of it. I was wrong in an instructive way: AI killed commodity writing but made a distinct human voice more valuable, not less. Generic content collapsed while opinionated, experience-backed writing became the differentiator. I now run a personal branding business built on the exact skill I had declared dead. — Bhavik Sarkhedi, Ohh My Brand
13. FRINGE BENEFITS
I over-invested in the fringe benefit trend: complimentary coffee bars, snack stations, an employee lounge, parties, and appreciation lunches. All well intentioned, and some added value. I was wrong to think perks were a main driver of culture. The real shift came from appropriate pay, comprehensive health benefits with no employee premium, and clear role expectations so people could turn off when away. “Work-life balance” is an overused and poorly understood phrase. What we built instead is a system that honors employees through compensation while genuinely respecting their need for a life outside of work. — Charles Stanton, Transient, Animal Rights Foundation
14. QUICK COMMERCE GOING MAINSTREAM
I overestimated how quickly Q-commerce would become a durable, mainstream business model. The promise was compelling—extreme convenience, dense urban demand, and technology making delivery almost instant. But convenience alone doesn’t solve difficult unit economics. It reminded me that rapid consumer adoption and a sustainable business are two different things. Now, when I evaluate a trend, I spend less time asking, “Will people use this?” and more asking, “Can this work repeatedly at healthy economics?” — Artem Lalaiants, RiskSeal
15. AI TRANSFORMING THE JOB APPLICATION PROCESS
Initially, I believed that generative artificial intelligence would automatically transform the entire process of resume writing and job application submission. However, I soon discovered that, although AI is good at organizing information, it still lacks the quality of being human. AI-generated resumes are impersonal and one-dimensional. In reality, building successful career capital relies on a person’s self-reflection and storytelling capabilities. — Volen Vulkov, Enhancv
16. AI’S STRATEGIC POWER
I underestimated the strategic power of AI. I mistakenly presumed that it would remain a productivity tool for a lot longer than it has, instead of becoming the strategic advantage that it is now. My lesson here is straightforward: Don’t dismiss technology early even when it’s emerging. Being able to adapt to new technologies early gives your company the upper hand. — Michael Gargiulo, VPN.com
17. THE DEATH OF TRADE SHOWS
I thought tradeshows were permanently on their way out, but their return proved me wrong. It reinforced for me that there is still real value in face-to-face connection and relationship building. For leaders, it’s a reminder to distinguish between temporary shifts in behavior and lasting changes before making long-term assumptions. — Kimberly Ferro, Kawneer
18. THE DEATH OF THE QR CODE
I was convinced the QR code was dead. It felt clunky and like a technology marketers kept trying to force on people. Then smartphone cameras made scanning effortless, and the pandemic taught nearly everyone how to use them. Today, QR codes are one of the simplest ways to connect a physical experience to a digital one. It was a good reminder that sometimes a technology isn’t dead . . . it’s just waiting for the right behavior and infrastructure to catch up. — Greg Aiken, Air Compressor Services
19. HOW LONG IT WOULD TAKE FOR AI TO BECOME PART OF DAILY WORK
Honestly, one trend I was wrong about was how quickly artificial intelligence would become part of everyday work. I assumed it would take years for most teams to really start using it, but suddenly it was everywhere, helping with writing, research, customer communication, and routine tasks. What surprised me most wasn’t the technology itself; it was how quickly people found practical ways to use it. That experience reminded me to stay curious and avoid saying, “This will take a long time.” Sometimes change arrives much faster than we expect. — Ashish Fernando, EDMO
20. THE INFLUENCER ECONOMY
I assumed the influencer economy was a bubble that would pop. I thought it was too unstructured and too dependent on individual personalities to become real infrastructure. I was wrong. It professionalized. It built actual businesses. And it changed how brands reach people permanently. I came around, but I was late. I have tried to remember that the next time I am tempted to dismiss something because it looks unserious. — Kristin Marquet, Marquet Media, LLC
21. THE DECLINE OF GATHERING IN PERSON
I thought technology might reduce the importance of gathering in person. In fact, the opposite has happened. As artificial intelligence accelerates and more of life moves into digital spaces, the desire for real life experiences have become even more meaningful because they satisfy something technology can never replicate—our fundamental need to connect, belong, and share moments together with one another. — Brian Mirakian, Populous
22. RETURN TO OFFICE MANDATES
I was wrong about return to the office. I had predicted that companies would require employees to return to the office full time. Instead, the hybrid culture has become the norm, which I believe is the best of both worlds. — Stefan Evers, Infosec
23. THE POWER OF THE “EDUATIONAL” LABEL
For years I believed leaning into the “educational” label was the safest way to win parents’ trust. I was wrong. Parents aren’t looking for an education technology app. They’re looking for something their kid genuinely wants to spend time on. Once we repositioned ourselves as entertainment first, with learning built in rather than bolted on, everything from engagement to how proud parents feel about screentime changed. — Cristobal Viedma, Lingokids
24. RISE OF THOUGHT LEADERSHIP ADVISORS
I underestimated how quickly employers would embrace thought leadership when choosing advisors. Years ago, I believed referrals and relationships would always be enough. Today, employers want to learn from you before they ever contact you. Publishing educational content, sharing expertise, and building credibility online have become essential parts of earning trust. — Jennifer Schaefer, JS Benefits Group
25. SOCIAL MEDIA’S STAYING POWER
When social media first came out, I said, “This will never last. It’s the CB radio of the ’90s.” I was wrong. It’s the CB radio of the ’90s, ’00s, ’10, ’20s . . . — Bob Sprague, Yes&
26. 1980’S MUSIC AS A FAD
I thought ’80s music was a passing fad. Now it’s getting remixed into rap. Who would’ve ever . . . Leadership can take a lesson from that. That something iconic gets repurposed, not replaced. What’s old is new again? — Nicholas Wyman, IWSI America
27. MORE COMMUNICATION TECHNOLOGY TO CREATE STRONGER CONNECTION
I once believed that adding more technology and communication tools would automatically create stronger connection and collaboration. I have learned that tools only work when people have clear expectations, shared priorities, and consistent follow-through. Technology can support a healthy culture, but it cannot create one. — Angie Hughes, Lets Do Good
28. AI AS JUST AN EFFICIENCY TOOL
I was wrong about AI being just an efficiency tool. I thought its biggest impact would be helping support agents respond faster. Instead, I now believe its greatest value is preventing issues altogether through proactive insights, automation, and better customer self-service. That’s changed how I approach support and operations. — Nandish Dave, BitGo Inc
29. DIGITAL DESIGN TOOLS
I thought digital tools would make design more efficient and therefore give us more time to design. Instead, they often just increased the amount of design we can produce. Faster tools don’t automatically create better work. Judgment is still the scarce resource. — Ryan Woods, Beck Group