AI is the new avocado toast—every brand wants a slice. But much like putting avocado on literally everything, not every AI application is smart, useful, or even tolerable. While the promise of artificial intelligence is transforming business, the execution often feels. Intelligent
I run an Executive Network with 725+ members and have observed way too many companies, both brands and tech vendors, using AI incorrectly. Here are five missteps I’ve noticed and how your company can avoid being a cautionary tale.
1. Focusing on Technology and Features Instead of Performance
“Look at our shiny AI!” — Someone who forgot to make it work.
You’ve seen it: websites and product decks boasting about “Llama 3 models,” “AI-powered personalization,” or “hyper-automated workflows.” It all sounds impressive—until you actually try using it and realize it’s marketing speak, not anything that will move the bottom line.
Fix It:
Focus less about the tech, more about the outcomes. AI is a means to an end in the majority of use cases and you should be thinking about the problems you want to solve and how you’re going to measure the success. Make sure you’re asking vendors about average performance and what flexibility you have if the performance level isn’t in line with what you need to see
2. Not Coaching Teams on Change Management
“We rolled out AI! Why is everyone confused and panicking?”
Dropping AI into a workflow without prepping the team is like handing a Ferrari to someone with a learner’s permit. You’ll get resistance, mistakes, and a lot of panicked Slack messages.
Fix It:
Change management is as critical as the tech itself. Make sure you’ve established a clear process internally for how you’re going to implement the vendor, who will be in charge of it, and what’s the process for sharing learnings across the organization.
And if you’re a tech vendor, your content needs to be about the change management, customer success, and training materials you can provide. Don’t assume your clients know how to get the full value out of your platform. Bonus points if you can focus more broadly and add real value without divulging into yet another hidden sales pitch.
3. AI Silos That Don’t Talk to Each Other
Your email AI doesn’t know your product AI exists. It’s basically tech middle school.
Many companies have AI working in different pockets—email, customer service, search—but these systems don’t communicate. It’s like having five different departments each writing their own script for your brand, and none of them have met.
Fix It:
Unify your AI strategy. Connect the dots between systems so data flows seamlessly and experiences are personalized across the board. Try to prioritize vendors that can vertically integrate your stack or at a minimum export your data. It should be your property, not theirs.
4. Ignoring AI Search Engines
Be where your customers are!
AI search engines like Chat GPT, Gemini, and Perplexity are becoming the new front doors to your content. But many brands are still writing content as if Google circa 2011 is the only game in town and are missing how much product and vendor discovery is starting to move to these search engines.
Fix It:
Think AI-first SEO. Use structured data. Write with clarity. Anticipate user intent. And maybe don't title everything "Ultimate Guide to [Generic Term]" like John Roman sees too many times in the vendor emails he posts on LinkedIn
SEO is one of the biggest untapped opportunities for vendor GTM now - I’m seeing it for my clients and hoping the SEO from this article gets me higher in GPT rankings too!
5. Lack of Brand Safety in General-Purpose GPTs
“If you’re selling yoga mats, don’t promote fast food.”
General-purpose AI is powerful—but when left unsupervised, it can go rogue faster than a kid with their parents’ Amazon login. Without controls, you risk brand-unsafe content, tone mismatches, or worse.
Fix It:
Employ human oversight to every AI implementation and look for partners that use closed models instead of open ones if you can. Treat AI like an intern: hard worker but not someone you want to leave unsupervised with the company credit card—or your tone of voice.
Final Thought: Do Less AI Theater, More AI Value
AI doesn’t need to be flashy to be effective. In fact, some of the best AI is invisible—silently making systems faster, recommendations smarter, and experiences smoother. The more you anchor your strategy in performance, cohesion, and human context, the more your brand will stand out in a good way.
--------------------------------------------------------------------------
Stan Rosenberg is the Founder of Storm King Consulting, a firm dedicated to helping MarTech companies and agencies grow through community-building, strategic partnerships, and referral-driven strategies.
He also leads the Storm King Exec Network, an invite-only community of senior leaders in eCommerce and marketing. Members include professionals from top brands such as DVF, Theory, Camping World, and Sephora. The Storm King Exec Network is committed to being vendor-free, completely free for marketers, and focused on peer-to-peer conversations to help members tackle real-time business challenges. The network offers a robust job board, has facilitated consulting and full-time placements, and regularly hosts industry roundtables and in-person gatherings.
Interested in joining the Storm King Exec Network, a free community of 750+ marketing leaders or just jamming about AI? Reach out to Stan@stormkingconsulting.co for more details.
0 comments