When AI tools started becoming genuinely useful for marketing work, I had what I can only describe as a déjà vu experience. Not because the technology was familiar, but because the mistakes being made around it were. The overconfidence in new channels. The assumption that better tools eliminate the need for better thinking. The rush to deploy before understanding. I'd seen all of it before — with digital advertising, with social media, with marketing automation, with every wave of new capability that came through the industry in the past two decades. The platforms change. The human patterns around them are remarkably stable.
Over 20 years in marketing gives you something that no amount of technical knowledge about AI can replicate: pattern recognition across cycles. You've watched the hype build, watched the early adopters claim transformation, watched the mainstream rush in without a strategy, watched the correction, and watched the handful of operators who did it thoughtfully emerge from the cycle with durable advantages. I've watched that arc play out four times in my career. AI is the fifth. And the lessons that actually matter are not the ones being written about most.
Positioning Is Still the Foundation
Before AI. During AI. After whatever comes after AI. Positioning is the foundation. Positioning means knowing — with precision and conviction — who you serve, what problem you solve for them, why you're the right answer rather than a dozen competitors, and how to articulate that in a way that resonates with the specific people you're trying to reach. No amount of AI-powered content production, ad automation, or personalization technology overcomes weak positioning.
I've seen businesses deploy sophisticated AI content systems and flood the market with well-written, SEO-optimized, on-brand content that didn't convert. The writing was good. The positioning was muddy. The message reached people who didn't have the problem being solved, or reached the right people with a message that didn't land. AI amplified the volume without improving the signal. More noise at scale.
What over 20 years teaches you is that the marketing that consistently works is clear, specific, and built on a genuine understanding of the customer. That understanding doesn't come from AI. It comes from conversations, from research, from years of watching what causes people to say yes and what causes them to disengage. AI can help you distribute and personalize a message. It cannot manufacture the insight that makes the message worth sending.
AI will help you reach more people faster. Positioning determines whether reaching them is worth anything.
The Best Marketers Were Already Building Systems
Here's something that surprised a lot of people but didn't surprise me: the marketers and marketing organizations that adopted AI most successfully were not the ones who were most excited about AI. They were the ones who had already been building systematic, documented, process-driven marketing operations. They had editorial calendars, documented audience personas, defined campaign frameworks, brand voice guides, and performance review rhythms. When AI tools arrived, these operators had something to connect the tools to.
The marketers who struggled most with AI adoption were the ones whose previous approach depended entirely on individual talent and creative inspiration. They were often skilled — genuinely skilled — but their process lived entirely in their heads. There was no documentation, no repeatable workflow, no structured framework. When they tried to incorporate AI, the tools had nothing to anchor to. The outputs were generic because the inputs were generic.
Over 20 years in marketing taught me to build systems even when individual talent could carry the day, because individual talent doesn't scale and doesn't transfer. AI has validated that lesson more thoroughly than anything else could have.
Timing Still Matters — And AI Can't Feel It
One of the things that separates good marketing from great marketing is timing. Knowing when to push and when to pull back. Knowing when the market is ready for a message it wasn't ready for six months ago. Knowing when a competitor's stumble creates an opening, or when a cultural moment makes a campaign land with unexpected force. This is timing in the broad, strategic sense — not just “send on Tuesday at 10am.“
AI can identify patterns in historical timing data. It can tell you when your audience historically opens emails, when content historically performs, when campaigns have historically converted. What it cannot do is feel the qualitative texture of the market — the mood shift after a major industry event, the sensitivity in a particular moment that makes a usually-fine message suddenly land wrong, or the opportunity created by a gap that just opened because a dominant competitor changed direction.
- Historical performance pattern analysis
- Optimal send-time predictions based on audience data
- A/B test scheduling and result monitoring
What still requires human judgment in marketing timing:
- Reading cultural and market sentiment in the moment
- Identifying strategic windows created by competitor or market shifts
- Knowing when to hold a campaign that's technically ready to go
I've stopped multiple client campaigns that were fully produced and scheduled because something in the market changed that made the timing wrong. That judgment saved those clients from what would have been at best wasted spend and at worst a meaningful brand misstep. No AI tool was going to catch that. A human with market experience and a felt sense of the moment did.
Data Without Interpretation Is Just Noise
Marketing has always had a data problem — not a shortage of data, but a shortage of interpretation. Before AI, the volume of data was already overwhelming most marketing teams. Now AI can generate, collect, and surface more data faster than any team can meaningfully review. The bottleneck has shifted from data collection to insight extraction.
Over 20 years taught me that data tells you what happened. It almost never tells you why, and it rarely tells you what to do next. The most important marketing thinking I've ever done was not in response to dashboards — it was in response to real conversations with customers, prospects, and people who chose competitors. It was in response to pattern recognition across years of seeing what worked and what didn't across different categories, different audiences, different market conditions.
AI surfaces more signals faster. A skilled marketer still has to do the interpretation work that turns signals into decisions. The risk I see in the current environment is that teams use AI-generated insights as a replacement for that interpretation rather than as an input into it. “The AI says this content type is performing best, so we should do more of it“ is not a marketing strategy. It's a performance optimization. Marketing strategy asks why it's performing, whether that performance reflects genuine audience connection or gaming a metric, and whether doubling down serves the long-term brand.
Data tells you what happened. It takes a human with context and judgment to understand what it means and what to do about it.
The Relationship Between Brand and Automation
Here's a tension I've navigated carefully in my own businesses and in client work: the relationship between brand authenticity and automation. Audiences are getting more sophisticated about what feels automated versus what feels genuine. AI-generated content, at scale and without careful curation, starts to acquire a texture — a smoothness and interchangeability — that sophisticated audiences increasingly recognize and distrust.
The solution is not to avoid automation. The solution is to be intentional about where automation appears in your brand experience and where it doesn't. High-volume, lower-stakes touchpoints — certain categories of social content, SEO-oriented blog content, email operational sequences — can handle more automation without meaningfully eroding brand perception. High-stakes, high-relationship touchpoints — proposal content, personal client communications, thought leadership that establishes your expertise — need the unmistakable presence of a human perspective.
After over 20 years of building brands and watching what makes audiences trust or distrust them, I can tell you that the organizations winning in an AI-saturated content environment are not the ones producing the most content. They're the ones whose content feels like it comes from a real perspective. That perspective has to be yours. AI can help you express it more efficiently. It cannot manufacture it.
What Survives Every Technology Wave
Every major technology shift in marketing has eventually arrived at the same equilibrium: the tools become commoditized, the advantage reverts to the operators with better strategy and deeper market understanding, and the organizations that over-invested in the technology without investing in the thinking behind it fall back. Digital advertising. Social media. Marketing automation. Each had its moment where technological early-adoption created advantage. Each is now table stakes where the winners are winning on strategy, not on tool access.
AI is on the same trajectory, and it's moving faster than previous cycles. The differential advantage of being an early AI adopter is already compressing. Within two to three years, most of the capability gaps between organizations will close. What will remain is the gap between organizations that used the AI transition to sharpen their strategy, deepen their customer understanding, and build better systems — and organizations that just added AI tools to existing confused operations.
Over 20 years of watching technology waves in marketing suggests pretty strongly which group will win. It's always the ones who understood that the tool is not the strategy.
