In 2017, I started studying for the Marketo Certified Expert exam.

Marketing ops was still relatively new. And I was technical enough to hold my own, but people who specialized in the practice could absolutely run circles around me. 

I remember seriously considering paying to take the certification exam with my own money because I thought that was the move.

That’s how you prove you’re serious. Specialization was how you built credibility.

If I didn’t pick a lane and go deep, was I eventually going to stall out? Would people figure out I wasn’t as skilled as I thought I was?

Halfway through studying, it finally hit me: I don’t like this stuff at all.

Not “this is hard.” Not “this isn’t valuable.” Just “I don’t want to do this every single day for the next decade”.

Walking away felt risky, but it also felt honest.

Over the next few years, the urge to specialize showed up again. During interviews, I’d hear some version of, “You’ve done a little bit of everything, but we’re looking for someone with deeper experience in X.”

It made me wonder if range was actually a liability.

But I read a few of Emily Kramer’s posts on the “Gen Marketer” this week, and it mostly confirmed something I’ve already been feeling. 

It’s not that specialization isn’t important. It’s that it’s not enough by itself anymore.

Emily’s POV is that the marketers who win now don’t just sit in one lane and wait for work to show up. They can go deep, but they can also connect the dots. They can actually run campaigns across teams. They use AI as part of how they work, not as some side experiment.

All of this felt true. Mostly because I’ve been living it myself for a long time.

This is the part I didn’t see coming

I didn’t expect AI to make me feel more confident about being a generalist.

If anything, I assumed it would expose the gaps. All the places I didn’t go deep enough. Instead, it forced me to confront how much context actually matters.

For a while, I was paralyzed by prompts. I thought I needed the perfect input to get the perfect output. That’s a trap. You can burn hours chasing something that never quite lands. It starts hallucinating. It contradicts itself. You close your laptop more frustrated than when you opened it.

What’s changed since isn’t AI. It’s how I use it.

I use LLMs to get from zero to 80 percent quickly. Not to 100. Not to polished. Just to something solid. Then I take it from there.

I’ll ask, would this meet Kyle Lacy’s very high bar as a marketing leader? How would Devin Reed tear this down? Would Jen Allen-Knuth laugh at this as a seller?

AI tools don’t have the context of experience. They don’t know the room. And they definitely don’t know when the math is wrong. I’ve had it spit back charts that looked incredible. Better conversion rates. Numbers that would make you feel really good about yourself. 

Then you double-check it and realize the calculations are way off. If you don’t know what good looks like, you won’t catch that.

I spent most of my early career thinking I should pick a lane. AI might be the thing that proves I didn’t have to.

The range I built, even when it felt chaotic at times, lets me use these tools without getting lost in them. It’s given me the judgment to flag BS and channel creativity when it matters most.

But not everyone wins here

There’s this idea floating around right now that some roles are safer than others. Writing is the obvious one.

The logic sounds simple. LLMs can generate content, sure, but they’re not really creative. They don’t get nuance. So if you’re a great writer, you’re fine.

I’m not sure that’s true.

Craft by itself won’t protect you. What will protect you is taste and knowing what will actually land.

That’s not some fluffy marketing concept. It’s looking at the idea of starting another B2B marketing podcast right now and knowing, without running a survey, how hard that’s going to be to stand out.

It’s also more boring than that.

Last month we postponed a spring CMO event in Jackson. The interest was real. The calendars weren’t. 

Because I’ve negotiated enough event contracts over the years, I knew exactly what actually mattered in the fine print and where we had wiggle room to negotiate.

That saved us real money. That’s what range looks like.

AI won’t fix weak judgment. It just makes it show up faster.

I see this most clearly when interviewing people

Curiosity, creativity, and self-awareness matter more to me now than a perfectly linear resume.

I don’t care if someone has “five years in X” as much as I care whether they can tell me what actually worked, what failed, and what they learned.

You can tell pretty quickly who has done the work and who has only read about it.

I also ask directly about AI. What tools are you using? What are you not using AI for?

Because the people who actually understand these tools know their limits. They know where they still need to think. The people who don’t will talk about them like they’re magic.

We also recently added three open-ended written questions to every job application we post. You can spot AI-written answers immediately. They’re clean. They’re structured. They’re technically correct.

They’re also forgettable. That’s an automatic no from me, dawg.

The candidates who stand out aren’t the ones who sound polished. They’re the ones who can talk through a mistake and explain what they changed. 

They’ve built pattern recognition across different roles. They’ve sat in enough rooms to know what matters.

It’s obvious once you’ve seen enough of it.

Also obvious: More isn’t the answer

Buyers were already skeptical before AI showed up—navigating a “sea of sameness” in B2B marketing content. Now, there’s just more slop in the mix.

There’s also more “good enough.” If everyone can generate something decent in minutes, decent stops being differentiated.

We’re seeing this play out with customer proof.

“A customer saw results” doesn’t cut it anymore. What starts to matter more is specificity. Industry. Persona. Company stage. The reality that actually matters to that buyer.

When you aren’t intentional about proof, it turns into 24/7 fire drills. Sales needs something hyper-specific, and you’re digging through old decks, spreadsheets, Slack threads trying to find a quote that kind of fits. 

Maybe it matches the use case. Maybe it’s close enough. You stitch it together. It works. No one feels great about it. Then you do it again the next week.

Without direction, it’s always a scramble. You’re solving the same problem over and over, and it keeps you busy instead of making you better.

As Emily Kramer pointed out in her guide, speed absolutely matters in a market flooded with AI-generated content. But velocity—speed + direction—matters more.

AI can help people move faster. It can also make the gap between speed and actually doing things that move the needle more obvious.

For the first time, being a generalist doesn’t worry me anymore

I still feel imposter syndrome all the time. That will never go away.

There are moments where I wish I had gone deeper in one lane. Where I wonder if it would feel more legit to point to one thing and say, “That’s what I do.”

But for the most part, legitimacy feels different now. It feels like knowing what I’m actually good at. Knowing where I can go deep. Knowing where I can’t. Not pretending otherwise.

If AI disappeared tomorrow, my range would still matter. I’d still have to sit in rooms with other teams and every marketing function we have. I’d still have to connect the dots.

For a long time, I thought not picking a lane was going to hurt me.

Right now, it feels like the opposite.

Stuff I’m digging this week

• The new UserEvidence homepage—We shipped a V1 last week after reworking the homepage all January. It’s the first version that actually reflects where the product is headed as a multi-product platform. Already working on V2. Check it out and reply with what you like (or what’s confusing).

• Long Strange Trip podcast with Brian Halligan—I know I dunked on launching another B2B podcast earlier in this newsletter, but when Brian Halligan talks, you listen. I love the format. It’s founders and operators talking like humans, not marketers talking about marketing.

• Huckberry’s YouTube channel—Alex Eaton has been in my ear about Huckberry’s marketing for a long time. I finally checked out their YouTube page and got sucked into way too many videos. It’s one of the best I’ve seen. It made me think B2B still hasn’t really figured out YouTube, which probably means there’s more opportunity there than we’re admitting.

Hold up—what does UserEvidence do again?

Most companies have great customers. What they don’t have is a way to activate them.

So the stories, the insights, and the people who would actually help end up scattered across decks, tools, and Slack threads. Things get messy. Quotes get lost. Advocates get overused. Deals slow down.

UserEvidence fixes that.

We give your GTM team one platform to activate your customer base. You can find new advocates, capture customer proof, and publish original research— then use all of it where it actually moves the deal forward.

• Advocacy — Activate your best customers without burning them out.

• Evidence — Verified, believable stories and results buyers actually trust.

• Research — Original long-form research content that shapes your narrative.

UserEvidence helps B2B marketing teams make it easier for customers to show up authentically when trust matters most.