Podcast
August 19, 2026
Episode 104: Your RevOps Career Can't Depend on a Tool
Hank Taylor on why RevOps careers can't depend on tool mastery, which GTM platforms will survive AI, and the mindset that keeps operators relevant.
The era of building a RevOps career on deep mastery of a single platform — Marketo, Salesforce, HubSpot — is not exactly ending, but it is narrowing. The operators who thrive in the next cycle won't be the ones who know every click path in their admin panel. They'll be the ones who can look at a problem, forget what tools they already own, and ask: what's the best way to solve this right now?
That's the argument at the center of this conversation. Hank Taylor, a go-to-market advisor, interim operator, and host of the Code to Market podcast, joined co-host Matthew Volm to talk through what's actually shifting in the RevOps skill landscape — which tools are genuinely durable, which are more fragile than the hype suggests, and what a career-building mindset looks like when AI keeps moving the target. Hank brings an unusual vantage point: he got a RevOps title in 2017 before he'd ever met anyone else with one, cut his teeth at companies like GitLab and Vercel, and now consults exclusively with developer-first and AI-native startups. His take on tool dependency, GTM engineering, and what separates operators who stay interesting from those who don't is worth sitting with.
Hank's entry point into operations is instructive. On his second day at his first job out of college, his company's Marketo admin quit. He was ten minutes from Marketo headquarters, where they offered in-person certification. His manager's solution: he was going to that certification next week, and then Marketo was his problem.
That forced ramp taught him something that has shaped his perspective ever since.
"It taught me how important it is to be workflow, and process-oriented rather than tech-oriented."
— Hank Taylor
The problem he's observing now is a scaled-up version of the same trap. Operators who built careers on platform mastery — becoming the Salesforce admin, the HubSpot admin, the Marketo person — were genuinely valuable when those platforms were complex, hard to learn, and sticky. That's not going away overnight. But at smaller, faster-moving companies, founders and executives are increasingly asking a different question entirely: do we really need this tool, or could we build that ourselves?
This connects to a broader pattern that Episode 80: The RevOps Identity Crisis explored — the way RevOps professionals define themselves (by tool, by title, by function) has real consequences for where their careers go. Hank's version of that argument is concrete: if your first qualifying question when considering a new job is "are you a HubSpot shop?" or "are you using this or that tool," you've already narrowed your options in a way that will cost you over time.
Hank sketches three archetypes of operators navigating the current landscape, and the picture he draws is more complicated than a simple "learn AI or fall behind" narrative.
The first archetype is the experienced operator who's stuck. They're often the most capable person in the room — they know the systems cold, they care deeply, and they're impossibly busy. That busyness is the trap: when you're executing within systems you've already mastered, it's very hard to step back and ask whether a different approach might be better.
The second archetype is the new graduate who went straight into GTM engineering. They can spin up a workflow in Claude or Clay before they've ever thought through what problem it's solving. They know how to prompt their way to an output. They don't always know whether the output is right — or whether they're solving the right problem in the first place.
The third archetype — the one Hank holds up as the actual target — is the experienced operator who stayed curious. He cites a friend named Mitchell, a performance marketer turned RevOps operator who's now a GTM engineer at ClickHouse, as an example:
"If you talk to him, he'll tell you most of what I do is still just basic workflow automation. And people think it's AI, but it's actually basic foundational automation that you've always had and that people have always neglected for ten years."
— Hank Taylor
The lesson scales up. If you're new, don't try to reinvent everything with the newest tools — learn the paradigms people have used to solve these problems for years. If you're experienced and entrenched, find the forcing function to look up from the queue and ask what the new tools actually make possible. The operators who do both are the ones getting the interesting work.
This is closely related to what Episode 102: The RevOps Skills Nobody Teaches surfaced — that the durable skills in operations aren't the ones on the tool certification; they're the judgment calls that sit underneath the tools.
Hank's take on Clay is the kind of hot take that ages well or very badly, and he's aware of that. He passed on becoming a Clay affiliate despite being approached with "interesting money." His reasoning is worth unpacking because it's not really about Clay — it's about a category of tools that gave non-technical operators access to technical capabilities they couldn't previously reach.
"I just don't think it's sticky enough."
— Hank Taylor
The specific dynamics he points to: a lot of Clay's user base was leaning heavily on its enrichment capabilities, which happen to be among the most expensive ways to do enrichment. As base-layer data providers have gotten easier to access directly — People Data Labs, Waterfall, and similar — operators are discovering they can bypass the Clay markup and work closer to the source. And on the automation side, code generation tools are getting accessible fast enough that the "no-code" value proposition of platforms like Clay narrows with every Claude or Codex release.
Matthew pushed back, or at least complicated the picture: even if enrichment is becoming commoditized, there's a genuine need for the middleware layer — data orchestration between AI tools and core systems, with proper scrubbing, deduplication, and governance in the middle. That's not going away. The question is who owns that layer and what shape it takes.
Both takes can be right simultaneously: the specific waterfall enrichment revenue may not be sticky, but the orchestration problem it helped solve is real and won't disappear. What changes is which tools get to own that problem. This is a theme Episode 97: Everybody Has AI. Nobody Owns It. examined in depth — when everyone has access to AI tools, and nobody's coordinating how they connect, the governance and orchestration question becomes squarely a RevOps problem.
The broader landscape question is which incumbent GTM tools are positioning themselves durably versus those that are fighting a defensive battle. Hank's heuristic is straightforward: how is a company thinking about its revenue model, and how is it actually giving more power to its users?
His read on Salesforce is blunt. Benioff made a high-profile claim about agentic capabilities, and the real-world answer was underwhelming.
"What can you actually do headlessly? Basically nothing. It was a big nothing burger. It was like, cool, you can look at some stuff. No changes. They didn't even give full API access."
— Hank Taylor
Outreach gets a similar diagnosis. The company built its business on seat-based pricing tied to SDR headcount. When the market started questioning whether to hire SDRs at all — let alone more of them — Outreach's messaging was "AI will enhance SDRs, not replace them." Hank's read: that's a company trying to protect its revenue model, not serve its users.
HubSpot lands somewhere in the middle. It's attempting to be AI-forward, and having a unified database without integration debt is genuinely valuable as agents need clean data to operate. But Matthew's own honest take — that RevOps Co-op's HubSpot looks basically identical to a year ago despite all the Breeze announcements — is a fair signal of how much the AI layer has actually changed daily operator experience.
The winner in Hank's view, at least among the automation layer, is n8n. The reason isn't flashy: it's that they keep adding features that make it easier for operators to use AI directly, they haven't tried to obscure token costs, and their workflow files can be generated directly from a language model and imported. That combination of open access and LLM-compatibility is what Hank sees as genuinely durable.
The pattern he's identifying mirrors what Episode 88: Not Everything Is an Agent mapped out — the tools that survive the AI transition won't be the ones that bolt on AI features to preserve legacy pricing. They'll be the ones that give operators more direct access to capability rather than less.
The career development argument Hank makes isn't really about which tools to learn. It's about the posture you bring to every problem.
His framing: everyone is always at the start of their career, in the sense that if your career is tech-dependent, you can never assume that what you know today will be what the market wants 12 months from now. The learning mindset isn't optional. It's table stakes.
"You have to constantly be asking yourself, 'Okay, this thing I'm about to do — there's the way I know how to do it, or there's the way someone told me they know how to do it. You have to know that. And then you have to be able to ask, is there a new way to do this?'"
— Hank Taylor
One of his favorite interview questions gets at this directly. He asks candidates for their favorite project, then their highest-ROI project, and then — after two success stories — their biggest failure. The failure question is the disqualifying one. Candidates who can't describe a real flop, a project that cost a month of effort and didn't pan out, reveal something: they're not trying anything interesting enough to fail at.
Matthew adds the learn-by-doing dimension. His take is that for operators who might naturally lean toward planning before acting, the current AI moment requires more tolerance for trial-and-error than feels comfortable. Part of that is permission to revisit approaches that didn't work before. Door-to-door sales for a B2B company sounds absurd until it's working because nobody else in the geography is doing it. The same logic applies to GTM and operations experiments — doing things that have been tried before, in a new context, under new constraints, can produce real results.
That's the recurring thread in how RevOps careers actually grow — not by mastering the current toolset, but by staying curious about what's changing and willing to build something that might not work. For more on the career dimension of this, Episode 99: What Executives Actually Want From RevOps and Episode 77: Run Your Job Search Like a Revenue Engine are both worth revisiting.
Check out our blog, join our community and subscribe to our YouTube Channel for more insights.
Our average member has more than 5 years of RevOps experience. That means you’ll have real-time access to seasoned professionals. All we ask is that you’re generous with your knowledge in return.
