Same Job, New Toolkit

MOPs. Marketing Engineering. GTM Engineering. RevOps. Everyone has a new name for the work right now, and a strong opinion about which one is correct.

I think we're too caught up in the titles. The roles have different emphases, but there's a lot more overlap in the actual work than the title debates suggest. If you spend your energy defending a title, you're spending it on the wrong thing.

The job hasn't changed

Strip the labels off and the job is what it's always been: own the data, own the systems, and get a lead from a form to the right rep without it breaking along the way.

That's the function. It was the function five years ago and it's the function now. I keep coming back to this: the role is changing fast, but the function is more important than ever.

What changed is the toolkit

It used to be that being good at this job meant knowing every admin screen in Salesforce and Marketo. Where the setting lived, which checkbox broke which automation, how to make the platform do the thing.

Now it's prompts, APIs, and building agents. I didn't come up through traditional software engineering, and AI helps me write the code. But I'm shipping things that run in production, and that wasn't on the table for a small ops team a couple of years ago. You don't have to become a software engineer to start doing engineering work.

That shift in skills is real. It's also not a different job. It's the same job with a new set of tools.

So if you're worried about staying relevant, focus on skills, not titles. Learn to build a little.

AI takes away the parts nobody wanted

AI isn't making ops less important. It's changing where the effort goes. The repetitive execution gets cheaper, and the value shifts toward designing the systems, defining the rules, and knowing when not to automate.

The best use of AI in ops so far is the boring stuff. List imports. Campaign setup. Checking a file before it goes anywhere. The work that was pure overhead, the stuff between you and the actual campaign. I built an agent for the list import problem and wired campaign setup into a single Slack thread, and in both cases the win wasn't speed for its own sake. It was removing the operational drag around the work so there was more time for the work itself.

Fewer steps between tools, fewer manual handoffs. That's the whole game.

The job moves up a level

Here's the part the titles debate misses. The job isn't disappearing. It's moving up a level.

Less clicking. More deciding what the system is allowed to do on its own, and where a human has to approve.

Just because AI makes the prep faster doesn't mean it should take on the risk. A machine can assemble the campaign. It shouldn't be the one who decides it's good enough to send. Sensitive sends still need a person and a second set of eyes, and the final call stays with someone on purpose. Designing where that line sits is the new version of the work, and it's more interesting than any admin screen ever was.

Building the tool is the easy part

The thing nobody warns you about: building an internal tool is the easy part. Getting people to use it is the hard part.

People will still ask a person for something an agent can do in 30 seconds, because asking a person feels safer. They're not just learning a new tool. They're learning when to trust it, what to check, and when to fall back to the old way. You don't fix that with documentation. Office hours and a few worked examples move adoption more than any doc ever has.

And most of the "AI problems" you'll hear about aren't AI problems at all. They're setup problems. Auth, permissions, the wrong command. Make that part dead simple and most of the complaints disappear.

Treat your workflows like a product

If I had to compress all of this into one habit, it's this: treat your workflows like a product.

Build it. Document it. Watch where people get stuck. Fix that next. Ship, watch, repeat. The same loop a product team runs, pointed at your own operations.

None of this replaces the fundamentals. Clean data and a clear process still beat any shiny tool, and they always have. Lifecycle design, lead routing, attribution, understanding how data moves across the stack. That foundation is what makes any of the new stuff work in the real world. What changed is the ceiling: how much a single person or a small ops team can actually build on top of it.

So pick whatever title you like. Learn to build a little. The work will keep evolving, and the title will catch up.