Marketing Ops Won't Disappear. But the Job We Built Around It Will.

This is one of the strange things about Marketing Operations and probably one of the reasons the function has sometimes struggled to explain its value.

When good Marketing Ops works, not much happens.

The email gets delivered to the right audience, the customer isn't sent something they shouldn't receive, the form works, the data gets enriched, the lead goes to the right salesperson, the campaign gets recorded correctly and eventually the dashboard gives leadership something reasonably close to the truth.

Nobody celebrates any of that because it is supposed to happen.

Chloe Pott captured this really well in an article that partly prompted me to write this. Her argument is that if Marketing Ops ownership disappears, the technology doesn't necessarily collapse; in many cases it carries on running while the controls around accuracy, attribution, compliance, routing and governance slowly become less reliable, something she describes as silent degradation.

I think that is exactly right, but there is another problem underneath it.

Because we became so good at quietly making the machinery work, we also allowed the value of Marketing Ops to become closely associated with doing the work.

That becomes a problem when somebody looks at the team and sees campaign builds, tickets, imports, integrations and platform administration, because eventually the obvious question gets asked — why do we need a senior Marketing Ops leader when we could just hire more people to do those things?

I've seen versions of that thinking throughout my career and I understand where it comes from. Marketing itself is constantly being asked to prove that the money going in produces something commercially useful coming out, so if the Marketing leadership team doesn't really understand what the operating layer does, it becomes difficult for them to articulate that value upwards as well.

Nobody is necessarily doing anything wrong, but the result is usually predictable.

The Ops team gets stretched, the leader gets dragged into execution, everyone becomes reactive and suddenly some very capable people are spending most of their week launching campaigns, investigating why a lead disappeared, reconciling numbers, fixing data, dealing with a field somebody changed in Salesforce and answering questions about systems nobody else fully understands.

Then because the team looks reactive, the answer becomes more execution capacity. Another specialist. Another agency. Another platform. Sometimes another entire marketing automation system.

And that can help for a while, but it doesn't solve the underlying issue because you cannot keep resourcing your way out of an architectural problem.

Small companies can survive with surprisingly messy operations because humans compensate for the system.

Everybody knows everybody, volumes are manageable, somebody remembers why that field exists, somebody else knows the spreadsheet that actually contains the right number and if a lead goes missing you can message Sales and work it out manually.

Then ambition changes.

The company wants to move upmarket, add countries, launch new products, introduce ABM, run PLG and sales-led motions together, hire more SDRs, create partner channels and give leadership much better revenue visibility.

This is normally where the cracks that were always there become visible.

The natural reaction is often to look at the stack. Maybe HubSpot isn't sophisticated enough. Maybe we need Marketo. Maybe Marketo is the problem and we should go back to HubSpot. Maybe we need another intent platform, another enrichment provider or a better attribution tool.

And sometimes that genuinely is the right answer, but technology cannot make the decisions that actually matter.

This is why I have increasingly thought that the senior Marketing Ops role is misunderstood when we reduce it to technical ownership. Of course the leader needs technical credibility, but their bigger job is understanding what the business is trying to achieve and then translating that ambition into a system that can actually support it.

What should the lifecycle look like, where does Marketing's responsibility end and Sales' begin, what information needs to exist at each point, which systems should own which decisions, what should be automated, where should people still make the call and what happens downstream when somebody changes something upstream?

Those questions sit above any individual platform. And increasingly I think they look much more like product questions.

Mike Rizzo from MarketingOps.com has been talking about this idea for a while and it is one of the more interesting directions I have seen the profession take.

In an Ops Cast discussion, he described the Marketing Ops practitioner as the strategist and key enabler of the GTM technology stack as a product — taking organisational goals, translating them into capabilities, building a roadmap, thinking through user stories and journeys, releasing functionality and then helping the organisation actually use it.

MarketingOps.com has since made that thinking much more explicit in its certification path, moving from Marketing Operations Professional to GTM Product Manager and then GTM Product Architect, with the GTM Product Manager positioned as someone who owns the GTM stack like a product and moves from execution into orchestration.

I don't think the product idea itself is new, but when I connect it with what I've seen in practice, the rise of Marketing Engineering and what AI is now making possible, the shape of the function starts to look quite different.

I like that idea because it changes what we think the product actually is.

The product isn't Marketo. It isn't HubSpot. It isn't Salesforce, Clay or whatever platform becomes fashionable next year.

The product is the capability those systems create together.

Marketing is a user of that product, Sales is another user, Product may be feeding signals into it, Finance consumes outputs from it and leadership relies on it to understand what is actually happening in the business.

Once you think that way the questions become much more useful.

A good product manager doesn't personally build every feature and that isn't really the point of a senior Marketing Ops leader either.

Their job is to understand the problem, understand the users, make the trade-offs, decide what needs building, bring the right people together and make sure whatever gets created produces an outcome the business actually values.

That is a very different way of thinking about the function from being the owner of a marketing automation platform.

The obvious question is where AI fits into all of this.

I think it is going to remove a lot of operational work and I don't see why we should be afraid of saying that.

Research will become faster, campaign production will become faster, analysis will become easier, integrations that used to require specialist support will become much more accessible and agents will increasingly move information between systems, investigate problems and perform sequences of tasks that currently involve several people.

McKinsey's research into AI-enabled marketing organisations is already finding widespread AI use but far fewer companies capturing value across complete end-to-end workflows, which is an important distinction because giving everybody an AI tool is very different from redesigning how the organisation actually works.

That is the part I think companies need to be careful with.

AI doesn't fix the operating model underneath it, it simply gives that model more speed and reach.

This is why the real opportunity for Marketing Ops isn't simply becoming the team that owns the AI tools. It is making sure the business understands what should be automated in the first place, how those automations fit together, where the data comes from, who owns the decisions and what happens when an automated system gets something wrong.

That starts to move Marketing Ops closer to product management, architecture and engineering. Which is where I think the next interesting change happens.

Marketing Engineering has existed for years and MarketingOps.com's original four-pillar model actually included Marketing Development/Engineering alongside platform operations, campaign operations and marketing intelligence.

The engineering pillar covered the more technical work — custom tools, front-end development and integrations — while its newer framework brings technology and data together and explicitly talks about teams containing software engineers, data engineers, product managers, administrators and analysts operating more like a product or technology organisation.

What AI changes is who can build and how quickly they can build it.

A Marketing Engineer increasingly doesn't need to mean someone sitting inside Marketing waiting six weeks to write custom application code. It can mean somebody who understands Marketing deeply enough to identify a problem and technically enough to build the thing that solves it.

That might be an agent that checks campaign briefs before they enter production, a workflow that connects product behaviour to CRM records, an internal application that helps marketers find the right audience, automated data-quality monitoring, APIs connecting systems that were previously handled manually or small tools that remove repetitive work from the rest of the team.

The technical barrier between having the idea and building the first version is getting much lower.

And that is where GTM Engineering becomes interesting as well.

Clay has pushed the term heavily, so it is worth recognising that some of the current narrative comes from a vendor with a clear interest in the category, but the underlying idea is useful: instead of repeatedly performing a revenue task by hand, build a reusable system that performs the task and then measure whether that system improves a commercial outcome.

Clay's own model separates GTM Engineering from core operations in a way I think makes sense. Operations provides the stable foundation — data, CRM, governance and reliability — while engineers build and test new revenue workflows on top of that foundation; once something proves useful, it can be industrialised and governed properly.

That distinction matters because I don't think the answer is to rename every Marketing Ops person a GTM Engineer.

Someone still needs to care about the foundations. Someone still needs to understand the lifecycle, the data model, permissions, compliance, system ownership, reporting and all the boring things that suddenly become extremely important when they stop working.

But alongside that stable core you can start building a much more experimental layer that continuously asks a different question.

I don't think there will be one standard org chart because a 100-person SaaS company and a global enterprise obviously have different needs, but the capability mix is becoming easier to imagine.

The names will change and some companies will combine several of those roles into one person.

What matters more is that Marketing Ops begins to look less like a service desk and more like an internal product and engineering capability for Marketing.

That also changes what I would look for when hiring the leader.

I would still want technical depth, but "Do you know Marketo?" would be quite low down my list.

I would rather give somebody the business model, explain where we want to go over the next three years and ask them what they think will break.

Platforms can be learned. Systems thinking is much harder.

And interestingly, this isn't only a Marketing Ops idea. Netflix's Chief Product and Technology Officer Elizabeth Stone recently described systems thinking as one of the most important skills she looks for in the AI era, precisely because increasingly powerful tools make it easier to create output while making judgment about how the pieces fit together more valuable.

That feels very relevant to where Marketing is heading.

For a long time we maintained a convenient distinction where Marketing created the strategy and Marketing Ops made it happen.

I am not convinced that survives.

You cannot really design ABM without understanding account data, you cannot design PLG without understanding product signals, you cannot personalise intelligently without understanding identity, you cannot deploy AI without thinking about workflows and governance and it is increasingly difficult to talk credibly about Marketing's contribution to revenue if you don't understand how the underlying measurement actually works.

That doesn't mean everybody becomes technical.

Creative instinct still matters, brand matters, product marketing matters, customer understanding matters and some of the very best marketers I have worked with have had an instinct for markets and messaging that no operating framework can manufacture.

But I think the strongest marketers will increasingly understand enough of the operating environment to know how their ideas become real.

And the strongest operators will understand enough about the customer and the business that they are not simply waiting for somebody else to tell them what to build.

The gap between those two people starts getting smaller.

Which is why, strangely enough, I am quite optimistic about Marketing Ops.

AI probably will remove some of the work. It may reduce the number of people required to perform certain tasks and some roles that exist today probably won't exist in exactly the same form five years from now.

I don't think pretending otherwise helps anyone.

But for years very capable Marketing Ops people have spent enormous amounts of time doing things that stopped them doing the work they were actually capable of.

Building another campaign. Moving another list. Pulling another report. Investigating another sync error. Copying information between systems because the systems couldn't do it themselves.

If AI takes some of that away, good. We don't need to defend every task simply because it happened to become part of our job description.

The opportunity is to use that space to move upwards — understand the business, spend more time with Sales and Product, understand Finance, become better at designing operating models, learn enough about AI and engineering to know what is possible and become much better at explaining operational problems in commercial language.

For companies the starting point isn't particularly complicated either.

Before adding another ten marketers, another platform or another layer of AI, follow one customer signal through the organisation and see what actually happens to it.

Where does it enter, which systems touch it, who changes it, who makes decisions from it, where is context lost, where are people manually compensating for a weak process and where have teams quietly created workarounds because fixing the underlying system became too difficult?

That is your real operating model. Then ask whether it can support where the company wants to go next.

If it can't, adding more activity isn't going to solve it. Give somebody ownership of fixing the system.

Marketing Ops spent much of the last 15 years becoming very good at operating increasingly complicated marketing technology, and there was genuine value in that, but I think the next chapter is more interesting because we can start treating the environment as a product, bring engineering much closer to Marketing and use AI to remove work that probably never needed a human in the first place.

For people working in Marketing Ops today, I wouldn't spend too much energy worrying about protecting the old definition of the job.

I would learn the business above the technology.

Learn how the company makes money, understand the customer journey beyond Marketing, learn how Sales actually works, get comfortable with product thinking, experiment with AI and automation, learn enough engineering to build simple things yourself and become the person who can look across all of it and see where the connections are missing.

We have spent years learning how the engine works.

Now, perhaps for the first time, we have the tools and the opportunity to help design the engine itself.

And that feels like a much better future for Marketing Ops than simply getting faster at operating the one we inherited.