Tech-savvy Head Of Marketing Shares How Marketing Budgets And Teams Must Be Restructured For AI
See how a marketing leader uses AI to scale content, restructure teams, and improve campaign execution, while keeping strategy and positioning human-led.
Simone Morellato
Head of Marketing of vCluster
AI Role: AI significantly enhances technical marketing by increasing efficiency and narrative consistency across various channels.
Leadership Insight: Marketing leaders must balance AI's capabilities with human judgment for strategic messaging and positioning.
Operational Shift: AI adoption in marketing requires shifting to a cross-functional model, enhancing collaboration and efficiency.
Tool Investment: Decentralized AI-tooling budgets empower marketers with flexibility, boosting productivity and innovation speed.
Enablement Role: An AI-enablement role centralizes learning and operationalizes AI advancements, preventing fragmentation in teams.
Simone Morellato is a technical marketing leader and the Head of Marketing at vCluster. He began experimenting with AI early in a previous role and is now using it as a force multiplier within his marketing teams.
We sat down with him to understand how AI is changing his marketing workflows, and his teams. He told us about the new roles and budget items that are must-haves.
Using AI As A Marketing Multiplier
I’m a marketing leader who has built my career at the intersection of deep technical understanding and go-to-market execution, primarily in infrastructure, cloud, and developer platforms. My journey started on the engineering side, where I studied electronics and computer science and worked as a software engineer building storage networking systems. That foundation helped me in marketing technical products because I always understood how they work, not just how to position them.
Want more from The CMO?
Sign up for a free membership to complete reading this article:
Have an account? Log In
Over time, I moved from engineering to technical product marketing at Cisco, and then into leadership roles at startups and enterprise companies, including Riverbed, Cloudian, and Apcera. My time at VMware was a major turning point in my career, and I spent close to a decade scaling the Tanzu Kubernetes business. I built and led a global team spanning product, developer, and solutions marketing, and helped expand the portfolio from a single open-source project into a suite of more than 20 products. While there, I experimented early with AI-driven workflows, building internal automation tools for analyst relations that significantly reduced cycle time and improved efficiency.
I now lead marketing at vCluster, building a full go-to-market engine across growth, product marketing, community, and brand. We improved efficiency by rethinking demand generation, scaling awareness from tens of thousands to millions of impressions, and significantly improving cost efficiency and funnel conversion. A key focus has been building developer-first narratives and community-led growth loops, particularly in the Kubernetes and AI infrastructure space.
Everything I have worked on converges at this moment in AI transformation. AI is now fundamentally changing both how software is built and how it is marketed. As a marketing leader, I focus on helping companies move from fragmented product messaging to unified platform narratives, using AI not just as a topic in the story but as a multiplier in how marketing is executed.
Balancing Developer Trust With Enterprise Demand
At vCluster, I oversee a full-stack, go-to-market function spanning growth, product marketing, brand, community, and events. The team operates across both open-source and enterprise motions, creating a dual mandate: Drive broad developer adoption at scale while also converting that engagement into a structured enterprise pipeline.
Organizationally, the function operates as an integrated system rather than isolated disciplines. Product marketing and developer relations work closely to shape the narrative and build technical credibility, while growth and content focus on scalable acquisition and distribution across channels such as social, SEO, paid media, and community-led programs. Brand and events extend that reach into ecosystem engagement, particularly within the Kubernetes and cloud native communities.
The complexity comes from the audience and ecosystem dynamics. We are marketing to highly technical personas, primarily DevOps and platform engineers, while also influencing enterprise decision makers. That requires balancing technical credibility with a clear articulation of business value. At the same time, we operate in a fast-moving, AI-influenced infrastructure space where category definitions are still forming. As a result, a significant part of the role is shaping the narrative, not just amplifying it.
The model is effective because a unified acquisition-to-activation-to-engagement loop ties everything together, supported by a strong data-driven approach to performance and rapid experimentation.
Where AI Ends And Judgment Begins
I use AI primarily for decision support and to accelerate work in high-volume, pattern-driven areas, or in those requiring rapid synthesis of large amounts of unstructured information. At the same time, I maintain a clear boundary where human judgment remains essential, especially for context, accountability, and long-term brand implications.
I rely heavily on AI for three areas:
- First, narrative development and message exploration. When shaping positioning or campaign themes, I use AI to quickly test variations, identify messaging gaps, and pressure-test clarity across personas such as developers, platform engineers, and executives. It compresses what used to be multiple iterations into a much faster loop.
- Second, content and campaign scaling. AI now powers the first draft of most content assets, including social content, blog structures, and campaign variants. It also adapts technical inputs into persona-specific language, which is especially important in infrastructure and AI platform marketing, where the same concept needs to resonate differently with different audiences.
- Third, performance synthesis. I use AI to analyze campaign performance signals across channels and surface patterns that might otherwise be missed, particularly when dealing with fragmented data across organic, paid, and community channels.
Explicitly human tasks involve strategic tradeoffs and accountability. Positioning decisions, category definition, and core messaging architecture remain fully human-led because they require a deep understanding of market timing, competitive dynamics, and company ambition. The same applies to final creative judgment on brand voice and high-visibility campaigns.
Positioning decisions, category definition, and core messaging architecture remain fully human-led because they require a deep understanding of market timing, competitive dynamics, and company ambition.
I also maintain human ownership of prioritization decisions. AI can surface options and insights, but deciding what not to do remains a leadership responsibility, especially in fast-moving categories like AI infrastructure, where it is easy to overreact to signals.
So, AI effectively expands the surface area of thinking and execution, but it does not carry responsibility for outcomes. Marketing leadership still makes the final calls that balance speed, coherence, and long-term brand equity.
How One Structural Shift Took Impressions From Thousands To Millions
I shifted our marketing model from human-first content production to AI-assisted narrative and asset generation. Humans now focus on direction, validation, and distribution instead of creating from scratch.
Practically speaking, I introduced an AI-driven content system. We use LLM-based workflows to turn a single technical input — such as a product release, a demo, a customer insight, or an engineering narrative — into multiple downstream marketing assets. This includes developer-focused posts, long-form content, social variations, campaign messaging, and even early-stage positioning drafts. Instead of separate authorship for each, a shared prompt and structure layer now generates, refines, and aligns all assets.
AI generates these in parallel at a strong draft level, not sequentially from scratch. We then add human review and refinement, in which product marketing ensures narrative integrity, growth optimizes distribution, and brand maintains voice and consistency.
And we moved from a campaign-based production cycle to a continuous content system. Most marketing organizations still operate in a traditional campaign mindset: You define a campaign, create a set of assets, launch, and then move on. That model was built for a world where content production was expensive and slow. Now, we run parallel generation and iteration loops. This reduces cycle time and allows us to respond to product and market signals much faster.
How AI Benefits Production Efficiency And Narrative Consistency
On the positive side, the most immediate impact of AI has been scale and efficiency. Introducing AI into content and campaign production workflows significantly increased output without increasing headcount. At vCluster, this translated into scaling from roughly 36K to about 7M monthly impressions, while also reducing CPM from around $5.58 to $0.29. While multiple factors contributed, a meaningful driver was producing and iterating content far more quickly and consistently across channels.
We also improved execution speed. Content cycles that previously took weeks — especially for cross-functional assets involving product marketing, brand, and growth — were compressed into days. This allowed us to respond to product releases and market signals in near real time, which is critical in a fast-moving AI ecosystem.
Another positive outcome was narrative consistency. AI helped standardize how we translate technical inputs into messaging across channels, reducing fragmentation between developer-focused content, sales enablement, and demand generation campaigns. This alignment improved both the quality of engagement and downstream funnel performance.
Qualitatively, one of the biggest shifts was organizational. The marketing team moved from a primarily execution-heavy function to a more orchestration-focused role. Humans now spend more time on direction, strategy, and validation, while AI handles first-pass production and variation. This has improved team focus and reduced operational bottlenecks.
How AI can lead to over-generation
Challenges also emerged. Early on, we saw a tendency toward over-generation. AI produced too many variations without enough initial constraint, creating noise and requiring us to build stronger editorial and prompt governance layers. We also actively managed the risk of message dilution, ensuring speed did not come at the expense of sharp positioning or category clarity.
Another learning was that AI can amplify both good and bad inputs. When the underlying narrative or brief is weak, AI quickly amplifies that weakness. This forced us to invest more in upfront strategy and tighter input design before generating anything at scale.
Overall, the net result has been clearly positive, but only after establishing clear guardrails.
Why AI cannot deliver on differentiation and positioning
AI has also underdelivered on early expectations for strategic differentiation and positioning.
Initially, we assumed AI could meaningfully help generate stronger category narratives or even identify positioning angles to give us an edge. In practice, we found that, while AI excels at remixing existing market patterns, it struggles to produce truly original or defensible positioning.
It tends to converge toward the median of what already exists, which is the opposite of what you want when trying to stand out in a crowded infrastructure or AI platform market.
Why marketing leaders must constrain AI first
An early AI-assisted campaign we ran to reposition a technical platform for a broader DevOps and platform engineering audience provides a clear example of where AI fell short in providing strategic differentiation.
We initially used AI to generate positioning angles and messaging variations based on our product capabilities, customer use cases, and competitive landscape. The output was fast and broad. It provided multiple directions, such as “simplifying Kubernetes operations,” “improving platform efficiency,” and “accelerating developer productivity.”
None of these angles truly differentiated us. When we pressure-tested them against the market, we found that adjacent tools were already heavily using them in the Kubernetes and cloud-native ecosystem. AI effectively amplified common industry language rather than helping us break away from it. If we had moved forward as generated, the campaign would have blended into the category instead of shaping it.
We significantly tightened the human-led input phase. Instead of starting with broad product capability prompts, we imposed sharper constraints on three areas: our unique architectural advantage, the trade-offs we were willing to own publicly, and the audience segments we explicitly wanted to lead with, not just include.
Once we re-anchored the inputs in those strategic choices, we used AI differently, not to invent positioning, but to stress-test clarity, translate the message across personas, and generate execution-level variations for content and campaigns.
The outcome was a much sharper narrative that centered on a more specific and defensible angle: how platform teams could rethink infrastructure abstraction for modern workloads, rather than generic efficiency or productivity themes. That shift improved the quality of engagement and made the messaging more memorable in technical communities, especially among Kubernetes practitioners, who are very sensitive to overly generic positioning.
So, the bottleneck is no longer production quality or speed; it is originality and sharpness of thinking. Standing out now requires a much stronger point of view, clearer positioning, and a willingness to take explicit trade-offs. Without that, AI tends to pull everything toward the market average.
How AI Changes The Structure Of Marketing Teams
AI fundamentally shifted my marketing team from a functionally siloed one to a more fluid, cross-functional execution model.
Traditionally, we clearly separated roles. A content marketer would own content creation, then hand it off to distribution specialists for adaptation, while product marketing would separately develop messaging and positioning that then needed to be translated for campaigns. Each step involved rework and translation between functions.
With AI in the workflow, the structure has changed significantly. Now, every team member can operate across multiple parts of the content and campaign lifecycle. A content marketer can generate not only the initial draft of an article, but also the first versions of social variations, email copy, and even positioning angles. A growth marketer can take the same input and immediately generate campaign variants or testing frameworks. Product marketing can move from writing everything from scratch to refining the narrative, ensuring accuracy, and aligning messaging with strategy.
The key shift is that we no longer create work sequentially from zero. AI generates it to a high-quality draft stage by default, and then we hand it across functions for refinement, validation, and optimization. This has significantly reduced friction between teams and shortened iteration cycles.
As a result, marketers now focus more on judgment, editing, and strategic direction rather than pure production. The team is smaller in terms of manual-execution dependency, but more leveraged in terms of output capacity.
AI also made collaboration more dynamic. Instead of rigid handoffs, we now operate more like a shared system where we can transform any input into multiple outputs across channels, and the relevant specialist then refines them.
Why Marketing Teams Need Decentralized AI-tooling Budgets
If I were starting over, I would first invest the marketing budget not in traditional campaign spend or even headcount, but in giving every marketer direct access to a flexible, decentralized AI-tooling budget.
I would allocate per-person budgets for marketers to adopt and experiment with tools that improve their day-to-day productivity, without waiting for centralized procurement or IT approval cycles. This includes AI writing tools, data analysis assistants, creative generation platforms, automation tools, and niche utilities that emerge quickly in the AI ecosystem.
The reason is simple. In an AI-driven environment, the pace of tool evolution outpaces enterprise procurement cycles. Centralized access inevitably standardizes teams on a stale stack, which slows experimentation and creates a gap between what is possible and what is actually used in execution.
I have seen this firsthand when introducing AI into marketing workflows. The teams that had the freedom to test and adopt tools early moved significantly faster in content production, campaign iteration, and data analysis. Teams constrained by centralized approval processes often reverted to slower, legacy workflows, even when better tools were available.
By contrast, a distributed tool budget creates a different dynamic. It turns every marketer into an operator who can continuously optimize their own stack. It also surfaces innovation organically because the best tools naturally spread across the team through use rather than by mandate.
Second, I would prioritize building a lightweight enablement layer around that freedom, including shared best practices, prompt libraries, and internal workflows to prevent experimentation from becoming fragmentation.
Central control optimizes for consistency, but distributed enablement optimizes for adaptation. And adaptation is what creates competitive advantage in fast-moving marketing environments.
Central control optimizes for consistency, but distributed enablement optimizes for adaptation. And adaptation is what creates competitive advantage in fast-moving marketing environments.
How Claude Cowork Creates Real Leverage
I’m particularly excited about one AI feature from Anthropic: Claude CoWork.
It moves beyond the traditional chat interface and feels like a true working environment for knowledge work. Instead of isolated prompts, it enables more structured, persistent workflows. This is exactly where AI offers real leverage.
I also appreciate how quickly Anthropic iterates based on real user behavior. They pay attention to what the community builds and uses, then integrate those patterns directly into the core product. This reminds me how fast concepts from projects like OpenClaw made their way into Claude CoWork. That feedback loop matters because the tool evolves to reflect how people actually work, not just how the product team imagines they should.
Practically, it also reduces the need to stitch together multiple tools or extensions. I don’t have to recreate workflows, evaluate new vendors, or worry about security and data handling across a fragmented stack. I already trust Anthropic as a company, so when they release new capabilities, I can adopt them quickly and confidently.
Why Every Marketing Team Needs An AI-enablement Role
With that said, the biggest gap I’ve seen in AI adoption is not tooling, but the absence of a dedicated internal capability that continuously tracks, translates, and operationalizes AI changes within the marketing team.
Most organizations assume that if teams have access to AI tools, adoption will naturally follow. In practice, awareness and synthesis are the limiting factors. The AI landscape moves so fast that no individual marketer can realistically keep up with new models, tools, workflows, and best practices while executing their core responsibilities. When everyone is partially responsible for staying current, nobody is fully responsible for it.
In my experience, creating a dedicated AI enablement role within the marketing organization has worked much better. This is not just a tooling admin or a prompt engineer; it is someone deeply curious about AI who actively follows developments across models and applications and continuously translates them into practical workflows for the team.
This person effectively becomes an internal intelligence layer. They evaluate new tools, test use cases, synthesize what is useful versus noise, and then operationalize it into repeatable workflows. Equally important, they act as a two-way bridge. Team members can bring them problems or inefficiencies, explore AI-based solutions, and bring back structured recommendations.
This role primarily impacts focus. It removes the cognitive load of “keeping up with AI” from individual contributors, centralizing it into a single function whose only job is to stay ahead of the curve and effectively distribute that knowledge. Without this, teams tend to fragment. You get uneven adoption, inconsistent practices, and duplicated experimentation. With it, you create a compounding effect in which learning centralizes while execution distributes.
Why Marketers Must Treat AI As An Operating-model Shift
The most important advice I would give marketing leaders is to treat this moment as an operating model shift, not a tooling upgrade.
It is easy to get pulled into evaluating tools, testing features, and chasing incremental productivity gains. That's important, but it doesn't provide the real advantage. The real shift is in how marketing work gets done, how teams are structured, and where human effort is applied.
A few principles prove particularly important:
- First, invest in input quality before output scale. AI generates as much content as you ask it to, but the quality of that content depends entirely on the clarity of your positioning, audience definition, and narrative. Leaders who skip this step scale noise.
- Second, redesign workflows, not just tasks. Instead of asking how AI can make an existing step faster, look at the entire system from input to outcome and rethink its operation in a world where drafts are instant, and iteration is cheap. That unlocks order-of-magnitude improvements.
- Third, create space for focused AI ownership. Whether it is a dedicated role or a small group, someone must stay on top of the landscape, test tools, and translate that into practical workflows. Without that, teams fragment and adoption stalls.
Follow Along
You can follow Simone Morellato on LinkedIn and keep up with his work at vCluster.
More expert interviews to come on The CMO Club!
