The AI Marketing Skills Portfolio
Shift investment from channel execution to domain mastery and AI operations
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 94%
Classify marketing capabilities into three directional groups. First, expect routine channel execution in areas such as paid advertising, search, and email to become more accessible as AI raises the performance of competent generalists. Second, increase investment in domain expertise that produces differentiation, including brand building, storytelling, creativity, and innovation. Third, establish AI operations as a foundational capability that connects applications, proprietary customer data, and engagement systems to create tailored customer experiences. The model does not require eliminating channel experts immediately; fragmented channels, immature datasets, regulation, and customer expectations can preserve their value. Its practical purpose is to rebalance hiring, development, and team design toward the technical and creative capabilities least likely to become interchangeable.
Origin
Extracted from Marketing Against The Grain during Kieran Flanagan and Kipp Bodnar's debate about how marketing teams will change in an AI-driven world.
Core principles
- 01AI commoditizes routine channel execution
- 02Brand, storytelling, creativity, and innovation gain value
- 03Technical operators must combine applications with proprietary customer data
- 04Human expertise remains necessary where data, judgment, and compliance are limited
How to run it
- 1
Inventory Channel Execution
List work organized around individual channels such as search, paid advertising, and email. Identify tasks that AI can already perform or materially accelerate.
Pro tip Separate strategic judgment from repetitive channel production before rating each role.
Watch out Do not assume every channel has enough reliable data for autonomous AI execution.
- 2
Identify Differentiating Domains
Map the brand, storytelling, creativity, innovation, and strategic expertise that makes the company distinct. Prioritize capabilities that depend on customer understanding and human judgment.
Pro tip Look for expertise that changes what the company says or creates, not merely where it distributes it.
Watch out Calling content a channel can obscure the domain expertise required to create valuable ideas.
- 3
Build AI Operations Capacity
Develop marketers who can connect AI applications, customer data, and engagement systems. Give them responsibility for building and maintaining small automated experiences.
Pro tip Pair technical fluency with genuine marketing knowledge rather than treating AI operations as pure IT.
Watch out Technical ability alone does not guarantee sound marketing strategy.
- 4
Preserve Essential Human Controls
Assign people to review strategy, accuracy, privacy, compliance, and customer impact. Maintain channel specialists where fragmented platforms or scarce data still demand judgment.
Pro tip Set explicit thresholds for when an automation must escalate to a human.
Watch out Hallucinations and unintended consequences can turn efficient automation into customer or regulatory harm.
- 5
Rebalance the Team Iteratively
Move budget and development time toward differentiated domain talent and AI operators as evidence accumulates. Reassess the balance as channels, models, and regulations evolve.
Pro tip Review the portfolio quarterly rather than committing to a fixed multi-year organization chart.
Watch out The direction of change may be clearer than its timing.
In the wild
A software company audits separate email, paid, and search teams. It automates routine production and reporting, retains one strategist to coordinate fragmented channels, expands its brand storytelling group, and creates an AI operations role that connects campaign data with its CRM. Human reviewers approve claims and privacy-sensitive workflows while the new structure is tested quarter by quarter.
→ The team handles more channels without sacrificing differentiation, compliance, or strategic oversight.
Common mistakes
Eliminating Specialists Too Early
Fragmented channels, small datasets, and immature models can still require deep human expertise. Reduce repetitive execution before removing strategic knowledge.
Hiring Technologists Without Marketing Judgment
A person who can connect systems may still lack the context to design a good customer experience. Evaluate technical and marketing competence together.
Automating Without Controls
Unchecked systems can hallucinate, misuse data, or violate customer expectations. Build review and escalation into every consequential workflow.
Is it for you?
Best for
It is best for marketing leaders redesigning teams or marketers deciding which skills to build next.
Not ideal for
It is not ideal for organizations that lack usable customer data or permission to automate customer interactions.
From the transcript
“So channel and expertise down, domain expertise up,”
“the role that is gonna be critical for every single marketing team in the future is really this AI ops persona who can stitch together…”
“marketing teams are gonna get much more technical over the next one to five years.”
From the episode
5 A.I. Marketing Skills You Must Learn In 2023 (#136)