Most marketers get far more value from sharpening how they think, prompt, and evaluate than from learning to code.
Some marketers, like the team behind Nike’s Halloween-themed AI ad, are already using AI to produce work that feels genuinely creative. As someone who typically frowns on the use of AI because of how … weird … it looks, even AI ads’ style and direction blew me away.
However, other teams are still finding their footing. Follow along as I break down the AI skills marketers need today, the human skills AI still can’t replace, and how to build both using HubSpot’s Loop Marketing framework — Express, Tailor, Amplify, and Evolve — as a practical structure for putting these skills to work.
I’ll also pull in fresh data from the 2026 State of AI in Marketing report to show how these skills play out at scale.
Table of Contents
- Build and Apply AI Skills Through the Loop Marketing Framework
- Skills AI Can’t Replace
- Top Skills Marketers Need to Work Effectively With AI
- How to Build AI Skills as a Marketer
- Develop the right skills, and AI becomes a tool you direct.
- FAQs About AI Skills for Marketers
Build and Apply AI Skills Through the Loop Marketing Framework
Knowing the skills is one thing, but over the years I’ve learned that applying them consistently is another. According to our State of AI in Marketing report, 98% of marketing teams are using AI in some way, but only 10% have reached a transformational level. Consistency is key if you want to see major, transformational results.
That’s where Loop Marketing comes in: HubSpot’s four-stage framework for marketing in the AI era: Express, Tailor, Amplify, and Evolve. Instead of a one-time skill checklist, Loop gives you a repeatable workflow for putting your AI skills, technical and non-technical, into practice, evaluating how well they’re working, and sharpening them over time.
Here’s where those skills show up at each stage.
Express: Define brand, audience, and point of view.
Express is where you document who you are before AI (or anyone else) defines it for you: your brand’s POV, your ideal customer, and the problems you want to be known for solving. The express stage relies almost entirely on non-technical skills.
Intellectual curiosity pushes you to dig into why your audience responds to certain messaging instead of copying a generic playbook. Creativity and holistic understanding help you connect your brand’s story across content, design, and product in a way that reads as consistent rather than assembled.
Best for: Marketers setting brand guidelines, messaging frameworks, or AI style guides for the first time.
Tailor: Use customer context and data responsibly
Tailor is where you personalize content and offers using customer data — and where prompting skill becomes central. A vague prompt produces generic personalization; a specific one, built with real audience context, produces something a customer actually recognizes as relevant to them.
The Tailor stage also demands empathy: data can tell you what a segment does, but empathy is what tells you how to talk to them. Objectivity matters here, too, since it’s easy to let a dataset’s built-in assumptions pass as facts rather than as one interpretation among several.
Pro tip: Before you personalize at scale, write out the assumptions your segmentation is making. If you can’t name them, you’re not ready to tailor responsibly.
The State of Artificial Intelligence in 2026
New research into how marketers are using AI and key insights into the future of marketing.
- Marketing AI Tools
- Practical Tips
- Trends and Statistics
- And More!
Download Free
All fields are required.
Form not available
Amplify: Adapt and distribute content across channels and AI search.
Amplify is about getting tailored content in front of the right people, on the right channels — including AI search and LLM-driven discovery, not just traditional web search. Adaptability is the core skill: testing new platforms and formats as they emerge instead of waiting until they’re proven.
Strategic thinking keeps that experimentation tied to a goal, so a new channel test serves brand recall or lead generation rather than distribution for its own sake.
What we like: Treating each new AI feature or channel as a small, contained experiment rather than a full rollout, because you learn fast without betting the whole campaign on it.
Evolve: Evaluate results, experiment, and improve.
Evolve closes the loop: reviewing what worked, questioning why, and feeding that back into the next cycle. Critical thinking is essential here in the form of pausing on a result that looks too good and asking what else could explain it.
Emotional intelligence and people management matter just as much, since evolving often means telling a team a campaign underperformed and working through why together, rather than jumping to conclusions alone.
Pro tip: Build a short reflection step into your workflow after every campaign, even five minutes of “what surprised us and why” compounds over time.
Skills AI Can’t Replace
The skills below direct AI rather than compete with it. I encourage you to think of them as prompts you need to use to guide AI well, evaluate what it gives you back, and improve on it before it reaches a customer.
Creativity and strategic thinking tend to show up earliest, in the Express stage, shaping what your brand says and why.
Empathy and emotional intelligence do their heaviest lifting in Tailor, where AI output has to land as human rather than generic. Critical thinking and people management carry the most weight in Evolve, where results get reviewed, and teams adjust.
In practice, all six show up everywhere in the loop, as it’s where they tend to matter most.
Critical Thinking
Critical thinking is what keeps you grounded when a dashboard looks too good to be true.
LinkedIn’s 2024 Future of Work Report lists it among the top five most in-demand human skills worldwide, and AI hasn’t changed that. If anything, AI has made the skill more urgent. An AI-generated performance summary can tell you that conversions dropped 12% last week without explaining why.
That’s on you to dig into: Was it seasonality? A tracking issue? A competitor promotion? Tools can generate the result. Reflection is what turns it into insight.
Loop stage: Evolve — this is the skill that keeps your post-campaign reviews honest.
Empathy
Empathy is the difference between AI output that talks at an audience and content that connects with them. Dove’s Real Beauty Sketches campaign went viral because it told real stories about confidence, not because it was polished.
That same instinct applies when you’re editing AI-drafted copy today: an AI tool can write a technically correct customer email, but it takes empathy to catch when that email is right in substance but wrong in tone.
Think of a churn-prevention email that reads more like a warning than a genuine check-in.
Loop stage: Tailor — empathy is what keeps AI-personalized outreach sounding like it came from a person, not a segment. Read every AI draft out loud before sending. If it doesn’t sound like something a person would actually say to another person, empathy is the gap.
Emotional Intelligence
Emotional intelligence is what you fall back on when a campaign doesn’t go as planned. When Microsoft’s CEO Satya Nadella rebuilt the company’s culture around empathy and listening, Microsoft’s market value tripled — proof that EQ scales even in large, technical organizations.
Dominik Maka, head of SEO at LVBET, puts it well: AI can’t sense the sigh of relief from a client whose problem just got solved, or the excitement in a teammate’s voice over a breakthrough idea; these are the gaps empathy and emotional intelligence fill.
The same applies to AI rollouts on a marketing team.
When a new AI tool changes someone’s workflow overnight, emotional intelligence is what helps you notice the frustration behind a “this is fine” and address it before it becomes resistance to every tool that follows.
Loop stage: Tailor — EQ is what keeps AI-personalized outreach from feeling cold, and keeps your team steady while you figure out how to get it right.
People Management
People management is about giving your team enough context to make their own decisions; AI can’t make them for them. If you’re the one reviewing every AI-generated asset before it ships, quality depends entirely on your bandwidth.
Share the “why” behind your brand’s AI guardrails, and your team can catch a bad AI output — or push back on an over-cautious one — without you in the loop on every single piece.
Loop stage: Evolve — this is the skill that lets a team improve its AI workflow without every fix having to go through one person.
Creativity
Spotify’s Wrapped campaign wasn’t a big-budget concept but a simple insight (people love seeing themselves in data) executed well, driving millions of shares. AI can now generate dozens of variations on a concept like that in minutes.
The creative skill isn’t coming up with the first draft anymore, but recognizing which of those AI-generated variations actually has a spark worth building on, and which are just noise.
Our report shows that 23% of surveyed marketers say AI improves creativity, but Andrew Boyd, co-founder of Forte Analytica, points out that AI can out-produce humans at scale on content, but it can’t offer an authoritative, real-world take on a product or service.
Prompted to write from someone’s perspective, it still tends toward skin-deep, sweeping generalizations because it hasn’t lived through the experience it describes.
Loop stage: Express — this is the skill that keeps your brand’s POV from sounding like everyone else’s AI output.
Strategic Thinking
Strategic thinking connects what AI is automating today to where you’re trying to go. An AI scheduling tool can post to social on a fixed cadence all week without you touching it.
Still, that automation is only valuable if it’s mapped to an actual outcome, like brand recall or lead generation, rather than running because it’s easy to set and forget.
Strategic thinkers are the ones asking what each automated task is actually supposed to accomplish before they turn it on.
Loop stage: Express — strategy is what determines what’s worth automating in the first place, before Amplify puts it into motion.
Top Skills Marketers Need to Work Effectively With AI
Intellectual Curiosity
AI rewards curiosity more than speed. Curiosity is what pushes you past a generic first output to ask why something works, and that mindset turns every new tool or trend into a chance to learn and improve outcomes.
Take a social media manager using an AI tool to build a content strategy. They might start with a generic plan for their industry. But instead of applying it as is, curiosity prompts them to ask: Why does this format work?
How can we tailor it to our audience or product? Breaking down each element — platform choice, content type, posting cadence — uncovers insights that make the strategy more relevant and effective.
Pro tip: Once a week, explore something completely unrelated to your role, such as a podcast, book, or tool outside your niche. Curiosity compounds when you draw connections others don’t see.
Loop stage: Express — curiosity is what shapes a sharper, more specific brand POV instead of a generic one.
Adaptability
New AI tools roll out every month, and automation platforms update their rules constantly. Adaptability isn’t optional, because the most effective marketers are those who experiment early and adjust quickly, rather than waiting for a tool to be “proven.”
When Canva launched Magic Studio and ChatGPT introduced a campaign idea generator, the teams that tested these features first learned to use them strategically rather than reactively.
Pro tip: Each month, test one new tool or workflow in your process. You don’t have to adopt it long-term; you just need to see what it changes in how you work.
Loop stage: Amplify — adaptability lets you test new channels and formats before they become conventional wisdom.
Objectivity
Data doesn’t speak for itself. Every dataset reflects human choices: what’s included, what’s ignored, and what gets defined as “correct.”
Google’s chief decision scientist, Cassie Kozyrkov, illustrates this with a Cat vs. Not-Cat example: an AI model sorting six animal images will call most of them easily, but whether a tiger counts as a “cat” depends entirely on a choice the model’s creator made rather than an objective fact.
As Kozyrkov puts it in her piece on AI and decision-making, machine learning’s “right” answers are usually in the eye of the beholder, so a system designed for one purpose may not work for another.
Blake Burch, co-founder at Shipyard, adds that AI works best when a result can be subjective (art, writing, idea generation) or when some error is tolerable — and that when something needs to be 100% right, you need another way to verify accuracy, like code execution or human-in-the-loop review.
Loop stage: Evolve — objectivity is what keeps you from mistaking a convenient result for a correct one when you’re reviewing what worked.
The State of Artificial Intelligence in 2026
New research into how marketers are using AI and key insights into the future of marketing.
- Marketing AI Tools
- Practical Tips
- Trends and Statistics
- And More!
Download Free
All fields are required.
Form not available
Holistic Understanding
Holistic understanding is about seeing how every part of a project connects rather than optimizing one piece in isolation.
As Brian Rhea from JobLens.ai puts it, humans are remarkably good at connecting seemingly unrelated information and experiences to draw a meaningful, creative insight, which is why posts like “What I Learned About Marketing from Game of Thrones” resonate: humans are better than AI at drawing lessons from patterns that recur but aren’t obviously related.
When you zoom out to see how design influences conversions or how a product update affects your storytelling, you start making decisions that serve the bigger goal rather than just your deliverable.
Pro tip: Before finalizing a project, map how your work connects to at least two other functions. It builds context and prevents blind spots.
Loop stage: Express — holistic understanding is what keeps your brand’s story consistent across every channel and format, not just the one you’re focused on.
Ability to Write Good Prompts
A vague prompt gets you generic results. A specific prompt with audience, context, and tone gets you exactly what you need. Good prompting is structured thinking — knowing what you want before you ask for it. That habit sharpens how you brief teammates and plan content, too, not just how you talk to AI tools.
Research on prompt structure supports this: a recent survey found that structured, context-rich prompts meaningfully improve large language model performance and reduce misinterpretation.
Example: Instead of prompting with “Write a blog intro about how to create a cash flow statement,” try: “You are a finance content writer creating a blog for startup founders and small business owners who struggle with understanding their company’s cash flow. Write a 180-word blog introduction for a post titled ‘How to Create a Cash Flow Statement.‘
The tone should be educational but conversational, similar to how Investopedia explains concepts — clear, structured, and jargon-free. Highlight why understanding cash flow is critical for decision-making and growth. End with a natural transition into the step-by-step guide that follows.”
Loop stage: Tailor — prompting well is the mechanism that turns generic AI output into something built for your specific audience.
How to Build AI Skills as a Marketer
AI skills, like any other skill, get stronger with deliberate practice. Harvard Business School researchers found that people who paused to reflect on their work performed 25% better than those who didn’t, and that same principle applies directly to working with AI. The marketers who improve fastest aren’t the ones with access to the most tools. They’re the ones who build a habit of reviewing their prompts, questioning their outputs, and adjusting their approach.
Here’s how to build that habit, and which AI skills to learn first.
1. Take online courses.
Structured learning gives you a faster, more reliable path to AI fluency than trial and error alone. HubSpot Academy’s AI courses cover prompt fundamentals, responsible AI use, and how to apply HubSpot’s AI Assistant inside your actual marketing workflows, so what you learn maps directly onto tools you’re already using, not a generic AI concept.
2. Practice active listening.
Active listening applies to AI the same way it applies to people: it’s the difference between skimming an output and actually engaging with what it’s telling you. Before you accept an AI-generated recommendation, sit with it.
Does the reasoning behind it hold up? Would you make the same call if a colleague suggested it? Slowing down here catches errors that a quick skim misses.
3. Ask questions.
Don’t take an AI output at face value. Before you use it, ask:
- What context did I give the model, and what might I have left out?
- Where might this be outdated or generic?
- Would this change if I gave it a more specific audience or brand context?
Getting into the habit of interrogating outputs — not just accepting the first version — is what separates marketers who use AI well from marketers who just use AI.
4. Question your biases.
It’s tempting to trust an AI output more when it confirms what you already expected, and to dismiss it faster when it doesn’t. Catching that tendency is part of using AI responsibly.
Before you approve AI-generated copy, a segment, or a recommendation, ask: Am I accepting this because it’s actually right, or because it’s convenient? Have I checked it as carefully as I would if it disagreed with me?
5. Seek out diverse opinions.
AI tools reflect the data and patterns they were trained on, which means they can just as easily reinforce a narrow point of view as challenge one. Run a new AI-generated campaign idea or piece of messaging by colleagues outside your immediate team before it ships.
A fresh perspective, especially from someone with a different background or role, will catch blind spots that neither you nor the model can see on your own.
Develop the right skills, and AI becomes a tool you direct.
The more you work with AI, the clearer it becomes that tools don’t define growth: skills do. What matters isn’t how many platforms you can use, but how you think, learn, and adapt when things change.
Every skill on this list, technical and non-technical, shapes how you work with AI at each stage of the loop. Some come naturally; others take time and a few mistakes to build. Start applying them today, and you’ll be glad you did.
FAQs About AI Skills for Marketers
What are AI skills?
AI skills are the capabilities marketers need to work effectively with artificial intelligence — from writing clear prompts to evaluating AI output and knowing when to trust it or push back. Most AI skills are non-technical, like curiosity, adaptability, and critical thinking, rather than programming ability.
How can marketers build AI skills?
Marketers build AI skills through deliberate practice: taking structured courses like HubSpot Academy’s AI courses, critically reviewing AI outputs rather than accepting them at face value, testing new tools regularly, and reflecting on what worked after each campaign. Applying HubSpot’s Loop Marketing framework gives AI skill development a repeatable structure.
What does an AI skill look like in practice with Loop Marketing?
It depends on the stage. In Express, creativity and strategic thinking shape a brand’s point of view. In Tailor, prompting skill and empathy personalize content responsibly. In Amplify, adaptability and strategic thinking guide channel experimentation. In Evolve, critical thinking and people management drive honest review and iteration.
Do you need coding skills to use AI in marketing?
No. Coding helps if you’re building custom AI workflows or integrations, but it isn’t required to use AI well. Most marketers get more value from sharpening how they think, prompt, and evaluate AI output than from learning to code.
Editor's note: This post was originally published in November 2023 and has been updated for comprehensiveness.
The State of Artificial Intelligence in 2026
New research into how marketers are using AI and key insights into the future of marketing.
- Marketing AI Tools
- Practical Tips
- Trends and Statistics
- And More!
Download Free
All fields are required.
Form not available
.png)
