The AI Fluency Report: What Advanced Marketers Are Really Doing With AI

[tldr]

Key findings from our AI fluency report

• AI usage is mainstream, but AI fluency is not: Nearly all of respondents use AI day-to-day, but only 12% qualify as hands-on builders or fully adopted developers.
• The biggest leap is moving beyond basic usage: Once marketers start using AI past basic prompts and tasks, they often start dabbling in building themselves.
• Advanced AI marketers are workflow designers: 80.8% of our advanced cohort referenced workflow automation or operations.
• AI use extends far beyond content: While 76.6% use AI for content and messaging, 60%+ also referenced research, analytics, strategy, and creative production.
• The best use cases keep humans in the loop: Almost no respondents framed AI around replacing humans entirely; most described using AI to increase speed, scale, efficiency, and clarity—with human input still very much needed.

[/tldr]

In 2026, the question isn’t about if marketers are using AI, but how—and how effectively.

That’s why Right Side Up’s vetting process to be in our talent collective now includes a thorough AI assessment. And while the primary goal behind this is to ensure we continue connecting our clients with top-shelf marketers, we’ve also gathered a ton of data that’s helped us better understand the relationship between AI and marketer. 

We’ve confirmed some common AI theories, invalidated others, and along the way, gathered tons of inspiration on how AI can make us better.

We wanted to share some of these findings, based on the input of 580 senior marketers. In this report, we’ll cover:

• The AI fluency ladder—and where most marketers land on this ladder
• What tasks AI-fluent marketers are using AI for the most (with some potential surprises)
• What tools are most commonly used (yes, there’s a hefty dose of Claude)
• Some of the most interesting use cases we noticed
• Themes we expected to see in the responses but didn’t

AI fluency in marketing: The results

Straight from the bat, there’s a clear discrepancy between who’s using AI (almost everybody) and what they can actually do with it. 

^{Data based on 580 answers from senior marketers within our talent collective}

There are two really interesting takeaways here.

There’s a Grand Canyon–sized gap between AI access and AI fluency

The most obvious takeaway (or important one, at least) is that while almost all marketers are using AI today, only 12% are using it to its fullest capabilities. That’s a huge gap—and those who can attain that AI-native level can truly differentiate themselves.

Most marketers who move past basic AI usage become builders

Most of our cohort resides within the AI-proficient category.

This means that once marketers move past the beginner level, they often find the confidence and skills to not just use embedded AI features, but also to become builders themselves. The hardest part, however, is attaining that final level of true fluency. 

The AI fluency ladder

Now, we’ll connect the survey results with a basic maturity model that explains what happens at each fluency stage: what the marketer does + a few examples.

Proficient AI users also use it for basic tasks

In addition to our general fluency assessment, we also asked advanced users additional questions to know exactly how their AI minds work. 167 who fall in the advanced or expert category responded.  

An interesting finding: Proficient AI users do not abandon basic tools, but rather stack them. 

74.3% of them use general AI tools, and 57.5% use embedded features. What this tells us is that AI-native marketers treat sophisticated tools as means to an end—this end being redesigning how they work entirely to make things as efficient and effective as possible. To do that, they use all tools in the AI box, from basic everyday prompts to complex systems.

Where AI-fluent marketers are creating the most value

The most AI-fluent marketers are finding recurring friction and building around it. They’re using AI across the full marketing spectrum, then apply that value through various delivery systems—playbooks being the most popular one.

Where AI-fluent marketers are creating value

^{Data based on 167 responses from our advanced cohort}

How AI-fluent marketers are delivering that value

^{Data based on 167 responses from our advanced cohort}

Let’s go over four key points this data reveals.

Content is still the main character, but not the whole story 

Content remains the most common AI use case among advanced and expert users, with 76.6% mentioning content, copy, messaging, SEO, or GEO applications. But the “AI just writes copy” narrative is becoming more archaic by the day, with research, analytics, product strategy, creative, and tools all clearing 60%.

Not only that, but the respondents mentioning content didn’t just use AI for copy; they used it to generate briefs, refine messaging frameworks, scale production, and build entire systems to support entire content workflows.

AI is showing up in strategic work, too

A potentially surprising finding is that nearly two-thirds of respondents referenced product strategy, planning, customer experience, or business decision-making. Instead of just treating AI as an execution tool, these marketers are using it to accelerate and validate strategic thinking (which is, yes, still led by humans for now).

Measurement is getting faster

Analytics, reporting, and measurement appeared in 65.3% of responses, making it one of the most common advanced use cases. Respondents described everything from automated dashboards to AI-powered reporting systems that surface insights, flag anomalies, and reduce the manual effort traditionally required to understand performance.

[tldr]

Related guide

Read our complete guide to marketing experimentation and learn to test, measure, and model what's truly driving results.

[/tldr]

AI-fluent marketers are workflow designers

Workflow automation and operations appeared in 80.8% of advanced-cohort responses, making it the top category overall. As mentioned earlier, advanced AI users are using all of their skills into redesigning how work gets done and creating repeatable better outcomes, whereas more novice users just focus on creating outputs faster.

What AI tools marketers are using

To understand where our marketers were building, we combed through responses and looked for tool and platform mentions. Though a lot of names were thrown around, one came up again and again.  

^{Data based on 167 responses from our advanced cohort}

Claude has become the default AI platform for advanced builders

Almost 90% of our respondents mentioned Claude and/or Claude Code as a tool of choice. Clearly, Claude has become more than just a chatbot for many advanced builders; it now also serves as the coding assistant, workflow architect, and agent builder. OpenAI’s stack ranked second at 60%, and all other platforms were mentioned much more sporadically.

The tool stack is getting more specialized

Beyond the major models, our cohort referenced a variety of tools that serve completely different purposes, including:

• Midjourney for image generation
• Lovable for app-building
• N8N for automation framework

This suggests that reaching AI mastery requires orchestrating multiple platforms—the same way it requires using several application methods (from simple prompts to sophisticated agents). In the end though, tool choice matters less than workflow-thinking: the most impressive examples in our data set weren’t tied to a specific model. 

Three of our favorite playbooks built by our talent

The custom MMM that brought enterprise-grade measurement within reach

One respondent built a custom marketing mix model (MMM) designed to help marketers make more informed budget allocation decisions without relying on expensive enterprise software. 

By combining historical performance data, media spend, and channel-level inputs, the system provided ongoing guidance on where incremental dollars were likely to have the greatest impact. In the end, the brand saved money while saving money.

The creative testing workflow that turns ad data into better briefs

One growth marketer built a reusable creative analysis workflow that combines raw ad performance data—spend, CPI, conversion rate, ROAS, retention, and winning creative assets—and turns it into structured briefs for the next testing round. The output includes new angle hypotheses, hook rewrites, and production notes for editors.

The win: Preventing friction between performance data and creative strategy, and ultimately enabling faster and more evidenced-based creative testing. 

The growth marketing dashboard that replaced manual performance analysis

Another marketer helped build a growth marketing dashboard that pulls daily data from seven ad platforms, normalizes it, and joins it with attribution data. The system can also analyze performance, then support optimization decisions through a draft-and-approve workflow. It also supports incrementality testing using open-source libraries.

What AI-fluent marketers don’t care about

Our data didn’t just reveal what AI-fluent marketers valued the most; it also showed us what they were indifferent to—something just as revealing. Here were some themes and keywords that simply never appeared when we hit Command + F and what these omissions tell us.

1. Prompt engineering was rarely mentioned—but not for the reason you think

Prompt engineering was rarely mentioned by our respondents. Has it gone out of fashion? Quite the contrary: good inputs are the building blocks of successful workflows and playbooks. But we’ve reached a point where prompt engineering skills are such a necessity that AI-fluent marketers don’t even think to mention it explicitly when discussing their work.

2. There were no mentions of AI replacing humans

In reassuring news, almost no one framed AI as a way to reduce headcount. Instead, the focus was about reducing manual drag and ensuring humans spent their time on the most impactful things—with scale, speed, efficiency, and clarity being the most common themes. Not only that, but the best AI-led processes systematically required human input at one point or another. 

3. They built AI ecosystems, not one-off tasks

The most advanced respondents built full ecosystems. Here’s what we mean.

A less mature use case might look like this: research → output.

A more mature pattern looked like this: research → analysis → dashboard → alert → action.

The latter connects multiple steps into a system that continuously generates insights, recommendations, and actions. That shift is the heart of AI fluency.

[tldr]

Bring AI-fluent marketing talent into your team

As this report indicates, AI expertise has become an integral part of modern marketing—and occasionally dabbling in ChatGPT isn’t enough.

Right Side Up connects companies with AI-forward marketers who know how to apply these tools where they actually matter, from building smarter workflows and reporting systems to optimizing content for GEO.

Need AI expertise today? Get in touch with us.

[/tldr]