Chad Hetherington

Need marketing help? Contact Brafton here.

A lot of content about AI in marketing sells the technology as a godsent solution for our industry woes. I’d like to think that we’ve taken a more practical approach on this blog by being objective about what AI can actually help marketers achieve, and where they still need to be the buffer between generative output and a published piece.

Many marketers still have hangups about AI that have kept them from using it regularly or at all, which we learned more about in Brafton’s second annual State of AI in Marketing Teams report.

About 19% of marketers surveyed have either never used AI or tried it and decided against continuing. That’s just the reality of the technology as it exists today. It’s not perfect, and might never be.

But among those who have concerns, there ought to be some who are still interested in exploring AI, but are unsure how without making concessions. So what are the best ways to do that without completely compromising personal priorities?

Let’s explore the seven leading reasons marketers said they’re avoiding AI, and some practical steps to experiment without completely ignoring very real and understandable concerns.

1. Moral, Ethical and Environmental Concerns (48%)

This was the most common reason marketers gave for avoiding AI, and it’s also one of the more difficult concerns to address. Demonstrating how much time the technology could save won’t resolve it.

Questions about copyright, creative ownership, labor practices and environmental impact deserve consideration.

For marketers concerned about sustainability, for example, one option is to investigate AI providers that put environmental considerations at the center of their offerings:

  • Ecosia is an environmentally focused search engine with AI search and chat features that prioritize smaller models.
  • GreenPT is an AI assistant that emphasizes renewable energy, European hosting (for all the EMEA marketers) and visibility into estimated energy consumption.
  • Viro AI also focuses on energy-conscious model selection and environmental impact estimates.

These aren’t necessarily zero-impact alternatives, and their environmental claims are worth evaluating. But they give concerned yet interested marketers more options to choose from.

For concerns about copyright and creative ownership, review providers’ training-data policies and suss out details that will help inform a decision about which providers to explore more or avoid.

Perhaps the best approach is to develop personal criteria and research how prospective AI providers address each one. Then, establish which tools and applications meet those criteria, and start experimenting with those.

2. Lack of Confidence in AI Output Quality (35%)

Questionable output is a general symptom of any AI tool. While better instructions and context can improve results, there’s only so much they can do if the tool itself isn’t intentionally designed for a specific task.

For marketers disappointed by AI-generated content, it may be worth experimenting with different tools — since different AI models have different strengths. Tools built for marketing work, for example, can offer capabilities that general-purpose chatbots struggle with.

Specialized platforms like contentmarketing.ai support brand-informed content creation, while its Google E-E-A-T Recommendations workflow helps identify opportunities to improve content quality and credibility.

Regardless of the tool, editorial review is still non-negotiable. But an AI tool whose purpose is to save time and help scale should get marketers closer to the finished product sooner in their process. In other words, the content it produces should be worth building on.

3. Data Privacy Concerns (30%)

AI tools can become more useful when they have access to proprietary information. But providing that information is a valid concern. Marketers are right to question what happens to the information they submit to an AI tool.

Here are a few ways to navigate that:

  • Review providers’ data policies and settings. Look into whether inputs are used for training, how long information is retained and what controls exist for deleting data or restricting access.
  • Start with public or synthetic information. Brainstorming topics, researching published competitor content and testing workflows with fictional customer data can demonstrate a tool’s usefulness without exposing confidential information.
  • Consider tools with stronger organizational controls. Business or enterprise offerings may provide additional data protections, administrative settings and contractual assurances.

For organizations handling client information or sensitive customer data, AI adoption should involve privacy and security teams, not just the teams using the tools.

4. Current AI Tools Don’t Meet Our Needs (22%)

There’s a difference between AI not being useful for a task and any particular AI tool being unsuitable for it. General-purpose chatbots can do plenty, but specialized marketing platforms may offer better results for specific activities.

Take SEO content planning, for example. A chatbot might suggest a list of blog topics, but those suggestions won’t necessarily reflect a website’s existing keyword rankings or competitors’ actual search performance.

contentmarketing.ai‘s Topic Cluster Analysis, on the other hand, examines a website and its competitors’ keyword rankings and topic coverage to identify pertinent and achievable content opportunities. In addition, its Existing Content SEO Analyzer can evaluate published pages against target keywords, compare them with competing content and recommend improvements.

Of course, these workflows or others like them don’t guarantee better rankings. But they’re built around actual SEO activities rather than general text generation, which can give AI-assisted SEO and GEO work a much more solid starting point.

For marketers who’ve been disappointed by AI tools in general but are still curious, identify one task that hasn’t worked well in the past, then test a purpose-built alternative.

5. Unsure Why They Aren’t Using AI (17%)

Some marketers couldn’t identify a particular reason why they hadn’t adopted AI, which is one of the more interesting responses from the survey. Maybe AI just hasn’t been a priority, or it’s difficult for some to see where the technology fits into an already established marketing workflow.

Experimenting at random can feel aimless, so for those who are unsure why they haven’t adopted AI but are interested, it might help to look at how other marketers are using it.

At risk of being too meta, this very blog — the ai marketer — explores tools, workflows and practical ways to apply AI to everyday marketing tasks.

Zapier’s Workflow Gallery is another great and practical resource, especially for marketers who are already using Zapier for other activities or are interested in it. The gallery showcases functional workflows created by real users for specific tasks.

A million and one outlets explore practical AI applications, and one is bound to have a convincing or creative use case for unsure marketers.

6. Lack of Budget for AI Tools (13%)

AI is often promoted as a cost-saving technology, but everyone knows how expensive subscriptions can get these days.

Fortunately, marketers don’t necessarily need a paid subscription to investigate whether AI could be useful.

Plenty of tools, like ChatGPT and Gemini, offer free options. And tons of specialized tools also offer at least a free trial period, which can help teams evaluate whether a particular use case justifies a subscription later on.

7. Lack of Training or Expertise in AI Systems (13%)

Marketers are busy folks already, and it’s not always easy or ideal to take time away from the work at hand for in-depth AI training. But finding useful AI applications doesn’t require mastery.

One option is to start with a guided workflow for a familiar task.

For example, contentmarketing.ai‘s AI & Search Visibility Brief analyzes competing pages for a target search term and recommends ways to create competitive content — all while giving marketers a structured process to follow.

For those interested in broader learning, Google’s AI learning resources and Microsoft Learn offer introductory AI training and practical exercises to work through.

Teams can also build their own shared knowledge library of successful prompts, workflow examples and common mistakes. That way, each person doesn’t have to figure everything out independently.

AI Adoption Doesn’t Have To Be All or Nothing

Marketers shouldn’t feel like they have to dismiss their legitimate concerns to explore AI. In fact, being particular and selective based on personal principles is a great thing.

When it comes to AI, marketers have the benefit of options. There are plenty of tools and providers available, and there’s bound to be at least one option — but most likely more — that directly addresses a specific concern.

Concerned but still curious? Consider your non-negotiables, investigate options, test them and think critically about a tool’s actual value. If something clicks, keep it.

In the end, AI adoption should be a practical decision, not a foregone conclusion.

Note: This blog was originally posted on contentmarketing.ai.