Marketing
AI can be your new best friend, if you keep it on a leash.
Here are a few do’s and don’ts when incorporating AI in marketing communications.
AI belongs in B2B marketing, helping marketing teams do more with less. But there’s a danger in letting it have too much say when it comes to your branding efforts. Here’s where we think it works and where it doesn’t.
AI is better for production, not inspiration
AI accelerates output, generates options, and removes friction from execution. A marketing team that uses it at the right stage moves faster, and that matters in categories where differentiation is hard to maintain.
Branding is a different kind of work. It is the set of decisions that determine what makes a company distinct: what it believes, who it serves, and how it sounds when it speaks.
In complex B2B categories, those decisions are especially consequential. Your buyers are comparing multiple technically credible vendors, and what separates you often comes down to how clearly and distinctively your brand communicates who you are and why that matters to them.
An AI model trained on content scraped from across the internet cannot produce a judgment unique to your organization. While in many instances the output may be professionally sound, it will not be singular or inspiring. More likely, the content that these models are trained on is a lot of the same B2B marketing jargon we’ve been decrying for years—what it returns will likely be jargony and generic, and easily replicable from anyone using the same tool and similar prompts. Why participate in a race to the middle?
Where AI earns its place
The question worth asking is not whether to use AI in branding work, but at what stage it belongs. The answer is execution, after the brand has been defined by people who understand what it needs to stand for.
1. Scaling content. Once the brand voice has been established, AI can produce copy variants, social posts, and product descriptions that stay within it. The brand is the constraint, and AI works efficiently inside it.
2. Generating options for human review. AI can write ten headline options in the time it takes a copywriter to write two. You can then select the one that sounds like the brand. AI accelerates the work. People make the decisions.
3. Research and synthesis. Competitive landscape reviews, customer language patterns, category trend analysis: AI handles large volumes of input faster than any team. The interpretation of what that research means for the brand still requires someone who understands what the brand is trying to do.
4. Consistency review. Does this copy match the brand guidelines? Does the tone fit the channel? AI can flag deviations from established standards faster than manual review, which can become a meaningful advantage once those standards have been set.
In each of these applications, the brand decisions have already been made by the people who know the business best.
Where AI costs you
The risk isn’t using AI. It’s using it too soon, before the brand questions have been answered. For example, here are three areas you will want to keep AI on a tight leash, or avoid using it altogether:
1. Positioning. What makes your company different from its closest competitors? AI will produce a credible answer. It will produce the same credible answer for every competitor who asks a similar question. In a category where buyers are evaluating multiple vendors with comparable capabilities, positioning built on a shared tool is positioning any competitor can replicate and doesn’t give customers a reason to pick you.
2. Voice. How a brand sounds is one of the few things that truly belongs to it. Voice is built from the people inside the organization and the customers they serve. AI constructs voice from averages, and average voice blends into the category rather than standing apart from it.
3. Values. What does your company believe, and what will it not compromise on? In regulated and community-facing industries, buyers feel the difference when those answers are clear and when they are not. No prompt retrieves them. They live in conversations inside the organization and the hearts and minds of its people.
One question before you approve anything
Before publishing any AI-generated brand content, ask this: could this have been written for any company in my category? If the answer is yes, AI made the brand decision. It’s time to rethink your branding, keeping AI tools out of your brand’s positioning, voice, and values.
You are in a competitive market. Your content needs to reflect what is unique about your brand and not sound like it belongs to your competitors. It should shine a light on what makes your brand stand out. Your AI output can only clear that bar if the brand work was done at the start. No amount of prompting will substitute for a unique value proposition.
