• by Ray Schultz , Columnist, July 27, 2026

Marketers at all levels are being sold on artificial intelligence. But AI won’t help them unless they have defined their business goals and understand the impact.

That’s the view of  Natalie Cunningham, senior vice president of marketing at Data Axle, who has an inside view on what works in AI — and what doesn’t. Email Insider spoke with Cunningham last week to learn how AI fits into the marketing operation. 

Email Insider: How are marketers adopting AI? 

Natalie Cunningham: With marketing pre-AI and now in the AI age, we make a lot of micro decisions about who our audience is, where they are in the buyer journey, what’s going to resonate with them, all day, every day.

The biggest challenge is in the agentic era of AI. We’re moving from outsourcing execution to outsourcing decision making, and that’s risky.

We’re overly focused on sort of downstream outcomes, subject lines, authentication, etc. But if you’re sending email to the wrong person because you outsourced to an AI agent, the right segment is never going to see your subject line anyway.  

Email Insider: Who is doing it well?

Natalie Cunningham: It varies based on the industry and on role that email plays. B2C tends to adopt a little bit faster than B2B because email has long been a core part of the customer experience. That's not a bad thing, but the focus now needs to shift from simply adopting AI models to making the right data decisions based on a strong semantic data layer.

Email Insider: What is the semantic data layer? 

Cunningham: The semantic data layer is where your organization defines the business meaning behind its data. In the past, we didn’t have data connected to autonomous systems — you might have a spreadsheet. You want to ask, what does this mean to our business, and how should we leverage it? In simple terms, it's where your organization defines its business rules and decisions so AI understands what your data actually means.

Start with the why -- what’s the problem in your workflow or efficiency that AI can solve -- and find the right data foundation, especially if you’re leaning into Agentic AI to make the right decisions

Email Insider: How are B2B marketers using AI?

Cunningham: B2B is a little behind. There are multiple people in the buying process, so you have to focus on who actually is in market. Your ability to do that depends on the quality of your data.

It comes down to having unified identity resolution, not just at the account level, but the individual level. You build a robust professional profile using information like a person's role, name, and how long they've been in the job. Then you can connect that professional profile with relevant consumer attributes to create a more complete view of the individual. That richer understanding gives AI the context it needs to make better decisions and personalize at scale, especially in B2B. 

Email Insider: How can Data Axle help?

Cunningham: We help marketers combine their own data with Data Axle's to create a more complete view of the individual, enabling more informed marketing decisions.

Email Insider: What are you advising clients to do?

Cunningham: First, take the time to evaluate your team, your technology and your data. That will help you determine where you are ready to apply AI where you are not. Too many organizations are asking, ‘How fast can I use AI?’ Instead, they should ensure they have the necessary skill sets, technology, and most importantly, the right data foundation before scaling AI. 

Email Insider: And then?

Cunningham: Customers are successful when they see AI as a thought partner, not just a production engine. With insights from your customer data platform, AI can provide a more holistic view of your audience. Instead of simply generating a list of prospects or customers, it can help identify your best-fit audiences and uncover patterns you may not have thought to look for. 

Email Insider: What about the cost?

Cunningham: It doesn’t have to be expensive — it depends on the use case and on where you are in your maturity. Lots of companies are making moderate investments and getting returns. It’s better to put the investment into the right data infrastructure than investing more in the newest model.

 2