Trending Now

Interview with Mike Moran: AI/ML in Marketing and Beyond

I had the pleasure of interviewing Mike Moran about the use of artificial intelligence and machine learning in marketing. We talked about how Mike is really training Artificial Intelligence instead of artificial stupidity, among other things.  I hope you’ll give it a listen.

The full transcript is in the YouTube video, but here’s the first question that I asked Mike:

Tim Peter: All right, so let’s start off. Obviously, AI and Machine Learning depend upon data. Can you talk a bit about the ability to leverage AI and how that’s tied to having your data in a good state to start with?

Mike Moran: Sure, yeah, I don’t think there’s anything more important than having your data in a clean state. And what that means for different applications can be very different. If you’re using supervised machine learning where you have a set of training data where you’re creating a model, it’s really important that not only the data itself be easy to import and all in the same format, but also that you’re very confident of the accuracy of your outcome data.

So, for example, if you were trying to predict the weather and the way you were doing that is you were taking all sorts of inputs of barometric pressure and wind speed, and you were trying to look at where the systems are a hundred miles away, and said, you were using that type of data to predict what was going to happen in an hour or three hours. Then, in order to do that you have to have good data that says that you agree on what it is that happens. If something as amorphous as the weather … For some people accuracy means, “How many inches of snow do we get?” For other people it means, “Did we get precipitation at all?” Because predicting snow versus rain is really difficult when it’s 32 degrees out.

And so, for what looks like a really bad miss that someone says, “Man, we got five inches of snow, and they said it was going to be raining,” it was because of two degrees of temperature that was wrong. And so, trying to figure out what your outcome actually is, what are you really trying to predict, that’s really important when you’re putting your data together.

The other thing that’s important is that a lot of times the outcomes you’re trying to predict are actually a stand-in for human judgment. So, think about the Watson application and how it’s trying to do medical diagnoses. What’s really happening here is that you have to be careful that your data is actually representative of the correct answer, not just the most popular answer. So we’re playing Jeopardy rather than Family Feud.

Everyone thinks that’s the right answer. That’s only good if it actually is the right answer. So making sure that you understand that your data was really compiled by experts who agreed with each other rather than just anybody that you gave a survey to, it’s a really important distinction. I think the most important thing for having your data ready is that you know where it is, you know that it is accurate, and you know how to import it into the machine learning environment.

So, the most easy way to do that is if you have data that’s all in a common format. But you can bring data in that’s in multiple formats as long as you really understand that the different fields in the data are all defined the same way, so it’s possible to do that. Having a data dictionary that says, “This is what it is, and we absolutely know how it was compiled and is being compiled the same,” is just as good as if it all came from the same source.

But the real problem, as I said before, it’s the outcome data. You have to know what your outcome is. You have to know that it’s been done in a way, that it’s being tabulated in a way that is actually the same thing you’re trying to predict. And you need to make sure that if it’s based on human expertise, that those are really experts and that there’s maybe multiple of them that agree with each other.

Tim Peter


Tim Peter built his first website in 1995 and loves that he still gets to do that every day. Tim has spent almost two decades figuring out where customers are, how they interact with brands online, and delivering those customers to his clients’ front door. These efforts have generated billions of dollars in revenue and reduced costs.

Tim works with client organizations to build effective teams focused on converting browsers to buyers and building their brand and business. He helps those companies discover how marketing, technology, and analytics tie together to drive business results. He doesn't get excited because of the toys or tech. He gets excited because of what it all means for the bottom line.

An expert in e-commerce and digital marketing strategy, web development, search marketing, and analytics, Tim focuses on the growth of the social, local, mobile web and its impact on both consumer behavior and business results. He is a member of the Search Engine Marketers Professional Organization (SEMPO), HSMAI, and the Digital Analytics Association.

Tim currently serves as Senior Advisor at SoloSegment, a marketing technology company that uses machine learning and natural language processing to improve engagement and conversion for large enterprise, B2B companies.

Tim Peter’s recent client work covers a wide range of digital marketing activities including developing digital and mobile marketing strategies, creating digital product roadmaps, assessing organizational capabilities, and conducting vendor evaluations for diverse clients including major hospitality companies, real estate brands, SaaS providers, and marketing agencies.

Prior to launching Tim Peter & Associates, LLC, a full-service e-commerce and internet marketing consulting firm in early 2011, he worked with the world’s largest hotel franchisor, the world’s premier independent luxury hotel representation firm, and a major financial services firm, developing various award-winning products and services for his customers. Tim can be reached at tim@timpeter.com or by phone at 201-305-0055.

Join the Discussion

Your email address will not be published. Required fields are marked *

Back to top