I listened to an interesting podcast this week on the topic of artificial intelligence (AI). The host and guest were discussing explainability. That's the need to have an AI model that's explainable to the layperson, or business level knowledgeable stakeholder. The premise is that folks like the COO/CFO should be able to understand the objective of the AI model, along with understanding the key tenets of the model -- what it does, and how it does it. Explaining a model at a summary level, maybe a one slide Powerpoint level, is going to be pretty easy. But, if you dig down into the details of data sources, classifications, etc. it becomes a different story.

If I remember right, the podcast scenario was something like an AI model to optimize the efficiency of an airplane wing. They were talking about the depth of engineering involved and how it was well beyond the realm of explainability to the layperson. They got down to the question of, “Do you want to build a model that's good, or a model that's explainable?". Sometimes you can't have both. That really resonated with me. If an engineer has to dumb down a model to the point that a layperson understands it, the model may not be very good.

Apply that to digital marketing, and trying to explain the tactics and practices to an uneducated legacy marketer. Just because they don’t understand it, doesn’t mean it’s not the right approach. And, while most digital tactics can be explained at that one Powerpoint slide level, it often time requires some domain knowledge to drill down into the details.

So this is my counter to the gobbledygook blog post. While consultants need to keep the gobbledygook and technobabble out of the conversation, marketers need to earn their way into the conversation that they want to have. If you’re not willing to hone your craft. You shouldn't participate in the game. Today's marketers can't hide from tech. Understand it enough to do your job or you'll probably get left behind.