Provocateur:

Thinking about AI makes your heart beat faster. You feel giddy. Maybe you even get butterflies. It’s not love. It’s just a crush.

In reality, all the hype around artificial intelligence has led you to fall in love with the idea of using AI and machine learning to do better, faster, cheaper, always-on personalization of content and offers, understand buyer’s needs before they know them, and even provide smart “helper” apps to field marketers and merchandisers.

Sure, AI can help us see more trends and even crunch data in the background, spitting out new insights or recommendations. And soon enough, the hype insinuates, we’ll all be on the beach while our smart marketing bots scurry around doing our work for us. If only.

Those of us who were at the frontlines of AI back in the early days (or at least the mid-90s) have experienced this hype before. What’s feeding it now? Well, certainly the volume of new funding going into AI-related start-ups, which spiked 4x from 2012–2016 according to CB Insights. Big companies are falling over themselves to spend more on smart apps and be AI-savvy. In fact, research house Forrester projected a 300 percent increase in corporate AI investments from 2016 to 2017.

AI is super-exciting to marketers. And yes, machine learning (a sub-set of AI, but sometimes not) offers to revolutionize the way we monitor, model, pattern match, and interpret all the big and small data swirling around us. But no, it’s not all going to happen overnight. Most businesses and consumers still don’t have a clue about the best ways to select, apply, and monetize AI for practical, everyday stuff.

Marketers are no different despite the claims to the contrary.

The good news is that many of us have become more data driven and are using tools that have smart stuff baked in, whether it’s in our SEM or CRM or favorite ad tech platform. Even so, many of us want AI to be a magic bullet. We marketers love shiny objects. And AI is glistening up on the hill.

So, before we get too lovestruck, I offer five ways to make sure that crush you have on AI doesn’t get stoked by the hype machine.

1. Start small, stay focused. Creating a general-purpose thinking machine is really hard. Creating an intelligent agent (or bot) that gives us advice or automates a single or small set of everyday, repetitive, “standard” tasks is a lot more tractable. Just as the key to early AI was finding narrow but high-value applications — such as improving search performance or predicting which of three offers would drive more sales of a product — we need to apply the same type of “think global, act local” approach to marketing use cases for today’s AI. For the same reason, starting with small-ish data versus super-large data sets can make sense when applying analytical techniques to many marketing applications.

2. Experiment early and often — especially with machine learning. It’s not easy to get machine learning right the first time. Or the second time. An agile approach with lots of experiments is the way to go because there are so many algorithms to choose from (Bayes, decision trees, regressions, and neural network models, aka “deep learning”). Of course, selecting and training most of these models means you likely have to call in your techy friends. But, there are lots of good resources and tools to get started, such as the excellent KDNuggets site by one of my colleagues from way back in my R&D days.

3. Recognize that data is (still) king. Getting close to customers, understanding their journey, tailoring their experience, and selecting just the right offer are all outcomes that insights powered by big and small data can enable. Generating these insights in a timeframe and cost that make them readily available to frontline teams (and consumers themselves) is where advanced analytics and techniques such as deep learning need to go. But, as mentioned above, what insights you get out is very much a function of what data you put in. Where will your training data come from? How will you prepare it? Who will test the performance? These questions are as important as what tool or algorithm you’ll use.

4. Focus on helper apps. Even as AI systems become more skilled at complex decision-making, and take over some “back of house” functions, marketers aren’t going to be retiring to the beach any time soon (sorry!). Instead, they’ll be supported by AI and “helper” apps — whether users are sitting at a desk at HQ or out in the field; for example, embedding smart algorithms into a CRM solution that help field managers spot trends in their territory and help reps know which customer to visit next and what to feature on the shelf for a specific store, market, or season. Bottom line: Some of the best, most impactful use cases will continue to augment rather than replace humans.

5. Build apps that improve everyday work. Start by turning small and big data insights into everyday value. There are established use cases for data-driven marketing that do just that, such as always-on personalization. Get ideas from sources such as TopRight Partners, which published a helpful framework for considering which marketing processes are mostly likely to be disrupted by AI, and McKinsey, which published a very cool study (and poster!) on the overall potential of automation in the workplace, called “Where machines could replace humans.”

What tactics for getting real have you discovered as you’ve played around with AI in your marketing role? I’d love to hear how you get past the hype. Have your bot call my bot.

About the Author

Allen Bonde is VP of marketing at mobile CRM provider Repsly. He was previously SVP of marketing for Placester, VP of product marketing & innovation at OpenText, and cofounder of social marketing pioneer Offerpop (now Wyng). Earlier in his career he spent time at eVergance (now KANA), McKinsey, and Yankee Group, consulting with global B2B and consumer brands. He started his career as a data scientist and AI researcher in the telecom sector, and continues to advise several data-driven start-ups. He has also given keynote talks on four continents, has appeared on MSNBC, CNBC, and FOX TV, and blogs at Small Data Group.

Find Allen at @abonde and @repsly