When it comes to artificial intelligence (AI Marketing), there’s something big that I’ve learned: You’ve got to take an “all-of-the-above” approach. What I mean is that AI typically can help marketers in three ways: improving decision-making, systems and broader transformation. But marketers only tackle one or two at a time. It’s when you do all three at once that AI can really pay off. That can be hard (AI is hard) but the results can be magnificent
As a CMO who works hard to keep up with the latest MarTech developments, I see three main ways to use AI Marketing. The first is to make better decisions. Trying to decide on channels? How much to invest in a campaign? Which demographics to target? AI can ingest the relevant data and then produce forecasts and analyses to help you make the right choice.
The second way is to allow AI to simplify your everyday tasks for you or help you do them better. Is your team tired of putting together reports? Let a data-AI pipeline do the work, with real-time results and analysis. Are creatives spending time searching for information or just figuring out what they need to do next? Allow AI Marketing to organize their workflow and desktop.
The third way is to make digital transformation come to life. Do you want a marketing function of the future that constantly iterates campaigns as data comes in by forming, dispersing and reforming as challenges arise? You’ll need AI (supported by cloud and data) to make it happen.
Each of these goals is ambitious. But if you just do one at a time, you’ll probably miss out.
In a recent survey of leaders actively working in AI Marketing, the biggest takeaway was that companies working on AI in three dimensions at once (enhanced decision-making, systems modernization and business transformation) are far more likely to get valuable business outcomes.
Typically, marketers don’t do this. It’s common to use AI Marketing to modernize marketing systems, such as AI-powered online recommendations and chatbots. A recent survey also found that marketers expect the use of AI or machine learning for prediction and measurement (which would support better decisions) to triple in the next three years. But putting it all together is rare. Even among digital marketing leaders, only 17% are using AI across the function.
How You Can Do It
One reason an all-of-the-above approach works well is scale and synergy. If you’ve got AI powering chatbots and online recommendations, for example, they’ll produce consumer data that another AI Marketing model could use to help you make decisions. Another reason is that an all-of-the-above approach obliges tech teams to work with you. Don’t get me wrong, I love my firm’s techies. But collaboration is essential to confirm the tools are valuable and understandable.
Your CEO probably doesn’t call you up to ask for AI Marketing advice. But there is, in fact, a lot you can do as a marketer to get leadership what they need and to help out the whole organization.
1. Begin with outcomes. Don’t start by identifying a great AI Marketing tool thinking you’ll figure out how to use it later. Instead, start with the outcome you want—say, more power to predict and measure a campaign’s impact—and then assess how AI and other MarTech can help. This approach will help you and IT look at all the ways AI can help deliver the goods.
2. Make connections. As a marketer, you’re used to addressing different audiences. Direct that talent inward, working to align the business and IT around your project. If your executives see the value of your project (which they likely will if it supports better decision-making, modernized systems and digital transformation), your odds of success rise.
3. Upskill and engage. Too many initiatives fail because of a lack of adoption. New, AI Marketing-powered data analytics can lead to better decisions and a data-driven marketing function only if people actually learn and use them. Emphasize not only training but ways to make training and usage appealing.
4. Make a case for responsibility. As a marketer, you’re a steward of the brand. Insist on processes to assess AI models for explainability, robustness, bias, fairness and transparency. There’s no point in having AI Marketing if it doesn’t produce accurate results—or if it leads to bias issues or privacy violations.
AI isn’t easy, but if you take the right approach, you can raise the odds that one of the biggest investments the marketing function may ever make will be a roaring success.