What if I told you that you could take the blue pill and keep marketing the same way you always have — or the red pill, and open new doors of perception through which you’ve never peered? Which would you choose?

Market in the dark, or in the light? Stay stuck in the same old reality, or unveil new dimensions you haven’t seen (but nonetheless which very much exist)?

Now what if I told you that red pill was a special recipe to cure many of your marketing woes and was available to any marketing team?

Artificial intelligence is that antidote, but unlike the 1999 movie The Matrix, there’s no ploy involved! Thankfully, AI wasn’t spawned from some malevolent marketer bent on cultivating your data for its own agenda.

The opposite is true: AI is becoming one of the most highly sought after technologies in e-commerce, and for good reason. AI is helping marketers uncover new insights, predict future customer events, and anticipate how customers will behave along their journey through enhanced lenses.

E-commerce marketers use AI to become proactive

Give me the Whitepaper

Using AI to Glitch the Marketing Matrix

What do I mean when I say “new dimensions” of marketing? What kind of new information am I talking about and, ultimately, how does this new information impact the bottom line?

I always assumed that state-of-the-art marketing automation technology, rich personalization capabilities, and solid campaign blueprints were the best that e-commerce marketers could hope for… and until now, that was sort of true. And that was a darn good combination of tools to do all kinds of real-time execution, 1-to-1 content customization, and more.

But then I started digging deeper into the truth about what AI can actually do. It surprised me to learn just how much marketers are missing — and how much they don’t know — without leveraging (real) AI woven into the fabric of a marketing platform/personalization engine.

Let’s look at three ways AI works to unlock new information.

AI helps uncover new insights about who customers are

Personalization and automation are the end result of what AI can do. They’re the output.

Some marketers don’t really care about what happens before that, or what kinds of information they can leverage to inform those outputs. But the insights hiding within existing data can point to information about further events likely to occur down the line.

Daniel Eisenhut

“AI interprets data points that are collected across the consumer purchase journey. It is not generating new data; it only enriches attributes based on initial captured behavior. Therefore, the underlying captured data needs to be as accurate as possible.”

Daniel Eisenhut • VP, Services & Support, Emarsys • @eisenhut_dan

AI helps analyze (putting large quantities of big data into understandable, meaningful clusters) a plethora of information about the “who”, including:

  • Buyer Predictions
    • Likelihood to move from first- to second-time buyer
    • Likelihood to go cold
    • Likelihood to defect
  • Lead Predictions
    • Likelihood to convert or buy again
    • Likelihood to become cold or churn
    • Likelihood to remain inactive or defect
  • Revenue Predictions
    • Next cart value
    • Value over time

AI marketing gives businesses a fighting chance with today’s finicky buyers — to be able to say, with a high level of accuracy, who is likely to buy or convert and how likely they are to do so. This is all based on previous behavioral/purchase data.

You can also see which sets or segments of customers are likely to buy again, remain inactive, or go away for better targeting. Marketing is becoming a game of levels, and AI offers a new kind of tiered understanding of your database that hasn’t been available before.

AI helps predict customer events and what to do about it

The key ingredient to AI systems (compared to traditional automation) is the self-learning element.

It’s one thing to do A/B testing, experiment with sample segments, or even make educated guesses about what content to send and to whom. It’s another entirely to use self-learning artificial intelligence algorithms to anticipate, with confidence, numerous attributes about customers, campaigns, and communications.

So, how does AI predict these customer events, and which elements does it predict?

Raj Balasundaram, SVP of AI here at Emarsys, describes how BrandAlley U.K. leverages predictive intelligence to do just that:

Raj Balasundaram

“BrandAlley focuses on what data they collect, keeping it clean, making it usable, trainable, and predictable. In the past 6-8 months, they’ve started predicting who is about to churn before it actually happens. When they moved from reactive to proactive, the same campaigns gave them 4x in revenue within half the time period.”

Raj Balasundaram • SVP of AI, Emarsys • @RBalasundaram

twitter “You can use #AI to do a whole lifecycle of predictive #marketing to see who will buy/not buy & churn/not churn,” says @RBalasundaram CLICK TO TWEET

Using engagement score, purchase history and other behavioral patterns, algorithms can make predictions about a contact’s LTV as well as their likelihood to purchase, to visit your website or churn, for instance, in a given time frame.

Then, armed with that information, it automatically tailors products, content, and incentives to each individual, for each use case and across all channels. AI predicts and presents:

  • Color schemes, tones, images, categories, and the same content can be presented in different ways to different customers. Preferences for images varies from person to person, and AI understands visual affinity to create 1-to-1 content.
  • Product recommendations. AI understands product affinity — who is likely to buy what, and which offers will spur purchases.
  • Algorithms do the heavy work to deliver dynamic pricing and incentive recommendation, along with incentive usage predictions.
  • Best Send Time. AI identifies the exact time to engage each contact across all channels, including when each is most likely to buy at different points in time.

AI unlocks the full customer profile and anticipates value

Next-level insights (and increased relevance and accuracy) with AI are only really possible when it’s embedded in an end-to-end solution with a knowledge layer across every connected channel.

While channel-specific AI tools can work well in narrow use cases, they lack the ability to integrate data back with a unified profile. Ultimately, any insights and experiences will be incomplete and disjointed.

But, with an underlying layer of self-learning algorithms that govern multiple channels, AI can shine. Why? Algorithms thrive on data. And quality data in high quantities across all of your channels give your machine as much clarity, completeness, and transparency as possible.

Ashwin Ram

“Machines need to know stuff about the world. The way they learn is through experiences, data, and examples… those experiences give them the data from which to learn. But you also need the right set of algorithms. With machine learning, if you take in enough data, you can train a system to do almost anything. With modern AI, there is a lot of big data involved, but also new algorithms.”

Ashwin Ram • Technical Director of AI, Google, AI Researcher, and Entrepreneur • @ashwinram

Nothing will be hidden, and nothing will be unknown about who a customer truly is. The floodgates will be open, and only truth will come forth.

With all of that data, you can use AI to make strikingly accurate projections about:

  • Next cart value
  • Customer lifetime value (CLTV)
  • Value of segments of customers
  • Likelihood of those segments to take certain actions
  • Business revenue over the next quarter

Final Thoughts

A new paradigm shift is taking over the e-commerce space. Marketers are finding deeper truths about so many new aspects of the business… all aided by the most powerful technology to date: artificial intelligence.

The fun part is this: we’re just scratching the surface. With AI, we’re unlocking these new dimensions about our database never seen before, and it’s enabling more intelligent automation across every channel.

But imagine what we’ll be able to do in the next 10, 15, 20 years. Mere analysis, segmentation, and automation will be things of the past. Predicting customer events based on behavioral data will easily be table stakes. True 1-to-1 marketing will come to mean something entirely different than it does now. I’ll leave you to ponder the possibilities.

In two weeks, we will take an extended look at how AI is empowering marketers to escape the marketing matrix on the Marketer + Machine podcast, so be sure you’re subscribed.

Handpicked Related Resources:

AI predicts buying probability, likelihood to convert & CLV. Learn more!

Download the Whitepaper