5 Powerful Ways to Use Artificial Intelligence in E-commerce

Raj Balasundaram
Raj Balasundaram
SVP of Artificial Intelligence
Newsletter Signup
Yes, I would like to receive Emarsys email updates on products, services, news, and events. I can unsubscribe by using the 'unsubscribe' link at the bottom of each email, or by emailing unsubscribe@emarsys.com.

Unless you’ve been hiding under a rock the past couple of years, you’ve heard the rumblings. Artificial intelligence is here, and here to stay.

The way some people have hyped it up, though, you may have thought there’s some three-eyed, two-headed AI monster — coming to steal your job, create havoc among your tech stack, and overtake, mess up, and mutate all of your precious customer data.

But this fear-based approach stems from widespread misunderstandings about what AI is, and how it can help e-commerce marketers. For teams who are ready to embrace this game-changing technology, the benefits are becoming evident.

How AI Is Impacting E-commerce

Let’s get real. AI is fundamentally altering the way online brands can market to their consumers. This is happening whether you believe it to be so, or not. And when it comes to AI, I have some good news and bad news. Which would you like first?

Since I’m a “glass half full” type of marketer and passionate proponent of AI, let’s focus on the good first.

Artificial intelligence is unequivocally changing the e-commerce space. And it will augment aspects of your job (the repetitious, monotonous, data-intensive parts). But it won’t distort your data or cause inconsistencies with your communications. The fact is that there are dozens of tangible use cases for AI in e-commerce that are helping (not hurting) the organizations which have employed it.

Now that we’ve got that out of the way, are you ready for the bad?

This is it: AI is still new. It’s unproven because so few companies are actually using it across the organization. The other piece of this is that AI is easier to implement and much more effective when it’s adopted across channels from an end-to-end solution… which most companies just aren’t using.

There’s nothing wrong with using point-solution tools which claim to be infused with AI, but a centralized, unified customer profile and personalization engine is critical to being able to execute across a multitude of use cases.

As you may have inferred, the “bad” doesn’t actually have anything to do with the technology itself — and true AI is an enabling technology, not a separate product that you can flip on or off.

For most forward-thinking marketers willing to take a little risk, like U.K. retailer BrandAlley, the upsides to leveraging AI far outweigh the negatives.

Related Content: 10 AI Trends Marketers Should Watch for In 2020 [+ Bonus Infographic]

Key Benefits of AI in E-commerce

The outward or customer-facing benefits of AI have been pretty well covered to this point. Content automation, product recommendations, best send time, last-touch attribution, and more can all be significantly uplifted with AI.

But I want to go beyond those and share other more strategic ways AI is changing the game — ways which not very many marketers understand or are talking about. In fact, without AI, many marketing leaders and practitioners don’t even know what they don’t know.

AI has out-of-the-box potential to drive immediate value and help marketers go from reactive to proactive.

What does this mean? In essence (and this is not a stretch) AI can make it possible to predict the future. Marketers using AI are able to predict dozens of metrics including CLTV, likelihood to purchase (purchase probability), and overall revenue.

Making such dynamic (and accurate) projections about customers, segments, and the business has tremendous implications: visibility toward what’s working and what’s not, the ability to tie results to campaigns, and to get in front of churning/defecting customers with individualized, “in-the-moment” content.

See how CMOs unlock new dimensions of their marketing with AI.

5 Ways to Use AI in E-commerce

The move from reactive to predictive, from guessing to knowing, is evident across a number of use cases among the customer lifecycle. Here’s five where we’ve seen, first hand, AI work to drive measurable business results.

1. AI predicts lifecycle segments for buyers.

Scalability and automation aside, AI can help actually differentiate messaging for a variety of permutations of priorly unseen segments across the lifecycle. For instance, you can send different messages to contacts who have made:

  • At least one purchase previously BUT are likely to remain inactive
  • More than one purchase AND are likely to churn in the next 30 days
  • Exactly one purchase AND are likely to continue converting

In these cases, not only does artificial intelligence find and identify the segments, it also arranges the best content for each and orchestrates execution across all.

twitter“#ArtificialIntelligence identifies the right segments, arranges best #content for each, & executes” says @RBalasundaram       CLICK TO TWEET

2. AI drives contacts from one lifecycle stage to the next.

AI helps identify who falls into each lifecycle category, what they need to convert, and when to communicate. It can take first-time buyers who are likely to convert and encourage the second purchase with an offer. AI can also identify active buyers who are likely to convert and provide an offer most likely to secure the next purchase while simultaneously maximizing cart value.  

U.K. retailer BrandAlley saw a 9.4% uplift in conversions among repeat buyers with AI:

average order value

If you know that Customer A buys every 90 days, and Customer B buys every two weeks, you know each has very different purchase cycles. AI understands which customer is of higher value, and, if they go cold, it automates delivery of the content most likely to convert them before it’s too late.

3. Predict who will convert on which offers.

With a strikingly high level of accuracy, AI can predict who is likely to buy or convert. Based on past purchases and other behavioral data, self-learning systems can “sense” (to make things a little more human) who will buy.

Similarly, AI also understands which sets of customers are likely to remain inactive or defect, and can anticipate which defective contacts are most likely to return.

4. Predict how much customers will spend.

AI predicts, at an individual level, what a customer’s next cart value is going to be (then automates execution of the best content for them). In short, you’ll be able to say:

  • “Customer A will likely spend $160 on her next purchase.”
  • “Customer A will buy every 90 days whereas Customer B will buy every 2 weeks.”
  • “Customer C, who used to be a high-value customer, is going to churn in the next 30 days.”

5. Predict overall revenue.

Based on cumulative projections for individuals — as well as the trajectory of business growth — you can anticipate revenue over time. Business leaders and marketing executives can say:

  • “The business will make $X in revenue next quarter.”

Not only will this allow better attribution and planning, it will finally let marketers directly prove the impact that their work has on the bottom line!

Successfully Integrating AI into E-commerce Strategies

Enterprise AI adoption is a journey, not necessarily a destination. This is especially true for big companies with millions of database contacts and multiple touchpoints already being used.

In fact, that’s one of the main challenges to AI adoption — dispersed systems and data silos that keep information separate from one another. AI works best when it’s implemented from one source where different channels are connected and integrated. Other challenges inhibiting organizations’ ability to use AI may include:

  • A “business-as-usual” philosophy. AI adoption requires a mind shift that more traditional or conservative brands may simply be unwilling to make.
  • Misunderstandings and misconceptions. Myths about AI have been circulating for way too long now. They need to be dispelled and replaced with facts and case studies.
  • Proving the value of AI. It’s difficult to prove return for an untested technological capability (which you’ve never used before). The C-Suite usually wants concrete evidence that you’ll be able to drive measurable impact (which is hard to “prove” upfront). Ironically, this ROI is the very thing AI offers.

Whether you’re adopting AI across the enterprise or wanting to successfully integrate it into existing strategies you have up and running, it all starts with one question: WHAT IS THE OBJECTIVE?

One surefire way to waste your time and money is to “do AI” just because… because it sounds neat, or because other brands are doing it. The most successful companies invest in AI-enabled technology because they want to achieve something they can’t currently do, like:

  • Segment the database efficiently using less people and less time to get a campaign out the door
  • Increase conversions across the site — such as automation of first-time buyers to active
  • Convert churning customers back to active, or prevent them from churning at all

Successful enterprise AI adoption requires clear objective setting and a clear understanding of whether AI is the right enabling tool to make it happen. At the end of the day, the process requires some up-front research, diligent vetting of tools and options, and, truly, an unyielding willingness to jump in.

Related Content: From Reactive to Proactive: Unlocking New Dimensions of Marketing with AI

Essential AI Tools for E-commerce Businesses

Tools and tech should be the last “prong” to be considered when thinking about AI. We’ve covered how the “Objectives-Strategies-Tactics” model works, so ensure you’re beginning with what you want to accomplish, along with your strategy for getting there prior to looking at tools, features, and functionality.

When it comes to AI tech, what kind of tools are available within a unified solution, and what kinds of capabilities can you expect with them? Here’s a few to keep in mind:

For best results and optimal performance of the algorithms, these solutions and more like them should all be interwoven within a unified marketing platform.

Final Thoughts

AI is here and here to stay. What’s more, it’s helping e-commerce organizations transcend the ways they can predict customer events, respond to them, and drive value at an expedited pace.

I can’t overstate the importance of going from reactive to proactive. Getting in front of customers proactively goes beyond real-time marketing to predictive marketing. This is the future of our industry.

Handpicked Related Content:

____________________________________

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

Raj Balasundaram

About the Author


Raj Balasundaram is SVP of AI at Emarsys, where he helps leading brands leverage their digital platforms and data to out-maneuver competitors and achieve superior financial results. Raj has delivered presentations and talks at a variety of global conferences and road shows including #DMWF Expo Global, evangelizing the art of data intelligence-based marketing. Prior to Emarsys, Raj worked at both Oracle and ExactTarget.

Connect: LinkedIn@RBalasundaram

 

 

 

 

Raj Balasundaram
Raj Balasundaram
SVP of Artificial Intelligence
Yes, I would like to receive Emarsys email updates on products, services, news, and events. I can unsubscribe by using the "unsubscribe" link at the bottom of each email, or by emailing unsubscribe@emarsys.com.

AI is Changing the Marketing Game.
LEARN MORE