AI in eCommerce: Examples & Applications For Every Budget

Staying within data privacy laws is the crux of good personalization. Here's how to win the AI in eCommerce game. 

Are you looking for ways to scale and automate your processes? 

Of course, you’re not alone. Being reactive is top of mind for all eCommerce professionals. Especially since the market – and with it consumer behavior – is changing so quickly. 

Against this landscape, the impact of AI in eCommerce is HUGE. 

After all, AI is being leveraged by eCommerce pros precisely to:

1. Automate
2. Scale

But as fast as AI is being used to help retailers out, the ethical argument behind leveraging AI is also a growing concern. How brands are collecting and using customer data is becoming more and more restricted. 

The challenge many of you are experiencing as you navigate AI in eCommerce is, on the one hand, a cookie-less future and, on the other hand, growing demands for personalization.

Plus, your shoppers are missing human interactions as they shop online. Two in ten shoppers lack the advice of sales assistants online. 

How do you personalize the customer journey while continuing to stay within data protection laws? 

And how do you leverage AI while retaining the human in the online shopping experience?  

This article will give you 5 AI in eCommerce examples unlike what you normally read. I want to make sure you have the tools you need to get started with AI today. 

Which means we'll stay away from AI in eCommerce that mentions things like augmented reality (AR) or the metaverse, or things you’re already using like smart merchandising or filters. 

Instead, my goal is to show you affordable and actionable examples of AI in eCommerce you can use right now. 

Let’s begin. 

What are AI and machine learning in eCommerce?

Free Guy movie: About a video game character who gains autonomy and consciousness. 

Unlike the AI technology from your run-of-the-mill Sci-Fi films, AI in eCommerce refers to the way customer and product data is collected and analyzed to automate processes and scale personalization. 

Meaning that your eCommerce AI doesn’t have to look like a warehouse robot (but a good example of AI being used to automate, albeit for brands with bigger budgets).

Source

eCommerce AI means teaching algorithms to a machine so that it can learn from patterns of data and generate personalization based on the data it finds. For example,

• Smart retargeting
• Social ads
• Tailored webshop experiences based on historical data

On a higher level, these examples are aspects of the customer experience that will leverage AI and machine learning. 

But these are also things that require lots of customer data and cross-channel tracking. Which, as we know, is becoming more problematic as customers are wary of third parties using their cookies for retargeting (see Apple’s ITP policy). 

Long story short, we can’t talk about AI and machine learning in eCommerce without talking about data privacy.  

AI in eCommerce marketing is a question of data privacy 

Orwell’s 1984 predicted a big brother state where technology facilitates supervision. This dystopia holds many warnings for present-day societies. 

The impact of AI in eCommerce is a question of data privacy. According to ROI Revolution:

• 79% of Americans feel concerned about how their data is being used by companies
• 42.4% of consumers don’t accept sharing their cookies when a website asks them
• Only 16% of consumers say current data privacy laws are enough for them
• Cookieless data is top-of-mind for 36% of digital marketers this year

As customer data becomes more protected, retailers are turning to product data to lead the way. This marks a shift in the traditional product-centric vs. customer-centric divide. 

Instead, the data feedback will be a continuous loop between product and customer (with anonymous tracking and cookies that stay in session). 

Love Stories personalizes product assortments based on the season and offers a style guide in their summer email campaign.  

Switching the focus to product data will allow brands to personalize their product offering in terms of:

• Product messages 
• Product promotions
• Product merchandising and assortment 

Matching users to the perfect products for them is one way AI in eCommerce is revolutionizing the data game. 

“Product personalization brings back several elements that retailers have neglected since the shift from product to customer-centricity. By understanding and properly communicating the characteristics of your products to the right person at the right time, you are catering to both goal-oriented and browning shoppers while reducing choice overload.” - Janelle De Weerd, Chief Operating Officer at Crobox

To leverage a tool that collects first-party data (i.e., customer ‘sanctioned’ data that stays within the GDPR), check out Crobox’s Product Finders. 

So if AI in eCommerce is being regulated, while at the same necessary for growth…where do you start?

4 AI in eCommerce examples: How to personalize while staying within data privacy laws

1. Personalized product recommendations 

You know that fight you get into when you’re tirelessly browsing Netflix and you and your partner/family/cat can’t decide on anything? 

Netflix actually makes 75% of its sales from recommending movies and shows to its users. Otherwise, we’d all suffer from choice overload and would probably quit. 

It’s no wonder that Netflix is still one of the most popular streaming platforms today. 

The power of personalized product recommendations lies behind its success. And, in retail, there’s nothing better than recommending products, especially in a market saturated with so many similar product choices. 

ASOS

You can make recommendations to your customers based on:

• Complimentary add-ons: Provide an instant add, especially if you’re offering this on the cart page. Here, the recommendation should fit the shopper’s interests or can cause friction.  
• Set completion: For fashion brands, this means helping shoppers complete an entire outfit. But you can also bundle complementary products together. If you opt for a bundle, make sure to discount the final price. 
• Similar product themes: This provides cross-category recommendations based on the interests of the customer– a good place to leverage your styling authority as a brand, with things like ‘shop the look’!  
• Frequently bought together: Leveraging Social Proof, you can collect product data based on which products others have bought in tandem and then recommend to drive up or cross-selling.
• New products/lines: This leverages the behavioral principle of Novelty, where many individuals will be more inclined to purchase a new product or line if it is labeled as such. 

What webshops leverage product recommendations best? Check our design trend report, where we analyze the most persuasive webshops based on the designs they use. 

2. Dynamic Messaging 

Crobox Dynamic Messages on ASICS’ webshop. 

72% of your shoppers will only engage with product messages specifically tailored to their interests. 

Dynamic Messaging leverages AI to personalize the messages your shoppers see at the right time. Dynamic Messages act as triggers that drive purchase behavior with visual cues that give customers more information about products, stock, or are persuasive messages designed to influence their choices. 

At Crobox, our Dynamic Messages are built based on theories originally found in psychology and behavioral economics. 

When executed well, they not only increase metrics like CTR and CR, but they make the customer journey more seamless by assisting in the decision-making process of your customers.

This means they effectively reduce choice overload, pointing out products that matter more to the customer (i.e., relevant information most likely to drive click-behavior). 

2. Conversational intelligence

One of the biggest hurdles for bringing the in-store experience online is the loss of actual humans in the eCommerce experience. 

But the upside of this for online shopping is that there is more intimacy, and customers can reveal more about their shopping interests and choices free from pressure. 

Which makes leveraging conversation still a good idea online. This is where your AI comes in, to make sure that conversation is natural. Conversational intelligence in eCommerce looks like:

• Conversational search: Integrate NLP (natural language processing) into your SEO. So it’s not about optimizing for the word “running shoe”, but for how people search for running shoes, e.g., “best running shoes for trail”. Searches that include  “I” in Google have reportedly increased by 65%.
• Chatbots: These guided selling tools communicate with your customers, but are often used for FAQs or to answer specific brand questions about delivery, in-store locations, or business hours. The most powerful chatbots will leverage conversational AI.
• Product Finder Quiz: A Product Finder quiz is basically a quiz that asks questions to the customer to determine their personality or profile, and then recommends them products, a set of products, or a category page based on their answers. The questions here are dynamic, based on the user’s answers, making this the perfect conversational intelligence tool. Not all Product Finders can offer dynamic answers – but Crobox does! To stay within data privacy, our Product Finders collect first-party data. Learn more.

3. Product intelligence 

Product intelligence is the process of gathering and analyzing your product data and making it actionable on your webshop or omnichannel. 

Your product analytics similarly answers the who, what, where, and how of your products. And based on these data points, product intelligence answers the why. 

In short, product intelligence shifts the focus to product data to personalize the customer experience. Here are a few examples of product intelligence at work:

• Predictive inventory management: By keeping track of product data in one place and analyzing changes to this data, you can begin to forecast future risks to your stock. This takes into account your supply chain, your assortment, and your shopper’s purchasing habits (historical and predictive data). 

• Personalize product taxonomies: The browsing behavior of individuals on your webshop will highly depend on your product merchandising and the categories available. Test which categories and product promotions drive behavior and then optimize these to fit in line with the shopper’s demands. 

• Promote products cross-channel: With your eCommerce AI in place to help you collect and leverage product data, you can start seeing insights on which product messages drive click behavior. For example, if sustainability is the best performing message for your line of coats, then you can bring more sustainability messaging in-store. Or, you can create a marketing campaign based on sustainable coats and deliver this through email and social. 

The future of AI in eCommerce: Where do you start?

The fact of the matter is, the impact of AI in eCommerce is no longer a prediction but proven by retailers’ KPIs. 

But for an eCommerce pro, this means coming up with a strategic approach to integrating AI and machine learning capabilities within your organization. 

So, how do you get started? Here are a few steps you can start with. 

Step 1: Invest in AI education and architecture will follow 

Take time to introduce innovation. Many retail teams suffer from legacy issues, and these could take time to undo. 

Instead of a complete 360 for your organization to AI, focus on educating your peers about machine learning and integrating it within your processes. 

Step 2: Acquire the right AI tools to help 

Then, understand what tools are right for you and make the switch. While many of you are working within a ‘Global’ IT structure, tech vendors can be a great workaround to give you some autonomy back. 

• If you’re looking for a search tool, Empathy is a great data-privacy first company that helps optimize the customer journey. 
• To analyze UX with real EEG scanning to see where attention spikes and dips in the brain, check out Braingineers. 
• If you’re looking to optimize the product discovery process, Crobox has a subset of features that include Product Badging, Product Finders, Product Stories, etc. Our mission is to stay within data privacy laws, and we personalize using AI and psychology.

These are only a few AI tools you can use to really personalize the eCommerce experience, giving you control over your webshop while staying within collecting first-party data. 

Step 3: Don’t let your AI do all the work, track micro-conversions

Otherwise known as Always Be Optimizing, making sure everything is up to date will ensure your AI architecture is solid for the future. 

Hotjar is a heat map tool.

Micro-conversions like:

• Heat maps, 
• Eye-tracking, and
• Transactional data

Are all data points that stay in session and will help you optimize the customer experience from the ground up, so to speak. 

At Crobox, we track the performance of Dynamic Messages in terms of clicks. Learn more. 

Tracking and optimizing micro-conversions will also facilitate behavioral segmentation.

Which, again, is something that will be made easier by AI, but should be overseen by someone on your team. As the behavior of customers on your webshop changes, so too will your behavioral segments. 

To sum it up, the future of AI in eCommerce will be:

Source

• Fully dedicated AI or innovation teams
• More predictive models 
• Advanced segmentation
• Hyper-personalization 
• Improved customer experiences online

But make sure your AI strategies are always updated. Without the human behind the machine, it will become more difficult to stay within AI laws while innovating per individual. 

Get started today!

So there you have it. We didn’t touch upon AR, VR, or the metaverse – big topics for a Monday (Tuesday, Friday?), I know. 

Instead, I hope this article brings you down to Earth a bit more. Brands are using AI in eCommerce today to scale and automate the customer journey. 

Which still relies on the most frictionless eCommerce experiences with small things like product recommendations, Dynamic Messaging, conversational intelligence, and product intelligence coming out on top. 

For more information on how Crobox leverages machine learning and AI, feel free to reach out!