AI's Role in Delivering Consumer Personalization: Key Takeaways from CommerceNext 2024

pink and purple speech bubbles
Published on
Jun 20, 2024
Written by
Kasey Hickey
Written by
No items found.
Thank you! You've been subscribed.
Oops! Something went wrong while submitting the form.

AI is transforming the way brands engage with their customers. Here’s how two seasoned retail industry leaders are using it to augment their teams’ impact.

At the recent CommerceNext 2024 event, marketing industry leaders Chris Maliwat from Victoria Beckham Beauty and Brendan Roeschel from Fresh Clean Threads shared their experiences and insights on integrating AI into their marketing strategies. 

Moderated by Attentive VP of Client Strategy, Elizabeth Ray, the discussion covered a range of topics from the initial excitement about AI to practical applications and future prospects in customer engagement. Here are some highlights from their conversation.

On the potential of AI in Marketing

Ray: What are you most excited about when it comes to AI? Do you think this will truly be a win-win for both brands and consumers?

Maliwat: Personalization and AI are at interesting points in their history. Ten years ago, personalization could mean 80 different things. AI is in a similar place today. The connection for me between AI and personalization is AI's ability to help us achieve true 1:1 personalization at a scale in ways we wouldn't have been able to do previously without these models.

Roeschel: It’s 100% a win-win for both the customer and the brand. We were using it for a use case of when customers come to site and abandon. In that scenario, you really have to meet the customer where they are. And while chat is great for capturing content on-site, SMS is that channel for us to continue the conversation throughout their journey. 

Without AI, it’s really hard to meet the customer where they are and at scale. By sending hyper-personalized outbound messages that are both relevant and engaging, we can effectively encourage our customers to respond to these messages and help them overcome any barriers to purchase.

On getting ​​started with AI 

Ray: In relation to AI, one of the more common questions I get from marketers is how do I get started, where do I get started. Can you describe Victoria Beckham Beauty’s approach?

Maliwat: At Victoria Beckham Beauty, we recognize that AI is still in its infancy. Therefore, we're adopting a very exploratory approach throughout our entire marketing lifecycle. We've opted for a holistic strategy that involves understanding every aspect of the marketing lifecycle, from inspiration and discovery to identifying motivators and barriers. Currently, our use of AI is mostly behind the scenes—augmenting our teams and helping them navigate through the vast universe of possibilities. We're starting to dip our toes into audience segmentation with AI.

For marketers looking to experiment with AI, I'd suggest identifying use cases that make sense throughout the entire lifecycle of your marketing efforts. It's important to focus on the prep work that happens before anything consumer-facing is developed.

Roeschel: We've implemented Attentive AI in our welcome and cart abandonment journeys.

To gauge the effectiveness of AI, we started with a 90-10 holdout—meaning we initially excluded 90% of our audience to minimize risk and maximize control over the experiment. This approach allowed us to incrementally test and adjust our strategies until we reached a 50/50 A/B test between AI-driven and static copy. 

The results were impressive: we captured 50% more revenue from abandoned carts and saw a 20% increase in overall built revenue. This not only contributed to substantial top-line growth but also significantly reduced our customer acquisition costs just given what we would be spending on ads to drive traffic to our site. 

On learnings from AI implementation

Ray: So the numbers are great, but I'm sure there were some learnings and surprises along the way. Can you describe some of those?

Roeschel: Absolutely, there were a few key learnings. Initially, we noticed that the AI was responding to messages that didn't necessarily require a response—like simple acknowledgments such as "thanks" or "will do." We quickly adjusted our approach to focus the AI on responding to actual customer barriers rather than offering acknowledgments. This was one of our first major tweaks.

Another significant focus was on leveraging what we learned about customer attributes, which we tag in our segments. This aspect was particularly cool because our brand ethos is "fit for everybody.” With AI, we were able to differentiate our responses based on specific customer data—like adjusting our approach for customers who abandon carts and might have different needs based on their size, whether taller or smaller. This customization allowed us to be more relevant and less reliant on broad data scraping, leading to more effective and personalized marketing.

On building trust and buy-in for AI

Ray: With gen AI tools, there's a lot to be skeptical about. It's new. It can be unpredictable and this, of course, can lead to concerns on the C-suite and leadership level. Chris, representing leadership here today, what advice do you have to build that trust and buy-in?

Maliwat: AI is really great at processing and helping you sift through and find insights in the data. For us, it was pretty easy to get leadership to understand the value of AI by really leaning on our hypothesis that we need to become more and more data-driven. Using AI to help us think through what all the data means and what actions we can take based on the insights we see was really our key to getting the leadership team’s buy-in. 

Final thoughts and advice on AI adoption

Ray: Can you talk to me about what's next on the horizon and what you're excited about and celebrating?

Maliwat: At Victoria Beckham Beauty, we're all about modern luxury, which for us is about convenience and anticipation. Technology's ability to understand people better than they’re able to understand themselves sometimes is something that we're very excited to see apply to our world. 

Our audience expects Netflix-level personalization and how to apply that level of thinking within our domain is what I’m excited about it. We’ve very much been a broadcaster: we have an idea and we disseminate it–everyone gets the same message. But now we’re able to to have personalized messaging that anticipates customers’ needs—it’s in our voice, and allows us to build relationships with customers at scale.

Roeschel: We're stoked about what we've achieved with AI-powered SMS. Beyond SMS, we're thinking about AI site personalization, like customizing pop-ups at the user level, and considering various customer touchpoints–like chat–to help us achieve an even bigger lift. 

Ray: Are there any other parting words, final thoughts that either one of you wanted to share?

Maliwat: If you or your management have questions about AI, you need to get specific. The more specific that we can get about ‌use cases and benefits, the faster it is to‌ bring people on board.

Roeschel: Crawl first, then walk, then run with AI. A lot of people think it's a switch that just flips on, but that’s not how I'd suggest going about it. Pick a specific part of the funnel, a specific channel, and consider running a holdout test. Measure and go from there. You can always expand into different customer communications later.

Looking to learn more about how you can start implementing AI into your marketing strategy, explore our AI resources and case studies and join our waitlist.

Related Articles