Increasing conversion rates is a big challenge for eRetailers. World Wide Lighting (WWL), a leading European online retailer specializing in lighting products, knew that it needed to reduce dropouts before and during checkout to increase revenues. The company turned to Celebrus for its machine learning platform and OnMarc, a Dutch-based digital consultancy that offers MarcSense Retain that predicts visitor behaviour on eCommerce stores.
The combination enabled WWL to increase personalized messages in real-time to prevent drop-offs. OnMarc has worked with Celebrus for years and turned to it again to deliver data collation of click, scroll and cursor movement data.
This new data set opened up huge potential. Kevin van Kalkeren, Manager of Product Management & Data Science at OnMarc, recalled, “Our customers already got real value from their data, but we had never looked into the potential of all that movement data combined at scale. So, with the challenge of the exiting visitor in our minds, we set out to unlock the potential of user’s movement data as well.”
Using the data, OnMarc identified visitors’ intent, linking what visitors physically did with their likelihood of leaving the site. With that analysis, they built algorithms that could propose personalised messaging to reduce dropouts. The study showed that MarcSense Retain can predict about 60% of visitors’ intent to leave within the first half of their visit.
The results were impressive.
The ability to detect whether a prospect is likely to leave a site, even before they consider doing so, enables eCommerce vendors to react. With personalised messages that entice the customer to stay longer and purchase. While OnMarc gave no specific examples of how these work. It could be based on the initial searches and views the customer went through. Perhaps providing alternative products that the customer might have missed that still broadly met their criteria.
Martijn Brouns, Marketing Manager at World Wide Lighting, commented, “OnMarc and Celebrus help us by revealing the true needs of our customers in a unique way. This allows us to communicate much more proactively and allows us to become more relevant and proactive in our communication. This cooperation is therefore very valuable, both operationally and strategically.”
There were some significant results after the deployment:
- WWL now have a 90% success rate prediction on the intent to leave
- 60% of abandoners identified within the first half of the visit, giving time for intervention
- An 800% performance increase using model vs without
- A 48% reduction in dropout at checkout
Bill Bruno, CEO of Celebrus, commented, “Since 2022, World Wide Lighting has used Celebrus data to power MarcSense Retain and gain valuable insight into customer behaviour. We are thrilled with the results seen so far and look forward to continuing to help maximize the efficiency of customer marketing efforts by leveraging live-time, comprehensive data.”
Enterprise Times: What does this mean
The Celebrus data management platform enables real-time data capture, enabling organisations to leverage that data quickly. Within eCommerce, watching a customer leave is frustrating. The combination of Celebrus and MarcSense has enabled WWL to not only detect how a customer leaves a site, it does so well ahead of the point they leave, thus giving WWL time to react and do something to prevent it.
What isn’t noted in the case study or press release is how WWL can detect which actions they take to make a difference. For example, do pop-ups further frustrate a customer or does a simple ticker message offering free delivery or another incentive retain their interest for longer?
Celebrus and MarcSense offer eCommerce vendors the opportunity to build interactive and intelligent websites. That enable customers to enjoy an experience akin to dealing with a store salesperson who adapts to physical behaviours and what customers say.