Skip to content

World Wide Lighting reduces drop-off 48% using online visitor behavior to predict exit intent in real-time



Reduction in drop-out at checkout 


Increase in performance in preventing drop off of website 


Success rate predicting intent to leave

About the customer

World Wide Lighting is a leading and fast growing e-commerce platform offering lighting solutions for the business market. They carry top brands such as Noxion, Philips, and Osram. Sustainability and energy saving is very important to them and thanks to their wide range of LED products, they're able to facilitate a cleaner planet together with their customers.

Use Cases

  • Increase in website performance
  • Reduction in dropout
  • Retention increased ROI


OnMarc was tasked with finding a solution to identify exit intent on client websites. They’d been working with Celebrus for years, enabling real-time collection of click, scroll, and cursor movement data. “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.” says Kevin van Kalkeren, Manager Product Management & Data Science at OnMarc. “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.” 

To assess if exit behavior was universally detectible, OnMarc conducted a groundbreaking study to evaluate the connection between mouse movement and emotion. The results clearly showed a direct link. Through testing, they uncovered the behaviors that determine if a visitor is going to leave the website. Then, they created a pipeline to make it work in real-time.


Our customers already saw real value from their data, but we 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.


Use Case

World Wide Lighting, an online retailer, faced the common challenge of visitors frequently leaving their website without making a purchase. They wanted to increase marketing ROI by proactively identifying customers who could potentially be retained.  

They struggled with understanding visitor behavior on their website, leading to ineffective popups that appeared too late. They needed to gather better and faster information to deliver personalized and relevant messages in live-time, aiming to retain visitors and convert them into customers.  

World Wide Lighting implemented OnMarc’s MarcSense Retain, powered by Celebrus, to capture all available data and identify patterns in clicking and mouse movements that signal intent to leave – long before the customer would actually abandon the website. With this crucial information, their decisioning engine fuels the website's content management system to deliver personalized content that encourages the visitor to stay. 

For World Wide Lighting, OnMarc refined the models to determine where they get the best results. For example, top of funnel visitors are given more broad personalization or a lower offer, while bottom of funnel are given a higher value offer. 

The Solution

With Celebrus taking care of the data, OnMarc leveraged Amazon Web Services to build MarcSense Retain. The solution identifies visitors who are likely to leave World Wide Lighting’s website with a high level of accuracy by leveraging Celebrus mouse movement data with probabilities up to 90%. Timely interventions are crucial, as waiting until customers reach the point of clicking the exit button is too late. Leveraging live-time data from Celebrus, the models detect exit intent early, sometimes 10 times faster than the actual exit, giving the team sufficient time to engage and retain customers before they’re lost. The results were impressive; within the first half of a visit, MarcSense Retain can predict about 60% of visitors’ intent to leave. And delivering personalized content to these visitors has a significant positive effect vs. standard “call me now” messaging. 


By identifying potential abandoners, and delivering personalized messaging in real-time, World Wide Lighting experienced a 48% reduction in checkout drop-off.  

Since 2022, World Wide Lighting has used MarcSense Retain and Celebrus to gain valuable insight into customer behavior and enhance their ability to deliver targeted messaging. This approach empowers them to engage visitors based on their mouse movements, increasing customer retention and conversions. Celebrus’ live-time, comprehensive data powers better attribution, predictive modeling, and next-best-channel recommendations to maximize the efficiency of their marketing efforts. 


success rate prediction intent to leave 


of abandoners identified within first half of visit


performance increase using model vs. without


OnMarc and Celebrus helps us by revealing the true needs of our customers in a unique way. This allows us to communicate much more proactively and relevantly with them.

Martijn Brouns, Marketing Manager at World Wide Lighting

48% Reduction in drop-off


This online retailer did it, and you can too!

Celebrus is the radical data solution that uncovers the deepest insight to reveal true customer context and identity - with a single line of code!

Finally solve gaps in data capture, customer context, connectivity with full compliance to elevate your marketing and personalization and explode revenue.

Connect with us today to experience the power of Celebrus for yourself!