For digital businesses, ensuring the user experience is optimized and personalized is a key strategy for keeping users engaged and retained over time.  Alongside a great user experience is a marketing team that keeps users engaged with your brand even when they are not actively on the site.  Typically offsite targeting of users encompasses one of the largest portions of a marketing spend, and ensuring you are targeting the right users can save large portions of your budget. This is especially true for campaigns being targeted at known customers, as the past behaviors of those users can lead to increasingly more relevant marketing campaigns.

This post will dive into the world of marketing campaigns, exploring how machine learning and user predictions can help retarget and re-engage the right users with the right message and keep the campaigns ROI positive.

Data Driven Marketing

The foundation to personalized marketing is understanding user behavior over time.  By understanding user activity across a customer base, companies can identify and predict the users most likely to convert and be profitable for the business over time.  What events should be tracked to enable effective predictions? Events can be as simple as page views and searches while also tracking conversions events such as purchases and subscriptions. 

At Vidora, we provide a Machine Learning platform that ingests these behavioral data sets and transforms them into user predictions.  These predictions are automatically retrained and updated with new data, ensuring these predictions are always up to date and can be integrated directly into the campaign workflow.  Examples of how companies are using Machine Learning to power marketing campaigns include: 

  • Predict which users are most likely to convert in the future, and stop retargeting users with the lowest-likelihood
  • Recommend and rank the top items for each user to create a dynamic and varied set of content to power retargeting

Immediately save 10% on Retargeting Costs

Not all users who visit the site are ideal users.  Some users are highly engaged and convert often, while others are infrequent visitors without any conversions.  When it comes to retargeting campaigns, it’s important for the ad spend to be optimized to target only users likely to re-engage. Otherwise, click through rates on the campaigns could be low leading to higher spend per conversion.

Using Future Events predictions, you can rank users on their likelihood to convert in the future, be that a purchase, subscription, or filling out a lead form.  This prediction can prove valuable for retargeting campaigns as it helps identify those users with the lowest likelihood of re-engagement.  By excluding these low-likelihood users from the retargeting campaign, even just the bottom 10% least likely, the ad spend can be optimized only for those users likely to click through. This targeting can save marketing teams large amounts of money.

Cortex provides a no-code solution to target high value users for offsite re-engagement with a few clicks. These users can be easily exported for offsite targeting. Learn more about exporting segments here

Dynamically Retargeting Powered by Content Recommendations

Even when a user isn’t live on the site they can still be engaged with personalized content.  However, a message or ad that is shown which is irrelevant to a user may encourage them to not return to the site at all.  

Content and category recommendations help solve this problem for many marketers.  By using the historic data from customers, it becomes easy to learn user preferences from past behaviors.  This information can then be used in marketing campaigns to ensure the message is always relevant to the end user.  Such examples include:

  • Individual item recommendations for an eCommerce company,
  • Video recommendations for a media company, or 
  • Category/Section recommendations for a news company

Cortex is powerful enough to generate recommendations to increase purchases, clicks, or long-term brand engagement.

In Closing

The above examples show how you can use future predictions to quickly save 10% of your retargeting spend while also increasing the conversion rates by powering the campaigns with dynamic recommendations.  These cost savings and increases in conversions help marketing teams increase the ROI from their retargeting efforts.  If you’d like to learn more about how your team can leverage Machine Learning in your campaigns, reach out to use at!

Want to Learn More?

Schedule a demo and talk to a product specialist about how Vidora’s machine learning pipelines can speed up your ML deployment and ultimately save you money.