Anticipate customer requests with Webtrekk
and create a conversion and return forecast for every website visitor

Based on retrospective data, machine learning makes it possible to forecast future customer behavior in the customer journey. What is the expected order value? What is the customer’s lifetime value? What is the probability that the prospective customer will turn into an actual customer?

The forecasts are continuously adjusted and are based on a variety of attributes, such as, visit duration, page views, click speed, device class, value and quantity of products viewed, and much more. Real-time predictive analytics requires a lot of data, data expertise, and data science. We are proud to be one of the few service providers that can offer you this functionality. Depending on the expected customer behavior and customer value, you can concentrate your investments on need-based marketing measures.

Benefits of Predictions:

  • Optimized remarketing with a focus on the most promising website visitors meaning often additional marketing activities are not required
  • Strategically planned conversions – depending on probability data automatically determine which onsite offers or actions should be initiated
  • Predict future sales trends automatically and interlink them with the expected customer lifetime value
  • Segment high-converting visitor groups in a targeted manner and address them using exit-intent triggers

Webtrekk’s Predictive Analytics won a Gold German Stevie Award

In 2019, Webtrekk received the German Stevie Award for Predictive Analytics. The jury recognized Webtrekk’s innovative approach, which not only enables real-time forecasts of conversion probabilities for website visitors, but also empowers companies to become directly involved in the conversion process.

Frequently Asked Questions:

Can I influence visitor activity in real-time to increase the likelihood they will convert?

Yes, for the first time, this is now possible. Depending on the conversion probability, companies can now initiate targeted actions on their websites and in their online shops. For example, it is recommended quickly showing a ‘free shipping’ banner or message to visitors identified as having a decreasing conversion probability (as a metric). Furthermore, on the basis of their conversion probability, website visitors can be placed on remarketing lists in order to optimize targeted bid strategies.

What is the methodology behind Predictive Analytics?

Such precise predictions are made possible by machine learning. By searching for behavior-based patterns, our algorithms build a comprehensive data set. The more extensive the data volume, the more precise the resulting forecasts become.

Which attributes are factored into the forecast calculation?

Our self-learning algorithms consider a multitude of attributes. These include metrics such as visit duration, page views, click speed, device categories, the value and number of products viewed, and the shopping cart value.