Sentient Ascend: AI-Based Massively Multivariate Conversion Rate Optimization

Abstract

Conversion rate optimization (CRO) means designing an e-commerce web interface so that as many users as possible take a desired action such as registering for an account, requesting a contact, or making a purchase. Such design is usually done by hand, evaluating one change at a time through A/B testing, or evaluating all combinations of two or three variables through multivariate testing. Traditional CRO is thus limited to a small fraction of the design space only. This paper describes Sentient Ascend, an automatic CRO system that uses evolutionary search to discover effective web interfaces given a human-designed search space. Design candidates are evaluated in parallel on line with real users, making it possible to discover and utilize interactions between the design elements that are difficult to identify otherwise. A commercial product since September 2016, Ascend has been applied to numerous web interfaces across industries and search space sizes, with up to four-fold improvements over human design. Ascend can therefore be seen as massively multivariate CRO made possible by AI.

Potentials of artificial intelligence in digital marketing and financial technology for small and medium enterprises

“…Firstly, a webpage is designed and users are directed to it for measuring how well it converts. "If the conversion rates are statistically significantly different, the better design is adopted" [46]. AI has made its way into boardrooms and has the potential to change corporate decision-making forever.…”

Ascend by Evolv: Artificial Intelligence‐Based Massively Multivariate Conversion Rate Optimization

“…When the Evolutionary CRO methods were taken out of the laboratory and into the real‐world application, it became clear that there were new and interesting challenges that needed to be met. First, in the original Evolutionary CRO framework (Miikkulainen et al 2017a; 2018), the evaluation of each candidate is performed in a static fashion. A fixed amount of traffic is allocated to each web design.…”

Research Updates from the Deployed Applications of the 2018 Innovative Applications of Artificial Intelligence Conference (IAAI‐18): Editorial Introduction to the Special Articles on Innovative AI Applications

“…This is no small feat! IAAI-18 had a diverse set of seven papers in this track: one on improving government human-resource services deployed in three cities (Zheng et al 2018); one on improving employee shift schedules with complex constraints used in two types of businesses (Hoshino et al 2018); one discussing the learnings from deploying sketch technologies in two science, technology, engineering, and mathematics classes (Forbus et al 2018); one on machine-learned revenue forecasting models used in a major software development company (Barker et al 2018); one on a question-answering system used in an enterprise information technologies helpdesk (Mani et al 2018); one on an interactive learning system used to optimize multivariate deployed user interface changes available commercially (Miikkulainen et al 2018); and one that was a unique insight into an analytics, analysis, and dynamic adjustment system integrated into the products of several game studios through their major publisher (Kolen et al 2018).…”