A Leading EdTech Company Builds a Customized Adaptive Engine to Make their Product More Competitive​

Case Study

A Leading EdTech Company Builds a Customized Adaptive Engine to Make their Product More Competitive​

The customized adaptive engine for assessments increases the test counts and improves scoring accuracy significantly.​​​


The client is one of the world’s leading education companies and the largest global comprehensive provider of educational assessment products, services, and solutions that promote learning and academic achievements. ​


The client is leading the evolution to more engaged, interactive forms of learning that cater to each learner’s unique needs. They aspired to make their product more competitive to increase the revenue share with higher student coverage by delivering improved learning outcomes and wanted to provide value-added services for one of their assessment platforms that helped colleges make accurate course placement decisions. They desired to implement an advanced adaptive algorithm for a set of new tests introduced in the application. Further, the client was also looking to generate personalized learning paths and provide guided course placements based on the learner’s performance. ​


With our deep expertise in 1PL (One Parameter Logistic) and 2PL in the IRT (Item Response Theory) model, Ness helped set up the required customized adaptive engine for the new set of assessments and developed a more tailored assessment system based on each learner’s potential. We set up a mechanism within the application to create rules that would apply to the score reports, determining personalized course placements based on a learner’s performance while also taking the learner’s other academic factors and background into consideration. A specific set of assessments decided the accuracy of personalized learning path recommendations for a learner, which helped derive the learner’s proficiency in different test sections within an assessment. These solutions helped colleges and institutions assess the incoming learner’s skill levels to decide the courses that best fit the learner.​


With the customized adaptive engine in place, Ness’s solution increased the test counts for the assessments by 13% to 3.2 million tests a year. It also improved the scoring accuracy by 9%, declined the score wait time by about 2%, and gained accessibility to 20% more learners. Further, smart implementation of the learning path generation made the resources easily consumable by learners.​