Introduction to Othot

What is Othot? 

Liaison’s Othot cloud-based predictive analytics tools for higher ed uses artificial intelligence to deliver insights that help you make informed decisions across the student lifecycle. Through simple interactions, our software identifies who is most likely to enroll and to retain and persist, and where to focus your resources for the greatest positive outcome. You can do more in less time and with fewer resources.  

Our Othot analytics software delivers made-to-order predictive and prescriptive models with: 

• Immediate predictions 

• Explainable AI 

• Continuous intelligence 

What is available in the Othot software? 

Cultivate a deeper understanding of the overall predicted outcome and individual student impact with AI-driven data technology that provides insights into the behaviors that influence recruitment and retention. 

• Explainable AI models 

• Real-time aggregate and individual predictions 

• Simulated prescriptions and optimization 

• High-level dashboards 

• CRM/SIS integration 

Who uses Othot? 

Othot is designed to be useful for nearly all stakeholders and team members in enrollment and student success. 

• Executive Leadership will find value in viewing dashboards and high-level predicted insights. 

• Directors and Managers can better understand historical trends and what’s leading to the predicted outcomes to best develop strategy, allocate limited resources, and be proactive throughout the life cycle. 

• Recruiters and Advisors will be able to understand what impacts the individual students they are responsible for with more insight to guide conversation and intervention than ever before. 

Who builds the model and when?

To ensure quality, models are built and maintained by the Othot Data Curation and Othot Data Science teams only. Models are trained using institutional historical data, ten fields appended from census data, and some Othot derived fields. 

Following the initial implementation, generally, your model will be rebuilt at least once a year after the active year’s cycle is complete. In this process (called a Rollover), the newly completed year of data is added to the historic/train data and used to make predictions for the upcoming year. 

While we attempt to keep the number of updates to models to a minimum to avoid disruption and changes to predictions, your model may also be updated at other points in the year to make changes or include additional variables that are necessary for making decisions and building your strategy. These model updates will be coordinated via your Client Success Director.