Skip to content


An example-based customization sharing platform for educators

Co-authors: Kavana Ramesh, Joanna McGrenere, Parmit K. Chilana

A screenshot of Customizer listing two home page customizations and an external Wikipedia app shared by a colleague
Customizer's discovery interface overlaid on top of a Canvas page, listing several customizations suggested by a colleague.


Educators are increasingly reliant on a range of software tools to support their teaching practices and engage their students. However, learning to use these tools and integrating them into a classroom environment can be challenging and time-consuming, especially for those without a technical background. With feature-rich classroom software like learning management systems (LMS), educators face a particularly steep learning curve, as they often need to customize aspects of the look-and-feel or functionality of these tools to match their own teaching practices and circumstances.

Customizer is designed to simplify this learning process, letting educators leverage the collective wisdom of their colleagues to discover useful, vetted customizations and seamlessly apply them to their own software setups.

The Customizer platform

Built atop the Canvas LMS, Customizer incorporates a context-aware recommender system for serendipitous discovery of example customizations used by colleagues. It also includes a Try It mode, where educators can interactively experiment with shared customizations in a safe testing environment that mimics the content of their real courses (avoiding unintended student-facing consequences).

Educators can import promising customizations directly into a Canvas course without the need to scour through settings menus or external help forums. Finally, Customizer offers robust tools for authoring and sharing example customizations with others, alongside rationale such as what pedagogical goals they support or how they improve the interface.

Designing Customizer

Customizer was built through a human-centered design process that began with interviewing K-12 teachers and university instructors, to build an understanding of the challenges they face with customizing their classroom software and to explore possible design directions. These interviews made it clear that most educators preferred to pick up software knowledge via over-the-shoulder learning — i.e., seeing a colleague demonstrate how they customized their software and trying to borrow their strategies.

Through a few rounds of prototyping and usability testing, I designed several tools to support this model of learning directly within a software application and implemented them as features of Customizer, which users install via a lightweight browser extension.

A screenshot of Customizer's sidebar listing several customizations relevant to the Canvas Discussions page
Customizer adds a collapsible sidebar to Canvas pages, where educators can quickly see relevant or widely-used customizations related to the current page (here, the Discussions page).
A screenshot of Customizer's Try It mode, showing a copy of the user's Canvas course with customized features highlighted
Customizer's Try It feature gives users a risk-free, fully-interactive testbed to explore how one or more customizations would affect the UI or functionality of their courses. This mode highlights any changes resulting from the selected customizations, and adds help badges that can show additional information about each change.

Evaluating Customizer in practice

I planned and conducted a multi-week field deployment of Customizer to gauge how educators would use and perceive its features in real-world contexts. As it turned out, participants found all sorts of creative ways to integrate Customizer into their day-to-day workflows, including for rapidly experimenting with new ideas, gaining confidence to explore LMS features, streamlining their knowledge-sharing routines, and learning at their own pace without bugging others.

Our conference paper about Customizer's design won an award at DIS 2020, and was followed by a further paper at CHI 2022 describing the field deployment and results. Looking to the future, the design ideas behind Customizer also have the potential to generalize well to other types of feature-rich software, such as image/video editing software, website builders, and highly-customizable IDEs.