LinkedIn — Global Search Feedback III (Mobile)

Hello, it's me again!

This is a rebound to one of the shots I posted earlier this week for the "Explicit Search Feedback" project I worked on while contracting at LinkedIn. The goal was to design an effective explicit user feedback framework where feedback signals could be employed to predict the quality of the LinkedIn search experience (e.g. searcher satisfaction), as well as other useful information such as result relevance and searcher preferences.

Challenges of using user feedback

Using user feedback also poses some challenges for improving search results. These include dealing with noise and bias, balancing trade-offs and conflicts, scaling and updating the data, as well as integrating and interpreting the data. Filtering out irrelevant, inaccurate, or misleading feedback or behavior is essential to ensure diversity, fairness, and privacy of the users and the results. Additionally, combining and analyzing the different types and sources of feedback and behavior data is necessary to handle the large volume, variety, and velocity of the data.

Feel free to hit 'L' if this design resonates with you. Happy Friday!