Playground XYZ is an advertising technology company that stands apart in the market by providing a campaign metric called Attention Time. In a nutshell, Attention Time aims to predict how long regular people will spend looking at a certain ad on the web. To predict and model Attention Time, the company runs a series of paid eye-tracking studies with thousands of users. The company needed a data inspection tool to understand what happens in the studies.
Focus of the project: Design a tool to inspect eye-tracking sessions and improve data quality
My role: Business requirements, technical feasibility, product design end-to-end
Background of the project:
The data is at the core of the business of Playground XYZ and there is a lot of data collected during eye-tracking studies. Unfortunately, not all the data collected in such studies is usable and there is no way to replay an individual eye-tracking study. To tackle this problem I worked on a project called "Playback" which aimed to understand more about eye-tracking sessions without violating the privacy of the study participants.
At the time none of the product managers were available and the tech team was working on a different project, so I stepped in to gather the business requirements and assess the technical feasibility.
Discovery work:
As the business requirements for the project were not clearly defined, I ran a series of workshops to get more information on the subject, align the teams on the pain points and identify the opportunities to improve current processes in the company. The workshops were the following:
-Feature canvas workshop
-Data quality workshop
-Affinity mapping workshop
Feature canvas workshop
Data quality workshop
Business requirements:
The outcome of the workshops was the business requirements document which was approved by the stakeholders.
The success criteria for the project were:
The success criteria for the project were:
-High-quality data
-More accurate models/products
-Saving resources
-Reduce back and forth to understand anomalies between teams
-Streamline our data collection and model training process
-Increased revenue $$
Success Metric:
-Define the causes of 100% of all high attention time impressions in eye-tracking studies
Design plan:
The design plan for the project consisted of the following steps:
- User interviews
- Competitor analysis
-Wireframes and high-fidelity designs
-Check-in with development team to assess technical feasibility
-Several rounds of usability testing
-Check-in with stakeholders
Wireframes
High-fidelity designs:
Home page with the list of eye-tracking studies
Eye-tracking study details page
Inspection page of the eye-tracking study done on mobile
Inspection page of the eye-tracking study done on desktop
Design components in Figma:
Outcome and learnings:
This project was one of the most challenging projects I have ever worked on because of the technical complexity and lack of resources in the company. I was happy that I was able to identify the main opportunities for data quality improvement and the data metrics important for the eye-tracking studies. This project was a big step forward in a data quality initiative in Playground XYZ.
