Roku Ad Management Platform Analytics Project

Analysis of Ad Management Platform Usage Data To Derive Strategic Insights.

Emilie Pourchet

Background

During the fall semester, I was paired up with two classmates and given an opportunity to work with Roku TV.

The goal of the project was for us to look at how data analytics were set up in their data analytics software Pendo, and find a way to draw insights from the data on how the customer is interacting with their platform/

Initial Work

The first thing we did was to try to get an understanding of the product. We did this by getting access to OneView, the data management platform, and trying to understand the steps required to create ad campaigns. We also spent a lot of time researching AdTech: we needed to understand the conventional ad buying process, and how use of platforms like Roku Oneview differs from those. We spent some time speaking with various groups at Roku to get an understanding of their different user personas and how they each use and benefit from the platform.

Roku advertisement for Oneview taken from https://advertising.roku.com/resources/blog/insights-analysis/oneview-runs-tv-streaming-campaigns-like-social

Getting access to Pendo Analytics to be able to see data on the OneView platform usage took a little more time than expected, so we started by performing an analysis of how feedback on the platform is provided and implemented by the product team. We were able to identify areas for improvement and propose solutions.

Final Deliverable

When we finally got access to Pendo, we were able to create funnels, flows and dashboards to get a better understanding of how users were interacting with the OneView platform. We quickly understood that the Pendo dashboards had some limitations, and decided to utilize the Pendo API to create dashboards of the data in Pendo. This also allowed us to pull in some excel data that had been given to us separately but could provide additional information to the Pendo data.

We created dashboards to show activity based on accounts, sectors. We looked at campaigns, spendings, unique accounts. We created a dashboard to show trends for number of events and time spent, as well as metrics per product area, page and feature. This allowed us to uncover 6 major insights about how Roku customers use the platform that Roku can take to their product team to address in future product decisions.

Sample section of Tableau Dashboard (one of multiple sections provided to Roku) with all confidential information hidden.

Project Takeaways

This project taught us a lot about adtech, how the industry works, and how companies like Roku are implementing new features to attract growing customer segments.

I learned a lot about data analytics, and especially about the capabilities and shortcomings of analysics platforms like Pendo. This helped give me a better understanding of a similar platform I use at my part-time work.

Finally, this project taught us to be creative, and pivot when things aren't working out. We were waiting for data access for a long time, but we didn't waste time and utilized other resources to provide value to Roku through analysis of the feedback they receive about their platform. This allowed us to get them a second deliverable, which was not initially within our scope.