Friday, December 3, 2021
No menu items!

Post Title

Wharton Customer Analytics

Analytics Accelerator Case Study

Customers of A Feather Engage Together:
Using K-Means Clustering to Maximize Retention

Objective

Fox Entertainment is an American television production company owned by Fox Corporation. They oversee Fox Alternative Entertainment, Bento Box Entertainment, and Tubi, Fox Entertainment’s video-on-demand service. Fox Entertainment sought to establish an analytical framework for expanding their knowledge of data analytics throughout the organization. They aimed to become less dependent on third-party analytics vendors, develop processes for building in-house analytics capabilities, and to deepen the company’s platform analytics to inform its digital media business.

In pursuit of these goals, Fox Entertainment partnered with Wharton Customer Analytics. They worked with a team of faculty-led students to create a new marketing mix model and social media approach for optimizing strategies to increase viewership of television premieres.

Approach

FOX Entertainment provided a series of data sets consisting of campaigns (television shows) with different variables, campaign-level promo data, and campaign-level social media data for fifty days leading up to the television premiere for each campaign.

The team started by cleansing the data, and then performing exploratory data analysis, unsupervised learning, and supervised learning/modeling. They sought a model to optimize advertising across each channel in order to generate the optimum number of viewers for the amount of money Fox spent. This includes the Fox Entertainment Channel itself, all the synergy platforms, paid advertising off-network, and its non-linear digital media platforms. The team analyzed campaign, channel, and awareness effects to determine an appropriate marketing mix.

Regarding social media branding, the team’s initial exploratory data analysis found interesting insights, but did not yield actionable conclusions. Rather than beginning with the data, the team pivoted to begin trying to answer business questions that might be valuable to Fox Entertainment. This uncovered a number of actionable insights.

Recommendations

Using the described methodologies, the team was able to arrive at a number of valuable conclusions and recommendations for Fox Entertainment, including the following:

NFL premieres are significant: if Fox Entertainment wishes to reduce advertising spend, they can ensure strong viewership by placing premieres immediately after NFL games – specifically dramas, a genre which typically sees conversion rates decline without an NFL game preceding it.

Recommendations

Monitor homework completion rate as a potential indicator of retention.
Assign inactive customers a behavior health specialist.
Adapt specific product offerings to align more closely with engagement KPIs.
Convert remaining SMS notifications to push or email notifications.

“It was great to have a fresh set of eyes in thinking about user engagement and retention. In a short amount of time, the team was able to effectively connect the dots between patterns in the data with actionable product recommendations we could test.”

Bill Lynch
Director of Data Science at NeuroFlow

Impact

In the months following the Analytics Accelerator, NeuroFlow adopted the team’s recommendations and utilized the findings to:

Inform A/B Testing. NeuroFlow implemented a new testing infrastructure and has begun testing across different channels and themes.
Enhance features within their platforms. NeuroFlow launched a feature that drives customers to work with a behavior health specialist, leading to increased engagement and business growth.
Further qualitative and quantitative research and discovery. NeuroFlow integrated the internal data team into the product development business process to strengthen its core offerings.

About the Analytics Accelerator

Every fall and spring semester, Wharton Customer Analytics hosts the Analytics Accelerator, an experiential learning program that pairs students with a company to solve a real-world business problem using the company’s actual datasets and the latest techniques including machine learning and AI.

Learn More

The post appeared first on Wharton Customer Analytics.

Read MoreWharton Customer Analytics

RELATED ARTICLES

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Most Popular

Recent Comments