Data Speaker Series: Shaun Lindsay on the Data Science of Marketing

shaun-lindsay-featured-imageFor the October Data Speaker Series, we welcomed Shaun Lindsay. Shaun is the Interim Head of Marketing at Honor, a home care company based in San Francisco, CA. But, before stepping into the role, he was a data scientist. This makes him uniquely suited to share some fun and interesting marketing challenges and common pitfalls marketing teams often encounter when analyzing data.

Shaun talked about his long and winding journey from software engineer to data scientist, to his current marketing position. Starting out as a backend engineer, Shaun’s first foray into machine learning came when he detected a bot net interfering with Meebo’s ad network. Faced with eliminating said bot net, Shaun started on his path to becoming a data scientist. After Meebo, he continued fraud detection and prevention at Google, before leaving to start a home care company, Honor. At first, as a developer and data scientist, Shaun worked on fun and fascinating logistics problems, before his SEO knowledge helped him become the head of marketing.

In this clip, he discusses a bit of marketing’s history, and introduces data challenges in the field.


Lifetime Value

Shaun spoke about customer lifetime value (LTV) and highlighted a subtle, yet commonly overlooked characteristic of the metric. Often, marketers will try to model LTV by using the average of existing customer lifetimes. Here, he discusses the bias this method introduces and explains the problem of censored data, giving some advice on how to appropriately model LTV.


Net Promoter Scores

Next, Shaun talked about net promoter scores (NPS). These scores are based on survey results, and thus come with their own set of challenges, which he discusses in the following clip. He talks about the classic statistics problem of confidence interval estimation, and reminds us that “data scientists are statisticians with better marketing.”


Experimental Design

Finally, Shaun discusses the challenges of applying the traditional experimental set up within the advertising space, by contrasting it with the more straightforward A/B testing scenario.

Interested in catching up on our Data Speaker series? Check out Charles Earl on Discriminatory Artificial Intelligence and Thorsten Dietzsch on Building Data Products at Zalando.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s