Media Mix Modeling - What, Why, and How?

And a comparison vs. attribution modeling

How should I allocate spend across marketing channels like TV and paid search? If you’re managing a company’s marketing budget, this may be the most essential question to maximize profit or revenue given limited spend. There’s three typical ways to answer it scientifically: Top down: take all of your data... [Read More]
Tags: machine learning, attribution, optimization

What Worked at Wayfair

And how to build a money printing machine

In four years as a search marketer, I built - in collaboration with my team of analysts and our partners in data science, ad tech, and storefront - $87 million of annualized, incremental revenue for Wayfair. Molto bene! Here are three frameworks that I used to pick bets and execute... [Read More]
Tags: management, analytics, professional development

Painting With DALL·E

DALL-E is spectacular. I’ve been using it to make everything from abstract inspiration for acrylic paintings to my favorite Nintendo characters reimagined as renaissance era royalty. But creating great AI art takes more than a button press: you also need a well-engineered prompt. In this article, I’ll give my advice... [Read More]
Tags: DALL-E, AI, art, prompt engineering

Banana Data Podcast

On the Banana Data Podcast, I discussed how to get the most value from your time and communicate based on your audience’s technical ability. Listen to it here!
Tags: wayfair, management, data science, communication

Teaching a Machine to Read Pitchfork

This week I built a classifier that determines whether Pitchfork’s album reviews qualify for “Best New Music” based on the review text. I rely pretty heavily on Pitchfork to find new music and better understand what I’m listening to, and analyzing the results of this classifier helped me investigate a... [Read More]
Tags: web scraping, machine learning, classifiers, data visualization