Made for marketers, built by scientists
All Haus products are rooted in explainable econometrics and causal inference models - no black box deep learning models found here.
Accurate measurement is difficult
Poorly run tests can lead to inconclusive results, or worse, lead you to make sub-optimal decisions based on faulty data.
Skip the long, expensive learning curve with Haus's cutting edge methods for unparalleled scientific rigor.
Our team members were responsible for:
%202.svg)
IDFA/Cookieless ad targeting
Economics and ML
Growth and marketing infrastructure
Data pipeline and engineering
Causal inference modeling
Traditional prediction models require historical truth for model training, but there is often no accurate historical record of incrementality.
We extract marketing effects without this dataset by using causal inference methods (like ones that won the 2021 Nobel Prize in Economics).
Precision from cutting edge methods
We maximize accuracy and precision by using frontier methods in control selection, anomaly handling, and analysis.
For example, instead of using matched markets which are not precise test and control comparisons, we build synthetic controls where Seattle might be comparable to 60% San Francisco, 30% Denver, and 10% San Diego.
Controllable and interpretable
Our models and outputs are transparent and interpretable. The Haus platform includes on-demand power analyses, allowing you to select your ideal balance of precision and speed of insights.
.png)
Some of our science leaders
Joe Wyer
PhD: University of Oregon
Joe Wyer
PhD: University of Oregon
Before Haus, Joe dug into investment decision problems at Amazon, where he developed the central regional testing data product and led a team in evaluating hundreds of millions of dollars in marketing investments annually. While building that team, he also integrated frontier long-term value models with Amazon's multiple internal marketing measurement models attributing company value to billions of dollars in marketing investments. He then moved to Amazon Prime Video, where he initiated and led a decision science team that concentrated on pricing, promotions, and content acquisition strategies. Joe is our Head of Science at Haus building the models that power your experiments.
Formerly
Vanja Dukic
PhD: Brown University
Vanja Dukic
PhD: Brown University
Professor at University of Colorado Boulder
Vanja holds a PhD in Applied Mathematics (computational statistics) from Brown University. After Brown, she spent over a decade at the University of Chicago, first as a postdoc and visiting professor in the Department of Statistics, and then tenured faculty in Biostatistics. Vanja is currently a Professor of Applied Mathematics and a courtesy Professor of Economics at the University of Colorado at Boulder. Her main research interests focus on Bayesian methodology, inference and modeling, decision theory, and computational statistics.
Formerly
Omri Perez
PhD: Tel Aviv University
Omri Perez
PhD: Tel Aviv University
Before joining Haus, Omri led the science for multi-touch marketing attribution (MTA) at Amazon responsible for reporting and optimizing Billions of dollars of annual spend. Following his PhD studies, he has worked and contributed to various domains in the industry, working in Computer Vision, Predictive Medical Analytics, Fraud prevention and Credit at Paypal, and Personalization and Marketing Measurement at Amazon.
Formerly
Philip Erickson
PhD: University of Iowa
Philip Erickson
PhD: University of Iowa
Phil came to Haus from Amazon where he was the lead economist for scaling causal modeling of customer preferences for Amazon Devices. In this role, he developed the science and software to measure customer willingness-to-pay for every major Amazon-branded consumer electronics product. He also owned the econometric models that measured the ongoing causal impact that Device/Prime joint marketing investment had on company finances. Phil has researched and taught econometrics and data science in both academia and industry, most recently as an Affiliate Assistant Professor of Economics at University of Washington.
Formerly
Graton Gathright
PhD: UC San Diego
Graton Gathright
PhD: UC San Diego
Graton specializes in causal measurement of marketing. Before joining Haus, Graton helped Amazon to incorporate incrementality measurement into their attribution systems, leveraging both causal modeling and calibration to RCT marketing experiments.
Formerly
Marie-Renée Arend
Masters of Science University of Washington
Marie-Renée Arend
Masters of Science University of Washington
Prior to joining Haus as a data scientist, Marie worked as a senior technical program manager at Amazon where she led core research and modeling initiatives for the Alexa speech science organization. In addition to driving continuous improvements to Alexa’s speech recognition capabilities in existing markets, she also launched new Alexa languages such as Brazilian Portuguese and Canadian French. She later joined Alexa Lab, the internal A/B testing platform for Alexa product teams, to support self-service experimentation for device feature launches.
Formerly
