YC
Analyzing Billions of Transactions to Understand Consumer Behavior - Michael Babineau and Kevin Hale
@ Y Combinator
05/30/19
Takeaways
MB
Michael Babineau
05/30/19
@ Y Combinator
Our initial idea was to use credit card data to help investors make better investment decisions, and now we also work with companies, not just investors.
MB
Michael Babineau
05/30/19
@ Y Combinator
We aimed to build an analytics platform that allows investors to answer their own questions using credit card data.
MB
Michael Babineau
05/30/19
@ Y Combinator
I realized there was a huge opportunity in the investment space when I discovered that many hedge funds lack in-house technical support, relying instead on a handful of analysts.
MB
Michael Babineau
05/30/19
@ Y Combinator
Investors are always looking for an information edge, and many are turning to tools like Google Trends to find leading indicators of consumer sentiment.
MB
Michael Babineau
05/30/19
@ Y Combinator
Before we entered the scene, investors had to rely on surveys to understand consumer behavior, which was a slow and costly process.
MB
Michael Babineau
05/30/19
@ Y Combinator
We provide direct observations of millions of US consumers' purchases, allowing investors to find answers quickly without the need for surveys.
MB
Michael Babineau
05/30/19
@ Y Combinator
Our first customer was a VC who ended up investing in us, and being part of Y Combinator helped us get meetings with many interested VCs.
MB
Michael Babineau
05/30/19
@ Y Combinator
Credit card data is excellent for predicting consumer trends, but it has limitations when it comes to companies that don't sell directly to consumers.
MB
Michael Babineau
05/30/19
@ Y Combinator
Venture capitalists primarily use our service for due diligence, helping them understand market dynamics and compare companies in the same space.
MB
Michael Babineau
05/30/19
@ Y Combinator
We help VCs gain visibility into not just the company they're evaluating, but also its competitors, providing insights into market share and customer performance. Our core product empowers users to answer any questions they have about U.S. consumer spending without us selling research directly.
MB
Michael Babineau
05/30/19
@ Y Combinator
We don't proactively conduct research; instead, we respond to specific questions from clients on a project basis.
MB
Michael Babineau
05/30/19
@ Y Combinator
We've started collaborating with the press, getting quoted in major publications like the Wall Street Journal and Financial Times, which benefits both us and the reporters.
MB
Michael Babineau
05/30/19
@ Y Combinator
We continuously publish updates on topics like Uber versus Lyft to keep our audience informed and engaged.
MB
Michael Babineau
05/30/19
@ Y Combinator
Our editorial team, which includes data scientists and writers, focuses on identifying interesting stories and trends in consumer behavior.
MB
Michael Babineau
05/30/19
@ Y Combinator
When analyzing Stitch Fix, we found it does not cannibalize department store sales; instead, customers tend to spend more on clothes after becoming Stitch Fix users.
MB
Michael Babineau
05/30/19
@ Y Combinator
Peloton has surpassed SoulCycle in active memberships, indicating a significant shift in consumer preferences in the fitness industry.
MB
Michael Babineau
05/30/19
@ Y Combinator
Amazon is increasingly resembling a subscription business, with a growing reliance on Amazon Prime members for revenue.
MB
Michael Babineau
05/30/19
@ Y Combinator
Even former Amazon Prime subscribers continue to spend more on Amazon after their subscription lapses, indicating lasting behavioral changes.
MB
Michael Babineau
05/30/19
@ Y Combinator
We hire data scientists who possess strong quantitative backgrounds and can approach problems from first principles, rather than just relying on tools.
MB
Michael Babineau
05/30/19
@ Y Combinator
Our interview process includes giving candidates messy data sets to analyze, assessing their ability to identify and address underlying assumptions in the data.
MB
Michael Babineau
05/30/19
@ Y Combinator
A common mistake among applicants is assuming the data is perfect, which can lead to distorted findings if they don't anticipate potential issues.
MB
Michael Babineau
05/30/19
@ Y Combinator
Successful data scientists are those who question their assumptions and seek to understand the foundations of the data they work with.
MB
Michael Babineau
05/30/19
@ Y Combinator
Building a solid foundation is crucial; without it, any analysis or product built on top can easily collapse.
MB
Michael Babineau
05/30/19
@ Y Combinator
Credit card statements often contain vague transaction descriptions that can confuse consumers, as they may not clearly indicate the merchant's name.
MB
Michael Babineau
05/30/19
@ Y Combinator
The challenge with credit card data is that different companies represent the same merchant in various ways, leading to a cardinality problem where multiple descriptions exist for a single merchant.
MB
Michael Babineau
05/30/19
@ Y Combinator
Second Measure analyzes over 50 billion transactions, revealing that there are 1 billion unique transaction descriptions, but not nearly that many actual merchants in the U.S.
MB
Michael Babineau
05/30/19
@ Y Combinator
The human element in transaction data processing can lead to inconsistencies, as different people may describe the same merchant differently or make typos.
MB
Michael Babineau
05/30/19
@ Y Combinator
Second Measure employs machine-based approaches to clean and analyze messy credit card data, aiming to create a useful output from raw transactional data.
MB
Michael Babineau
05/30/19
@ Y Combinator
The company has built two products: a pipeline for processing raw transactional data and an analytics platform for competitive analysis.
MB
Michael Babineau
05/30/19
@ Y Combinator
Investors and companies use Second Measure's platform to understand company performance, which has led to a growing customer base beyond just investors.
MB
Michael Babineau
05/30/19
@ Y Combinator
Second Measure's approach focuses on helping clients ask better questions about their data, rather than just providing insights, to maintain a competitive edge.
MB
Michael Babineau
05/30/19
@ Y Combinator
The investment community benefits from Second Measure's data by being able to ask more creative questions, which helps them find their own edge in the market.
MB
Michael Babineau
05/30/19
@ Y Combinator
Second Measure's data is essential for corporate competitive analysis, as companies need to understand their competitors' performance to stay ahead.