Alternative Data Weekly #225
February 14, 2025
Happy Valentine’s Day!
Special thanks to our sponsor Big Data Federation!
Boost Your Strategies with AI-Powered Fundamental Forecasts.
The theme that emerged in this week’s email is … AI is changing things.
QUOTES
“…Asking one question at every stage of your generative AI journey — ‘what is the business value here’ — can help keep your organization on track.” – Cassie Kozyrkov
News Articles
Podcasts
Cool Charts
Final Thoughts (AI Valentine)
#1 – Exabel published 2025 Alternative Data Research Report. January 2025.
My Take: Exabel surveyed 130 active portfolio managers for this report. See the “Charts” section for some interesting charts I’ve pulled from the report. The good news is demand remains strong. But it is still hard to ingest, organize, and generally make alt data work on your own, with 87% of respondents saying "It is more economical to use third party systems than invest in our own in house".
There is value in Alt Data, as demonstrated by the 85% of respondents saying budgets would increase.
I am curious how AI will smooth over some of the difficult areas. The two most frequently cited challenges, 1-combining data from different sources into a singular prediction (79%), and 2-processing raw data into a useable format (48%) … are, in theory, something for which AI would help.
#2 – Duncan Gilchrist & Jeremy Hermann published The Paradox of Optimism in Data Science. February 2025.
My Take: I like the idea presented about creating a “ladder” of projects: “Quick Wins”, “Strategic Projects”, “Transformational Bets”. You need occasional wins, even small wins to keep the team optimistic and to make sure the C-suite remains supportive. The more fun projects can encompass transformational ideas that might improve how everyone works day-to-day … but these can be hard and take lots of time. By the time you get around to delivering, the whole team will have been fired for lack of demonstrable progress.
#3 – Marc Rubinstein published AI Eats Equity Research. February 2025.
My Take: This one hits close to home as I was on the sell-side for a long time. As a Senior Research Analyst, at one time, you could value gathering and organizing all the information like initiation reports & earnings recaps (AI can do this better). At one time, you could add value by having a “good relationship” with management and hearing things first (technically illegal). You could get paid by proving that you are consistent and have a good thought process … help your HF clients “think about” stocks, knowing the right questions to ask (not sure AI can do this as well as a 15-year veteran analyst, but maybe with good prompting?). A Senior Analyst could get paid by having his team do grunt work no Harvard grad HF guy wants to do, like surveys or channel checks (AI can’t really do this, can it?).
At the big banks, the sell-side divisions are loss-leaders, and essentially marketing arms for the i-bankers to go get M&A deals. The death of that business has been incorrectly called a thousand times.
At the regional brokers, you need to do “real research” to justify payment. This will change, but there will still be a place for this type of work. There has to be a place for a good “big brain” researcher to add value for clients. It has always been changing and always will be changing.
That said, when things change, they change quickly. Just ask the human market-makers.
What else I am reading:
Abraham Thomas published Making Markets in Time. January 2025.
Cassie Kozyrkov published Endless Right Answers: Explaining the Generative AI Value Gap. February 2025.
Ethan Mollick published The End of Search, The Beginning of Research. February 2025.
Benn Stancil published For the brand, I guess. February 2025.
Jason Derise published Which food and beverage companies still have pricing power after half a decade of high inflation?. February 2025.
FactSet Acquires LiquidityBook For $246.5 Mln Cash. February 2025.
Claire Calmejane published The AI revolution is here in financial services. February 2025.
Sven Balnojan published A Summary of What I Wrote In 2024. February 2025.
Annabel Smith published Exchanges hit back at ‘inaccurate’ and ‘misleading’ accusations around market data costs. February 2025.
Source: Wall Street Journal’s Joanna Stern’s mid-January Interview with Ron Lefferts of LSEG & Jayee Koffey of BNY. Generative AI in Financial Services: From Innovation to Industry Game-Changer. January 2025.
My Take: Among big enterprises, we are seeing the move from experimentation to a focus on projects with real ROI (Lefferts shares a customer service example). Moving from POV to execution as we see increased usage and feedback from the real world. LSEG partnership with MSFT is starting to deliver AI products now. 2025 is a year of seeing how we can get to adoption at scale.
From a data perspective, data trust is essential. Tremendous focus on data quality & seeing how difficult this can be to do well.
Push & Pull workflows (24:30).
Highlights (32-minute run time)
Minute 02:00 – Ron Lefferts LSEG intro
Minute 04:00 – Jayee Koffey BNY intro
Minute 07:30 – Ron: what has LSEG seen that is indicative of change
Minute 11:00 – Jayee: how AI is being implemented at BNY
Minute 16:00 – Ron: quality of data is what determines the success and adoption
Minute 18:00 – thinking about ROI; science experiments vs real world
Minute 20:00 – Jayee: BNY focused on specific use cases and ROI; putting metrics around the effort
Minute 22:00 – The importance of personalization
Minute 24:30 – Ron: Push & Pull workflows
Minute 26:45 – Responsibility: 7 principles for LSEG
Minute 28:00 – question from the audience about AI talent
Source 1: Exabel published 2025 Alternative Data Research Report. January 2025.
The industry as a whole continues to grow, driven by increasing demand from investment managers across sectors. However, different data categories are experiencing varying levels of growth.
Source 2: Ben Lorica published 10 Things to Know About the State of AI Agents. February 2025. (Letta alternative source doc)
I am a sucker for a good Top Ten list.
Which AI Chat Interface are you in love with?
PROMPT: Please tell me the top ten most interesting statistics about Valentine’s Day.
Below find comparisons of answers from ChatGPT, Mistral, and Perplexity.
CHATGPT
Here are ten fascinating Valentine's Day statistics:
1. Big Spending Holiday – In 2024, Americans were expected to spend around $26 billion on Valentine's Day gifts, making it one of the biggest consumer holidays.
2. Average Spending Per Person – The average U.S. consumer spends approximately $192 on Valentine's Day gifts, including for partners, family, friends, and even pets.
3. Pet Love – Around 27% of people buy Valentine’s gifts for their pets, spending nearly $1.7 billion on furry friends.
4. Most Popular Gifts – The top three Valentine’s Day gifts are candy (57%), greeting cards (40%), and flowers (37%).
5. Roses Rule – Over 250 million roses are grown and sold specifically for Valentine's Day each year.
6. Jewelry Sales Boom – About $6 billion is spent on jewelry for Valentine's Day, making it the most expensive gift category.
7. Singles Celebrate Too – More than 30% of people who celebrate Valentine's Day buy gifts for themselves or their friends instead of a romantic partner.
8. Marriage Proposals – An estimated 6 million couples get engaged on Valentine’s Day, making it one of the most popular days for proposals.
9. Online Dating Surge – In the lead-up to Valentine's Day, dating apps see a 20-30% increase in activity as singles look for love.
10. Chocolate Craze – More than 36 million heart-shaped boxes of chocolate are sold each year, with more than 58 million pounds of chocolate purchased for the holiday.
Would you like any of these expanded with more details?
MISTRAL
Here are ten interesting statistics about Valentine's Day:
1. **Popular Day for Engagements**: Valentine's Day is known as the most popular day to get engaged, with 40% of men agreeing that it is an ideal day for proposals.
2. **Global Celebration**: Approximately 59% of consumers worldwide celebrate Valentine's Day, making it a widely observed holiday across the globe.
3. **Greeting Cards**: About 145 million Valentine's Day cards are exchanged annually in the U.S., making it the second-most popular greeting card occasion after Christmas.
4. **Spending Habits**: The average amount spent on Valentine's Day gifts is $130.97, with men typically spending more than women.
5. **Flowers**: Roses are the most popular Valentine's Day flower, with over 250 million roses produced exclusively for the holiday each year.
6. **Engagement Statistics**: Around 6 million couples get engaged on Valentine's Day, highlighting its significance as a romantic occasion.
7. **Jewelry Spending**: Jewelry is the top category for Valentine's Day spending, with approximately $5.5 billion spent in 2023.
8. **Belief in True Love**: Between 52% and 56% of Americans believe in true love, which is often celebrated on Valentine's Day.
9. **Average Spending**: The average spending on Valentine's Day is projected to be $192, up from $175 in 2022.
10. **Gift Preferences**: Most consumers spend their money on candy (57%), cards (40%), and flowers (37%) for Valentine's Day.
PERPLEXITY
Here are ten interesting statistics regarding Valentine's Day:
1. Approximately 59% of consumers around the world celebrate Valentine’s Day[1].
2. Valentine’s Day spending hit a record high of $14.2 billion on significant others in 2024[1][5]. Total spending reached $25.8 billion[5].
3. For Valentine's Day 2025, consumer spending is expected to reach a record of $27.5 billion[2][7][8]. The average person is expected to spend $188.81, up from $185.81 in 2024[2].
4. Roughly 6 million people get engaged on Valentine's Day[1]. It is known as the most popular day to get engaged, with 40% of men agreeing[1].
5. Over 8 billion candy hearts are produced every year for Valentine's Day[3][6].
6. Each year, around 250 million roses are grown specifically for Valentine’s Day[1][3][6].
7. A total of 145 million Valentine’s Day cards are exchanged every February 14[3][4][6]. Valentine's Day is the second biggest holiday for exchanging greeting cards, after Christmas[3][4].
8. The most popular gift is candy (56%), followed by greeting cards (40%) and flowers (39%)[2][9].
9. Teachers receive the most Valentine’s Day cards annually[4][6].
10. In 2024, 29% of Valentine's Day spending wasn't for a significant other; gifts were also purchased for friends, coworkers, and pets[1].
Citations:
[1] https://www.driveresearch.com/market-research-company-blog/valentines-day-statistics/
[2] https://nrf.com/media-center/press-releases/nrf-survey-valentine-s-day-spending-reaches-record-27-5-billion
[3] https://localpulse.com/2024/02/10-fun-facts-about-valentines-day/
[4] https://www.goodhousekeeping.com/holidays/valentines-day-ideas/a26863/valentines-day-facts/
[5] https://www.rioroses.com/valentines-day-spending-trends-what-the-numbers-mean-for-your-floral-business/
[6] https://www.weareteachers.com/valentines-day-facts/
[7] https://nrf.com/research-insights/holiday-data-and-trends/valentines-day
[8] https://www.statista.com/statistics/285028/us-valentine-s-day-sales/
[9] https://moosend.com/blog/valentines-statistics/
---
Answer from Perplexity: https://www.perplexity.ai/search/please-tell-me-the-top-ten-mos-EODNBU0hTJWgAHtLyDDqSA?utm_source=copy_output