Thanks for being here!
Announcements:
If you care about RBLX (the stock), you should be checking out our Gift Card Data. This is the highest conversion rate I’ve seen with data product (# of firms that review —> # of firms that enter contract).
We have a NEW daily data product that offers end-of-day account balances across various types of accounts (savings, checking, etc). Small regional & community banks. No PII. Very granular.
Theme that emerged in this week’s email is … AI needs data. And lots of it.
QUOTES
“All this data can be used not only to create new factors but also to enhance existing factors.” - David Blitz, Chief Researcher, Robeco
“The alternative data market is making huge progress and the insights and trends it can show are becoming more important for investment professionals.” - Neil Chapman, CEO, Exabel
News Articles
Podcasts
Cool Charts
Final Thoughts (Current Reading)
#1 – Jason DeRise from The Data Score published The Bull Case for Alternative Data. April 2023.
My Take: It should come as no surprise I am bullish all things data, including “alternative data”. The author states we are in the “trough of disillusionment” (Gartner Hype Cycle … see below for chart). I largely agree. There are too many data providers and not enough long-term winning use cases. A combination of improving data & improving tools to engage with the data will lead the industry towards fulfilling everyone’s great expectations (Gartner’s “slope of enlightenment”). The article offers three key factors to long term success:
Realign data and teams with critical client outcomes.
Make it easy to value a data practice.
Build network effects.
The product-market FIT discussed under point #2 above is tough to identify, tough to build, tough to price, tough to commercialize. But once in place, the data business that is delivering on product-market FIT is a very good business.
“The data businesses that work backward from critical client outcomes, align their product with the value created by decision makers using the data and set up their businesses to benefit from network effects have the best chance of sustainable long term growth and reaching the plateau of productivity.” - Jason DeRise
On my entrepreneurial journey, one of the many things I wish I would have done earlier is talk directly to customers. Given that I spent my pre-data career on the traditional sell-side, I thought I knew what my customer wanted … I should have asked rather than assume.
#2 – David Blitz, Chief Research at Robeco, published Why the best is yet to come for factor investors. March 2023.
My Take: The rise of alternative data driving “next generation” factor investing. Firms are trying to extend the useful life of new factors. Some fast signals do not survive after accounting for transaction costs. Alternative data can be a method to uncover more durable signals that will drive value to those that can identify & execute on the new factors or use these new data point to enhance existing factor strategies.
#3 – Joshua Benton published in Nieman Lab What if ChatGPT was trained on decades of financial news and data? BloombergGPT aims to be a domain-specific AI for business news. April 2023.
My Take: Please see related 65-page white paper. Bloomberg has great data on which to train models. I’d imagine there is some benefit to being a more focused domain than just everything in the world (maybe, maybe not). 52% of the 700B data points used to train Bloomberg GPT are proprietary or cleaned financial data. I am paying attention to “domain specific LLMs”. Seems like this would be the cleanest path to professional (i.e. revenue-generating) use cases.
At a minimum BloombergGPT can help novice Bloomberg Terminal users enter complex query language into the terminal (similar to how I use ChatGPT to help me with SQL code). I guess we need to see how Bloomberg customers will ultimately engage with this resource but given access to great data, Bloomberg is in a great spot to take the lead in this area.
BONUS: Snowflake Invests in Cybersyn, Bringing Unique Data Products to Snowflake Marketplace. April 2023. An investment of this magnitude signals the growing importance of data & the importance of the right tools engage with data (both internal & external). I use Snowflake daily and have found it allows me to focus on the important part: asking the right question of the data. Post from Founder Alex Izydorczyk.
BONUS 2: Tim Human published in IR Magazine Consumer spending data to give ‘outsized informational edge’, finds investor survey. April 2023.
“The alternative data market is making huge progress and the insights and trends it can show are becoming more important for investment professionals.” - Neil Chapman, CEO, Exabel
BONUS 3: John Author published in Bloomberg: How Moneyball Investing Ran Into a Data Squeeze Play. March 2023.
“The main takeaway is that efficacy is declining over the long term. Nobody is saying these concepts aren’t relevant. They all still matter. But as soon as a particular definition becomes prevalent and widely used, there’s more attention to it and that factor begins to atrophy.” - Jay Rajamony, head of alternatives at Man Numeric
BONUS 4: Stanford published The AI Index Report for 2023: Measuring trends in Artificial Intelligence. 386-pages (wow!). See a couple of the interesting charts below.
#1 – David Johnson of Domo hosting the podcast Domo Data Therapy. This is the first of three published episodes. Inspiring Curiosity with Ben Schein, SVP of Data Curiosity at Domo. January 2023.
My Take: Good podcast to put in your “data podcast” rotation. Good format, short, focused. Interesting to hear from Ben Schein about his time at Target and how he helped build the data culture at such a big place. Lots of lessons from the front lines. Some keys:
Minimize dependencies & increase agility
Small teams that can build trust
Small data product teams
Highlights (22-minute run time):
Minute 01:00 – interview starts and Ben’s background
Minute 03:00 – now VP of Data Curiosity at Domo coming from his role at target
Minute 04:15 – seminal moment when Target combined into central center of excellence; allowed them to find use cases that weren’t being addressed
Minute 06:45 – changing data culture at target and how Domo helped
Minute 08:15 – biggest drivers from people & process standpoint on which data leader can focus.
Minute 12:45 – Ben’s own frustration with his data
Minute 14:30 – moment of innovation from Ben’s career; the data brought together at target.
Minute 18:00 – Ben’s Advice for Ways to move in the direction of data maturity
Minute 20:45 – Ben’s question for the next guest: How do you spark the fire of data curiosity?
Source: BloombergGPT: A Large Language Model for Finance (page 6).
I thought his was an interesting list of training data used to train BloombergGPT.
Source 2: Stanford’s 2023 AI Index Report. There are hundred of charts in this paper. These are just a couple examples that I found interesting (page 200). I am trying to understand the “real world” use cases.
Source 3: The Data Score. April 2023.
I typically have 3-4 books that I am reading / flipping through, at any time.
Taking suggestions.
This is what I am reading this week:
Hump Pilot, Nedda Thomas. My Grandfather flew The Hump in WWII. Talk about purpose (see article below). As Chris Arnade has put it, "People need to feel a valued member of something larger than themselves".
Persuasive Copywriting, Andy Maslen. Trying to improve my writing.
The Almanack of Naval Ravikant, Eric Jorgenson. This is a book I have around and pick up often.
When Money Dies, Adam Ferguson. I had read this cover-to-cover many years ago. Revisited & flipped trough a couple chapters when I came across Balaji’s Bet.
Get Serious: About Purpose, Katherine Boyle. I find it interesting to pay attention to these long-term trends how see them play out through the difference I see between my generation (I’m late 40’s) and my teen kids.
Some good changes and some bad. But distinctly different.
From article:
Have a great week.