Alternative Data Weekly #238
May 16, 2025
Thanks for being here!
Special thanks to our sponsor Babbl Labs.
New: Extracting Alpha and Insights from YouTube Financial Video Content
The theme that emerged in this week’s email is … let’s simplify data access.
QUOTES
“There is a whole class of information that is waiting to be uncovered” – Scott Hamilton
News Articles
Podcasts
Cool Charts
Final Thoughts (World of DaaS)
#1 – Vin Vashishta published How To Accelerate AI’s Time To Value. May 2025.
My Take: I’ve seen a shift in client attitudes about AI from experimentation & learning … to ROI. The author suggests practitioners “start with a data strategy, focusing on contextual data, which enables the delivery of value incrementally, starting with simple data products and descriptive models trained on high-quality data.”
I have seen solution providers that have a very narrow focus gain traction as they are able to help clients deliver an “AI Win”. It is worth taking a step back to make sure you are future-proofing your efforts to avoid later backtracking.
#2 – Monda Acquires Amplify Data to Power Customizable Data Sharing for AI. May 2025.
My Take: The theme this week is simplifying data access (see podcast notes below too). This is easier said than done, but acquisitions like this indicate steps are being taken in the direction of, not just having the data, but making it easy to deliver, and help nontechnical business users engage & ask questions of the data.
#3 – Cassie Kozyrkov published Please Cheat at Your Job. May 2025.
My Take: I am stealing this as instructed (if you read to the end). The punch line is AI is great, and you should use it for your job, but you still have to be the adult-in-the-room; you must validate the output. It’s interesting because I am seeing my college-aged kids indicate that “AI is bad”. After all, it is viewed as a method of cheating, where the real world will demand they use AI, but be smart about it.
BONUS: The best of Uber’s data labeling, testing, and localization for your business. May 2025.
My Take: Interesting to see Uber get into this business called Uber Solutions. Perhaps validation the market for these types of services (labeling, testing, etc) is big enough to move the needle for a company as big as Uber.
What else I am reading:
Joseph DiTomaso published Private Credit and PE Are Sitting on a Goldmine of Data—But They’re Not Using It. April 2025.
Antonello Zanini published Scraping Historical Data From the Wayback Machine. May 2025.
Peter Baumann published AI Books - Review. May 2025.
Stan Altschuller published Which ChatGPT Enterprise Model is Best for the Job?. May 2025.
Lyden Foust published a 5-step masterclass on customer profiling. May 2025.
Nupur Anand published JPMorgan says AI helped boost sales, add clients in market turmoil. May 2025.
Storm King Analytics published From Noise to Insight: The DoD’s Next Battle with Public Information. May 2025.
The Fintech Times published American Fintech Council Urges Senate to Pass Credit Access Bill to Expand Use of Alternative Data. May 2025.
Duncan Gilchrest & Jeremy Hermann published The Rise of Data Slop. May 2025.
Source: Dash Media’s Michael Mo published Live Data CEO Scott Hamilton · "The Birth of the Data Grid". May 2025. This podcast follows a post from Scott, The Birth of the Data Grid, that we highlighted last week.
My Take: There is a whole class of information that is waiting to be uncovered. The key is accessibility. Knowledge becomes a commodity & everybody is going to “know” the same thing. This will unlock new insights. “At any time, any of us will have access to a common truth”.
The “data grid” is the concept that data will be metered & priced similar to electricity, simplifying data access, allowing everyone to become a “knowledge explorer”.
It will be less about having access and more about asking the right questions of ubiquitous data. Intuition and hypothesis testing will be key skills, & humans are better at this than computers.
Millions of interesting stories just waiting to be told … when people can more easily interrogate the data & seamlessly join multiple data sets.
Of interest: Live Data is seeing the tide shift, with people coming back from Austin and Oregon (the covid locales) … back to major tech centers.
Highlights (31-minute run time):
Minute 01:20 – what does Live Data do?
Minute 05:00 – knowledge becoming a commodity is a net positive for humanity
Minute 06:30 – consulting firms shrinking (no more exclusive access to knowledge)
Minute 07:00 – how will the data grid come into existence?
Minute 12:20 – “not sure who will be first, but everyone will want to go second”
Minute 13:00 – “Knowledge explorer”
Minute 17:00 – will big AI models will be commoditized?
Minute 20:30 – how will Live Data evolve given all this change
Minute 24:00 – what industries are seeing inflows of human capital?
Minute 28:00 – what kind of industries will be most likely to use this type of data grid application?
Source: Scott Brinker and Frans Riemersma published State of Martech 2025. May 2025. 137 Pages! Some cool charts in here.
BONUS: Joseph DiTomaso published Private Credit and PE Are Sitting on a Goldmine of Data—But They’re Not Using It. April 2025.
My Take: This “unprompted” or push workflow, where agents are proactive and predict what is needed before the user asks. … This is where the world of knowledge work is headed.
World of DaaS.
I had a chance to participate in the World of DaaS event in DC last week. It was wonderful to see so many people, some I was meeting “IRL” for the first time.
Special thanks to Auren Hoffman & Elianne Marcus Kiejdan.
There are a ton of good things happening in the data space. I have lived through some ups & downs, and it definitely feels like the “data space”, broadly defined, is in an upswing. Perhaps the AI tailwind is giving fresh energy to the people doing the hard work of creating valuable data.
It is fun to be a part of such a vibrant industry….having the chance to moderate a conversation with Roy Schwartz, CEO of AxiosHQ was the highlight for me!
