Alternative Data Weekly #231

Mar 28, 2025

Thanks for being here!

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 … leading a data business through different stages of growth is a challenge.

QUOTES

Successful products bring “data as a first-class citizen into the workflow” - Shaun Clowes

My Take: Making things simple is a skill. We are all tempted to show all our work and overwhelm the reader with stats, and words, and charts. The key is to communicate your message as cleanly and simply as possible.
This can be very hard to do.

My Take: “We are all learning at the same time”. I’ve noticed over the past 12-18 months, as I’ve discussed AI with prospects & customers, I’ve seen a lot of “AI Tourism”. These are people who take the meeting just to see what’s going on and take a pulse of the market. No one knows. No one person can keep up with the pace of change. This feels like the internet in 1999. Things are changing and no one exactly knows what the AI killer app will be. Good leaders lead through uncertainty. This is a period of wild predictions and uncertainty and FOMO. Stay on your front foot, be educated, ask questions, and bring your domain expertise to the party.

What else I am reading:

Cassie Kozyrkov published All My Blog Posts In One Place. March 2025.

Michael Spencer, The Pycoach, and Frank Andrade published What is Vibe Coding?. March 2025.

Source: Lenny’s Podcast interviews Shaun Clowes, Chief Product Officer at Confluent. December 2024.

My Take: Really interesting episode. Of most interest to ADW’s “data” audience is likely starting around minute 30. Successful products bring “data as a first-class citizen into the workflow” (Minute 33:00).
LLMs can only be as good as the data they are given. How do you get customer information into LLMs so you can probe while incorporating context (#contextisking).
90% of the “calories” are consumed as you get good, timely, well-structured data into the LLM. It is a data management problem.

“Right data, right time, right place” (36:20)

Product Managers: constantly seek and get feedback. “Spend 80% of your time focusing on the external factors—customers, competitors, and market conditions—rather than internal processes. Seek out the proof that you are wrong.
Product management is still an undeveloped discipline. Are there 10x product managers, just like there are 10x engineers? How can these types of PMs be produced?

Highlights (80-minute run time)

Minute 05:00 – interview starts, Shaun’s background
Minute 09:30 – how to level up as a PM
Minute 11:30 – are you talking to customers enough? Probably not.
Minute 14:00 – What tools work? Right size your efforts (straight-up LLMs are good enough)
Minute 19:30 – the way AI will impact PM is through data processing; information has a decay rate
Minute 24:30 – AI makes it easy to create products; when is Shaun’s sense here of the future (where is value created?) … this is unbelievably hard
Minute 27:00 – you can make this product your (Salesforce, Workday, etc.); is it the years & years of underlying evolution of the product to make it yours
Minute 28:40 – Agents will replace UI; Shaun does not see AI displacing entrenched products (Jira, etc)
Minute 31:00 – Distribution advantage is the biggest advantage (AI making this harder)
Minute 33:00 – There has to be some angle on which you are materially better
Minute 36:00 – Advice to founder trying to put data into the workflow:
Minute 38:00 – “Data for product managers” course
Minute 43:00 – Final goal is happy customers paying us money
Minute 50:00 – PLG – Does Product Lead Growth work?
Minute 55:00 – The rest is career advice and some conversation on past failures

Source: June Dershewitz published What Happens When a Data Org Is Off-Balance? (Part 2 of 2). March 2025. Part 1 can be found here.

The tricky balance between BI, data engineering, data science, and leadership.
The goal is to deliver obvious business value for stakeholders. As your business morphs through different growth stages, different groups will have overweight & underweight impacts. The key as a leader is to be aware of this and maintain balance as you grow to be a tailwind for your business.