Magis

Magis by Alex Izydorczyk explores the intersections of data, finance, and economics with a focus on DaaS (data-as-a-service) businesses, offering insights on fundraising strategies, data utilization in decision making, geopolitical implications, and the integration of AI in economic forecasting and investment management.

Data-as-a-Service (DaaS) Startups and Fundraising Alternative Data Economic Forecasting Geopolitics and Data Analysis AI and Machine Learning Investment Management Semantic Web Financial Markets

Top posts of the year

And their main takeaways
36 implied HN points 19 Jun 23
  1. Startups can raise large upfront rounds to de-risk future fundraising and take advantage of money's present value.
  2. Founders should benchmark their equity dilution against similar funding rounds to understand ownership implications.
  3. Raising more capital than necessary can lead to giving away equity at a discount, especially for capital-efficient startups.
27 implied HN points 15 May 23
  1. The post discusses an interview on the Alternative Data Podcast about Cybersyn and Coatue.
  2. The interview reflects on topics like culture at Cybersyn, alternative data adoption in asset classes, and opportunities in the public sector.
  3. The post suggests subscribing to the author's writing to stay informed.
Get a weekly roundup of the best Substack posts, by hacker news affinity:
3 HN points 22 Apr 23
  1. LLMs can potentially boost productivity for investment managers by automating tasks like creating Excel models and analyzing communication logs.
  2. Using tools like Gong to automate note-taking during Zoom calls can provide investment managers with valuable insights from meetings.
  3. Investment managers have the opportunity to leverage LLMs to extract structured datasets from unstructured data sources, enhancing their analytical capabilities.
2 HN points 03 Feb 24
  1. Credit card data remains valuable despite its availability because of the infrastructure and talent required to utilize it effectively.
  2. Having the computational resources and expertise to analyze consumer spending data gives larger firms an advantage over smaller firms.
  3. Success in leveraging consumer spending data depends on the rarity of talent that can understand and apply it effectively.
2 HN points 02 Jul 23
  1. Snowflake Summit 2023 introduced key features including a partnership with Nvidia, Snowpark Container Services for machine learning, and updates to the Native Application Framework.
  2. Snowflake announced new options for paying Marketplace Listings using Snowflake capacity contracts, custom billing events for native applications, and data governance features like Aggregation Constraints.
  3. Additional announcements at Snowflake Summit 2023 included updates in Snowflake SQL, a new Snowflake Performance Index, and the ability to set spending alerts and calculate cost run-rates.
1 HN point 20 Feb 24
  1. Simpson's paradox teaches us that aggregate metrics can lead to wrong conclusions when not considering the composition of the aggregate.
  2. A common construction of churn metrics can be misleading; decreasing aggregate churn rate may not always mean the revenue base quality is improving, and sudden increases can be due to new customers rather than a decline in quality.
  3. Churn paradox occurs when fixed-period aggregate churn rates ignore the sizes of customer acquisition cohorts, leading to skewed conclusions about customer retention and revenue base.
1 HN point 14 Feb 24
  1. Selling data for training generative models is challenging due to factors like lack of marginal temporal value, irrevocability, and difficulties in downstream governance.
  2. Traditional data sales rely on the value of marginal data points that become outdated, while data for training generative models depends more on volume and history.
  3. Potential solutions for selling data to model trainers include royalty models, approximating dataset value computationally, and maintaining neutral computational sandboxes for model use.
1 HN point 26 Dec 23
  1. Measuring web traffic is crucial for digital properties to understand visitor demographics and behavior.
  2. Measuring web traffic is challenging due to decentralized internet infrastructure and complex metric standardization.
  3. Commercial data providers struggle to accurately estimate web traffic, showing the difficulty in obtaining precise data.