August 2021 Summaries

Heap has introduced several new features to enhance dashboard usability and improve the Heap Connect experience, largely based on user feedback. The updates include the introduction of dashboard filters that allow users to customize data views without duplicating dashboards, the ability to star frequently used dashboards for easy access, and a new Metric Card that displays key metrics prominently. Additionally, Heap Connect now offers faster user migration in Redshift, new default column encodings for improved storage efficiency and query speed, and updated recommendations for Redshift cluster settings. These enhancements aim to provide users with more efficient data analysis and integration capabilities.

Statisticians often seek to measure the "interestingness" of a statistical result through metrics like "variance explained," with McFadden's pseudo-R² being a particularly useful tool for non-linear outcomes such as binary predictions. This post explores the practical application of pseudo-R² in Heap's Group Suggestions feature, which automatically recommends insightful groupings in funnel analysis by quantifying their potential impact. Pseudo-R² is appreciated for its ability to balance variation with group composition, rewarding groupings with common categories that have distinct success rates. Unlike other metrics like log-loss or binomial deviance, pseudo-R² normalizes results between 0 and 1, making it easier to compare models across different datasets. The metric's adaptability to handle multiple groups makes it particularly valuable in product analytics, where interpretability and robustness are crucial. The author's experience at Heap highlights the value of classical statistical methods in building scalable, automated solutions that deliver meaningful insights, underscoring the continued relevance of pseudo-R² in modern data science applications.

Heap Champions is a new initiative launched to celebrate and engage a select group of advanced users and brand ambassadors of Heap, a digital insights platform. This community comprises 28 distinguished analysts from various industries, including small-business SaaS and enterprise eCommerce, who have demonstrated exceptional leadership. These individuals will participate in a year of career development, knowledge sharing, exclusive speaking opportunities, and networking with other Heap power users, thereby contributing to the evolution of digital insights. Some notable members include data professionals and product managers from companies such as Amway, Carrier Global, SS&C Advent, and Workhuman, highlighting the diverse backgrounds and expertise within the group.

Tech debt, often perceived as an unavoidable hindrance, can actually signal success when managed well, as it reflects growth and learning within a product team. Instead of eliminating tech debt, the focus should be on managing it through a holistic understanding of customer needs, product priorities, and business goals, while leveraging the insights of Product and Engineering Managers as allies. Effective management involves setting clear expectations and goals, fostering trust through collaborative discussions, and incorporating feedback to ensure future projects are feasible. Plans should be adaptable, acknowledging that not every project can be completed as initially envisioned, and that technical investments should align with business objectives. Introducing small, manageable improvements, like addressing "papercut" issues, can provide significant value over time. These overlooked tickets, when consistently integrated into sprint planning, can reveal patterns of tech debt that need addressing and ultimately contribute to a more robust and adaptable product roadmap.

Heap, a digital insights platform, recently introduced Effort Analysis, a feature that allows users to assess the median interactions and engagement time between funnel steps. The development faced a performance challenge when scanning over a billion rows of event data, initially resulting in slower query speeds. The issue stemmed from a PostgreSQL quirk where functions and operators in the query prevented the use of an index-only scan, which is faster than a standard index scan. Heap overcame this by employing a workaround involving partial indexes, which allowed the query planner to utilize an index-only scan by matching the query predicate with the partial index predicate. This adjustment led to a twofold improvement in the p90 performance of the Effort Analysis feature. The company hopes to share these insights to help other teams optimize their PostgreSQL queries effectively.