Podcast Summary: The Analytics Power Hour – Episode #273: "Data Products Are... Assets? Platforms? Warehouses? Infrastructure? Oh, Dear." Featuring Eric Sandersham

Release Date: June 10, 2025
Hosts: Michael Helbling, Moe Kiss, Tim Wilson, Val Kroll, and Julie Hoyer
Guest: Eric Sandersham, Founder and Partner at Red and White Consulting Partners

Introduction

In Episode #273 of The Analytics Power Hour, titled "Data Products Are... Assets? Platforms? Warehouses? Infrastructure? Oh, Dear," host Tim Wilson embarks on an in-depth exploration of what constitutes a data product. Joined by co-hosts Val Kroll and Moe Kiss, and with special guest Eric Sandersham returning to the show, the discussion delves into the complexities and ambiguities surrounding the definition and classification of data products within the digital analytics landscape.

Defining Data Products

The episode kicks off with Tim Wilson introducing the central question: "Can a single chart be a data product?" (00:05). Val Kroll humorously recalls her initial skepticism but anticipates a challenging yet enlightening discussion (01:41). Moe Kiss from Canva shares her ongoing debates with her team, highlighting the lack of a definitive stance on the matter (01:50).

Eric Sandersham provides a foundational perspective, sharing his journey to define a data product. He contrasts traditional definitions, which range from curated databases to functional tools processing data, and points out the blurred lines in the digital realm where distinguishing between a product and its components becomes challenging (03:35–08:20). Eric emphasizes the importance of intentional curatedness and specific utility in defining a data product, arguing that it's not merely about data presence but about solving specific decision-making uncertainties (08:25–09:49).

Eric Sandersham (04:21): "If it's just there and it doesn't do anything to fix someone's decision making, then to me, I don't call it a data product."

Data Products vs. Digital Products

The conversation shifts to the comparison between data products and digital products. Val and Mo provide real-world examples, such as mobile banking platforms and their embedded features like remittances, questioning whether these features qualify as standalone products (11:38–15:01).

Tim Wilson adds to the complexity by discussing his misunderstanding of digital products years ago and how components of an e-commerce experience, like the checkout process, are treated as separate products with dedicated product teams (15:01–16:38).

Eric Sandersham elaborates on this by comparing data and physical products through analogies like peanut butter manufacturing. He distinguishes between externally marketed products and internally used data assets, questioning whether internal data processes should be labeled as products (16:38–22:54).

Eric Sandersham (22:54): "It’s a really good example."

Interactive Discussion: Is It a Data Product?

To further dissect the topic, the hosts engage in a playful yet insightful game to categorize various tools and dashboards as data products or not.

1. A/B Testing Sample Size Calculator (45:18–50:03):

   • Tim Wilson declares it is not a data product.
   • Val Kroll and Mo Kiss agree.
   • Eric Sandersham concurs, arguing that without scalability and reusability, it's merely a tool.

   Eric Sandersham (46:36): "It's a data solution for sure... but to say this has reusability, generalizability, I'm not sure because it's very bespoke."

2. Interactive Dashboard with Campaign Performance and Targets (50:39–53:55):

   • Val Kroll believes it is a data product if it aids in decision-making.
   • Eric Sandersham remains skeptical, focusing on whether the dashboard’s core value lies in its data integration versus mere visualization.

   Val Kroll (51:00): "I say yes, if targets are included... it's helping in full action."

3. Single Chart Based on a Data Asset (53:51–56:59):

   • Tim Wilson advocates for considering simple charts as data products.
   • Eric Sandersham suggests scalability and potential for broader use might qualify it.

   Eric Sandersham (56:59): "If you build an A/B testing engine where thousand data scientists can now use, reuse the same kind of functionality, then perhaps yes."

The game underscores the nuanced criteria for what makes a data product—primarily focusing on scalability, reusability, and the presence of curated data assets.

The Value and Implications of Data Products

Val Kroll steers the conversation towards the practical implications of labeling something as a data product. She explores how this nomenclature affects team responsibilities, maintenance, and business value.

Val Kroll (29:06): "What is the value of calling something a data product?... what's different from the other side?"

Moe Kiss highlights the transformation in the role of data professionals when adopting a data product mindset, emphasizing strategic problem-solving and reducing reliance on ticket-based service functions (29:06–34:26).

Eric Sandersham provides a counterpoint, cautioning against the overuse of the term "product" and its implications on P&L accountability. He differentiates between external-facing products and internal data solutions, advocating for clear boundaries to maintain focus and value (33:46–40:00).

Eric Sandersham (40:00): "Should products only be reserved for an external facing part of that business?"

The discussion reveals a clear divide between those who see data products as strategic assets that require dedicated product management and those who view them as tools or features within broader digital products.

Final Thoughts and Closing Remarks

As the episode nears its conclusion, the hosts reflect on the depth and complexity of defining data products. They recognize that while the conversation has not led to a definitive answer, it has provided substantial food for thought.

Eric Sandersham shares insights from his recent readings on large reasoning models and their limitations, emphasizing the nuanced challenges in AI-driven data solutions (57:35–61:21).

Moe Kiss discusses innovative features like Snowflake's Cortex and its integration into SQL workflows, showcasing practical advancements in data product capabilities (61:21–62:44).

Val Kroll appreciates the engaging discussions and highlights related content, fostering a community-driven conversation around analytics (62:44–63:37).

Tim Wilson wraps up with a nod to favorite data visualizations and the importance of continued analysis, reinforcing the podcast’s commitment to deep analytical discussions (63:37–69:26).

Key Takeaways

1. Ambiguity in Definitions: The line between data products, tools, and features remains blurred, especially in digital contexts where scalability and reusability are pivotal.
2. Strategic Value: Labeling something as a data product elevates its perceived strategic importance, necessitating dedicated management and maintenance.
3. Role of Teams: Clear delineation between data teams and product teams is essential to avoid overlapping responsibilities and ensure focus on value-driven solutions.
4. Scalability and Reusability: For a tool to be considered a data product, it must exhibit scalability and reusability beyond its initial use case.
5. Community Engagement: Ongoing discussions and debates within the analytics community are crucial for refining definitions and best practices around data products.

Notable Quotes

• Tim Wilson (00:05): "Does that mean that I built a data product? I mean, it's just one chart, but maybe I did."

• Eric Sandersham (08:25): "To me, perhaps then that would make a data product."

• Val Kroll (29:06): "What is the value of calling something a data product?"

• Moe Kiss (30:23): "When you're building something, quote unquote, that you want to call a data product, you're normally stealing some of those things from the product folk you're doing."

• Eric Sandersham (40:00): "Should products only be reserved for an external facing part of that business?"

Conclusion

Episode #273 of The Analytics Power Hour offers a comprehensive and thought-provoking discussion on the nature of data products. By dissecting various definitions, real-world examples, and engaging in interactive debates, the hosts and guest Eric Sandersham provide listeners with valuable insights into the evolving landscape of data analytics. While definitive answers remain elusive, the episode successfully highlights the importance of clarity, scalability, and strategic value in defining and managing data products.

Stay Connected:
For more discussions, insights, and updates, follow The Analytics Power Hour on Twitter, visit our website, join our LinkedIn group, or connect via the MeasuredChat Slack community.