Skip to content
Artwork for Let’s Solve IT! Podcast
Let’s Solve IT! Podcast · November 4 · 14 min

Ep. 18 - Is bad data really an IT problem?

Artificial intelligence has raised the stakes for data quality, but is bad data really an IT problem? In this episode of Let’s Solve IT!, host Matt Brown sits down with NetApp Senior Vice President and Chief Information Officer Hem Nerkar to challenge one of the biggest misconceptions in enterprise AI. While IT builds and manages the technology that powers AI, the business owns the data that determines whether AI produces meaningful results. Hem explains why improving data quality requires more than better technology. It demands executive sponsorship, business ownership, governance frameworks, trusted data sources, and continuous collaboration across the organization. Together, they explore how NetApp approaches data stewardship, measures success, establishes accountability, and creates the discipline needed to turn enterprise data into a strategic asset. In this episode, you’ll learn: Why bad data is a business challenge, not just an IT problem How business and IT share responsibility for building trusted, AI-ready data Why data governance and stewardship are essential for accurate AI and analytics How to identify trusted data sources, data lineage, and ownership across the enterprise Practical first steps CIOs and business leaders can take to improve data quality without trying to fix everything at once Why data quality is an ongoing discipline, not a one-time project Whether you’re leading AI initiatives, modernizing your data strategy, or trying to improve trust in analytics, this conversation offers practical guidance on where to start and why data governance is becoming one of the most important business capabilities in the AI era. You are not alone. Let’s Solve IT. Learn More: IT case studies | NetApp Connect with us! https://www.linkedin.com/in/cmattbrown https://www.linkedin.com/in/hem-nerkar-445b731 ----- Episode overview In this episode of "Let's Solve IT!," host Matt Brown talks with Hem Nerkar, Senior Vice President and Chief Information Officer at NetApp, about a question Matt Brown says he hears constantly from customers: when AI or a dashboard produces a bad result, is that automatically an IT problem? Hem Nerkar's answer sets the frame for the entire conversation. IT manages the technology infrastructure, but the business owns the data, including how that data gets organized and used to achieve company goals. Hem Nerkar explains why this question has become urgent now. The "AI age" has intensified the stakes around data quality, since AI hallucination traces back, in many cases, to poor underlying data. Hem Nerkar describes NetApp's operational response: a data governance framework built on partnership between IT and business units like finance, sales, and product teams. In this model, IT acts as a "caretaker" and "coordinator," providing the architecture and tools, while business leaders own the data, decide acceptable accuracy levels, and sit as named stewards within the governance structure. Hem Nerkar also walks through concrete mechanics: identifying true data sources (as opposed to second or third-hop copies that get "contaminated" as they move between systems), maintaining an extensive catalog of data entities and their owners, and consolidating sources through enterprise resource planning (ERP) implementation. Hem Nerkar closes with advice for CIOs starting this work: acknowledge the issue, quantify the challenge, and build the discipline step by step. Key takeaways The following takeaways summarize Hem Nerkar's guidance on shared data ownership and governance for IT leaders. IT manages the technology infrastructure, but the business owns the data, including how it's organized and used toward company goals. The maturing AI age has made data quality more urgent, since AI hallucination often traces back to the quality of the underlying data. Data accuracy is a business decision, not a CIO decision. The business must determine what accuracy percentage is acceptable for its goals. A formal data governance framework works best when it includes named stewards from each business unit, such as finance, sales, and product teams. IT's role in data governance is to coordinate, not own. IT builds the framework, tools, and architecture, while business units own the data itself. Identifying the true, original source of data is critical, since customer and product information often becomes distorted as it passes through multiple systems. NetApp consolidated data sources through an ERP implementation, reducing the number of places data was flowing from. NetApp maintains an extensive data catalog tracking entities, key attributes, business owners, and data stewards. That catalog requires constant maintenance, not a one-time setup. Getting business leaders aligned starts with empathy. Business owners have their own day-to-day priorities, so IT needs to show how the data challenge affects those priorities directly. Incremental success starts with acknowledging there's an issue, quantifying the challenge, and building visibility with executives step by step. Topics and themes This episode covers the shared responsibility model for enterprise data quality, framed around governance structure, source management, and the AI-driven urgency behind it. Shared data ownership: the distinction between IT's infrastructure responsibility and business's data ownership. AI as a driver of data quality urgency: hallucination and other AI maturity issues traced back to poor data. Data governance framework: structure, stewards, and cadence built in partnership between IT and business units. Business measurement of accuracy: acceptable data accuracy levels as a business decision, not an IT decision. Data source identification: distinguishing original data sources from downstream, contaminated copies. Data lineage and architecture: tracking how data moves across systems as a foundation for governance. Data cataloging: maintaining a living record of entities, attributes, owners, and stewards. ERP consolidation: reducing the number of data sources through enterprise resource planning implementation. Business and IT partnership: framing governance as joint ownership rather than an IT-only mandate. Incremental CIO strategy: a step-by-step approach to acknowledging, quantifying, and addressing data challenges. Notable quotes "IT is responsible for managing the technology infrastructure, but business really owns the data with respect to how that data is organized and how it's getting used for our business and achieving company's goals." (1:12) "We are in this AI age, but as AI gets matured, AI has a lot more issues right now. Hallucination is one of the big ones. One of the main reasons is the data, and that's where the data is even more important right now than it was ever before." (2:15) "IT's job is to make sure that we have that governance framework established in the organization with all of this key owners sitting there with us, we being the caretaker as IT, making sure that we have that process laid out, the governance laid out, cadence laid out." (3:54) "The challenge could be that the data goes through multiple hops. The customer information flows from one system to another system, and that's where a lot of contamination happens." (7:10) "It needs to be constantly care and feeding." (12:02) "Incremental success starts with acknowledging there is an issue, identifying where the challenges are, quantifying the challenges, focusing on the right area step by step, getting those visibility to execs and all of the team members and continue to work on it." (13:51)

0:00-14:56

transcript

No transcript — this publisher did not publish one.

show notes

Artificial intelligence has raised the stakes for data quality, but is bad data really an IT problem? 

In this episode of Let’s Solve IT!, host Matt Brown sits down with NetApp Senior Vice President and Chief Information Officer Hem Nerkar to challenge one of the biggest misconceptions in enterprise AI. While IT builds and manages the technology that powers AI, the business owns the data that determines whether AI produces meaningful results. 

Hem explains why improving data quality requires more than better technology. It demands executive sponsorship, business ownership, governance frameworks, trusted data sources, and continuous collaboration across the organization. Together, they explore how NetApp approaches data stewardship, measures success, establishes accountability, and creates the discipline needed to turn enterprise data into a strategic asset. 

In this episode, you’ll learn: 

  • Why bad data is a business challenge, not just an IT problem 
  • How business and IT share responsibility for building trusted, AI-ready data 
  • Why data governance and stewardship are essential for accurate AI and analytics 
  • How to identify trusted data sources, data lineage, and ownership across the enterprise 
  • Practical first steps CIOs and business leaders can take to improve data quality without trying to fix everything at once 
  • Why data quality is an ongoing discipline, not a one-time project 

 

Whether you’re leading AI initiatives, modernizing your data strategy, or trying to improve trust in analytics, this conversation offers practical guidance on where to start and why data governance is becoming one of the most important business capabilities in the AI era. 

You are not alone. Let’s Solve IT. 

 

Learn More: 

IT case studies | NetApp 

 

Connect with us! 

https://www.linkedin.com/in/cmattbrown 

https://www.linkedin.com/in/hem-nerkar-445b731 

 

-----

 

Episode overview 

In this episode of "Let's Solve IT!," host Matt Brown talks with Hem Nerkar, Senior Vice President and Chief Information Officer at NetApp, about a question Matt Brown says he hears constantly from customers: when AI or a dashboard produces a bad result, is that automatically an IT problem? Hem Nerkar's answer sets the frame for the entire conversation. IT manages the technology infrastructure, but the business owns the data, including how that data gets organized and used to achieve company goals. 

Hem Nerkar explains why this question has become urgent now. The "AI age" has intensified the stakes around data quality, since AI hallucination traces back, in many cases, to poor underlying data. Hem Nerkar describes NetApp's operational response: a data governance framework built on partnership between IT and business units like finance, sales, and product teams. In this model, IT acts as a "caretaker" and "coordinator," providing the architecture and tools, while business leaders own the data, decide acceptable accuracy levels, and sit as named stewards within the governance structure. 

Hem Nerkar also walks through concrete mechanics: identifying true data sources (as opposed to second or third-hop copies that get "contaminated" as they move between systems), maintaining an extensive catalog of data entities and their owners, and consolidating sources through enterprise resource planning (ERP) implementation. Hem Nerkar closes with advice for CIOs starting this work: acknowledge the issue, quantify the challenge, and build the discipline step by step. 

Key takeaways 

The following takeaways summarize Hem Nerkar's guidance on shared data ownership and governance for IT leaders. 

  • IT manages the technology infrastructure, but the business owns the data, including how it's organized and used toward company goals. 
  • The maturing AI age has made data quality more urgent, since AI hallucination often traces back to the quality of the underlying data. 
  • Data accuracy is a business decision, not a CIO decision. The business must determine what accuracy percentage is acceptable for its goals. 
  • A formal data governance framework works best when it includes named stewards from each business unit, such as finance, sales, and product teams. 
  • IT's role in data governance is to coordinate, not own. IT builds the framework, tools, and architecture, while business units own the data itself. 
  • Identifying the true, original source of data is critical, since customer and product information often becomes distorted as it passes through multiple systems. 
  • NetApp consolidated data sources through an ERP implementation, reducing the number of places data was flowing from. 
  • NetApp maintains an extensive data catalog tracking entities, key attributes, business owners, and data stewards. That catalog requires constant maintenance, not a one-time setup. 
  • Getting business leaders aligned starts with empathy. Business owners have their own day-to-day priorities, so IT needs to show how the data challenge affects those priorities directly. 
  • Incremental success starts with acknowledging there's an issue, quantifying the challenge, and building visibility with executives step by step. 

Topics and themes 

This episode covers the shared responsibility model for enterprise data quality, framed around governance structure, source management, and the AI-driven urgency behind it. 

  • Shared data ownership: the distinction between IT's infrastructure responsibility and business's data ownership. 
  • AI as a driver of data quality urgency: hallucination and other AI maturity issues traced back to poor data. 
  • Data governance framework: structure, stewards, and cadence built in partnership between IT and business units. 
  • Business measurement of accuracy: acceptable data accuracy levels as a business decision, not an IT decision. 
  • Data source identification: distinguishing original data sources from downstream, contaminated copies. 
  • Data lineage and architecture: tracking how data moves across systems as a foundation for governance. 
  • Data cataloging: maintaining a living record of entities, attributes, owners, and stewards. 
  • ERP consolidation: reducing the number of data sources through enterprise resource planning implementation. 
  • Business and IT partnership: framing governance as joint ownership rather than an IT-only mandate. 
  • Incremental CIO strategy: a step-by-step approach to acknowledging, quantifying, and addressing data challenges. 

Notable quotes 

  • "IT is responsible for managing the technology infrastructure, but business really owns the data with respect to how that data is organized and how it's getting used for our business and achieving company's goals." (1:12) 
  • "We are in this AI age, but as AI gets matured, AI has a lot more issues right now. Hallucination is one of the big ones. One of the main reasons is the data, and that's where the data is even more important right now than it was ever before." (2:15) 
  • "IT's job is to make sure that we have that governance framework established in the organization with all of this key owners sitting there with us, we being the caretaker as IT, making sure that we have that process laid out, the governance laid out, cadence laid out." (3:54) 
  • "The challenge could be that the data goes through multiple hops. The customer information flows from one system to another system, and that's where a lot of contamination happens." (7:10) 
  • "It needs to be constantly care and feeding." (12:02) 
  • "Incremental success starts with acknowledging there is an issue, identifying where the challenges are, quantifying the challenges, focusing on the right area step by step, getting those visibility to execs and all of the team members and continue to work on it." (13:51) 
links3