We Spent 50% of Our Time Cleaning Data, Instead of Making Strategic Decisions (Eva Netušilová, GasNet)

May 28, 2024

She has been leading a team of 20 analysts, strategists and data specialists for 3 years. What are they working on at GasNet? Why do they deal so much with data governance? Why do they have two data warehouses now? How is GasNet preparing for AI integration and what role does data quality play in leveraging these technologies? We asked Eva Netušilová, Head of Strategy at GasNet.

Watch the interview (CZ)

Listen to the podcast (CZ)

Read the podcast as an interview

(The interview was shortened and edited using ChatGPT)

Ivana Karhanová: Earlier you mentioned that up to 50% of your time at GasNet was spent on data cleansing necessary for strategic planning. How did it come to be that so much time is dedicated to this task?

Eva Netušilová: In reality, our data was heavily dependent on primary systems, meaning that whenever we needed data for specific use cases, like network capacity calculations, we had to extract it from these systems. This involved accessing multiple fundamental systems to retrieve geographical and customer data, which we then had to integrate. We found that this process, especially for capacity calculations, was performed up to 500 times a year, which is quite significant.

Ivana Karhanová: Does this mean the data is downloaded 500 times?

Eva Netušilová: Yes, each instance required the preparation and processing of data specifically for individual calculations. It was quite demanding and proved to be insufficient. Once we recognized this issue, we began to focus on streamlining processes to better concentrate on managing our gas infrastructure, which needs to be monitored from various perspectives, rather than just dealing with data from primary systems. This necessity sparked a change.

Ivana Karhanová: So, you needed to centralize and organize all the data?

Eva Netušilová: Ideally, we wanted the data to always be available in the same structure, correctly cleaned, and ready for use, which would allow us to focus on the calculations themselves. We had previously implemented a data warehouse, but it lacked properly set rules, leading to uncertainties on how to effectively work with the data.

Ivana Karhanová: How did you address the situation when the data warehouse existed but was not functioning efficiently?

Eva Netušilová: I think we initially underestimated the need for a comprehensive solution. While the original idea was sound, it wasn’t fully thought through.

Ivana Karhanová: So the data was being pulled into the data warehouse from the primary systems?

Eva Netušilová: Yes, but it could happen that data was sometimes duplicated or incorrectly linked because we hadn’t properly set up the layers above the data and defined data governance correctly. Moreover, we lacked sufficiently trained personnel, leading to a situation where people, although willing and from various business departments, lacked sufficient understanding of these processes. We, therefore, started working on how to create internal capacities and a system of rules that would allow us to better utilize the data.
Ivana Karhanová: What solution did you find?

Eva Netušilová: We realized that we needed a smaller but more effectively managed data warehouse and an expert partner to help us with its proper implementation. This led to a tender for a partner who could advise us on how to optimize our data strategy and implement it across the company.

Ivana Karhanová: Do you believe that the main benefit of an external partner is that they see your data from a different perspective for the first time?

Eva Netušilová: I’m not sure if it’s just because they are seeing it for the first time. I think a significant advantage is their experience in other sectors. The processes we are now implementing have already been proven in telecommunications or banking. The energy sector might not seem similar at first glance, but the principles of data management and governance are universally transferable.

Ivana Karhanová: You are now building a new data warehouse. Why did you decide not to continue using the existing one?

Eva Netušilová: We explored the market and discovered that there are solutions that are not only more cost-effective but also more efficient than our existing setup. New technologies will allow us to construct a more effective and economically viable solution.

Ivana Karhanová: Does that mean that renovating the old warehouse would be more expensive and complex than building a new one from scratch?

Eva Netušilová: Exactly.

Ivana Karhanová: And did management agree to abandon the old solution and start investing in a new one?

Eva Netušilová: When we assessed it from an economic standpoint, the new solution proved to be more advantageous. It’s not just about aesthetics but primarily about financial efficiency.

Ivana Karhanová: How challenging is it currently to operate with two data warehouses?

Eva Netušilová: We are gradually phasing out the old warehouse. We have already mapped all the information and are transitioning it to the new solution. Our goal is to complete this process by the fall to avoid the complexities and team discomfort of running parallel systems.

Ivana Karhanová: Was it difficult to convince management to invest in a new data warehouse when you already have data in existing systems?

Eva Netušilová: Our company is focused on innovation. Given that we are approaching technologies such as artificial intelligence and machine learning, it’s clear that the quality of our data will directly affect the success of these tools. Therefore, the decision to invest was relatively straightforward, and there was consensus at both the management and operational levels.

Ivana Karhanová: What would you say if someone asked what your 20-member team does?

Eva Netušilová: We have divided competencies into several areas. The primary one is the preparation of asset management strategy, which involves how we will care for, renew, and evaluate the performance of our assets. We collaborate with the Work Delivery division, which performs maintenance on these assets.

Ivana Karhanová: So, you determine what needs to be done, and they decide when and how to do it?

Eva Netušilová: Essentially, yes. For efficient operations, we require various other segments. We need to monitor the energy market and anticipate its development, which requires specific analytical skills. Additionally, we need tools for evaluating the condition of our assets and analysts who not only collect but also evaluate the data. We also have a department focused on capacity calculations, aimed at optimizing the capacity of our networks for the future.

Ivana Karhanová: What are the other roles within your team?

Eva Netušilová: Some team members serve dual roles as analysts and mentors. They oversee the quality of data in our master system, which involves gathering user requirements and monitoring data quality. Additionally, I lead a small team focused on simulating mixed media in our network, where biomethane is added to natural gas.

Ivana Karhanová: Are you currently distributing only natural gas?

Eva Netušilová: We distribute methane and now biomethane as well.

Ivana Karhanová: And you will soon add hydrogen to that mix?

Eva Netušilová: Yes, we will have three types of media in our network. Despite this diversity, we strive to minimize the need for additional data analysts by automating wherever possible. We have implemented an automated reporting system that allows us to efficiently monitor and process incoming data.

Ivana Karhanová: When you talk about optimizing data, do you mean the preparation or cleaning process?

Eva Netušilová: Yes, the entire data preparation process for further use, so we can effectively respond to new situations.

Ivana Karhanová: How long have you been building the second data warehouse?

Eva Netušilová: It’s been the second year now.

Ivana Karhanová: And do you plan to complete it this fall?

Eva Netušilová: It’s an ongoing process. Although we aim to have the system operational for new uses and reporting across divisions as soon as possible, the reality is that completion is a long-term endeavor. We will continuously require improvements and adjustments to the data request processes.

Ivana Karhanová: So, two years ago, you managed to build a new data warehouse next to the existing one and are gradually transitioning to the new solution. What do you see as the key factors for successful data implementation across the company?

Eva Netušilová: The key factor is user trust in the new solution. In our company, each division has personnel preparing data, and our goal is not to take over their tasks but rather to standardize them according to clear rules. We want to ensure that all data comes from a single reliable source to avoid interpretation discrepancies.

Ivana Karhanová: Have you introduced mentor roles to support other departments?

Eva Netušilová: Yes, we have mentors and also a team that manages the data warehouse and prepares data for other departments. These individuals are crucial for maintaining communication and explaining the possibilities that the new solution offers.

Ivana Karhanová: Do you think other departments understand this new approach?

Eva Netušilová: I believe they do, but it took a while, as finding a common language about something so new is not simple. A well-established training system helps, allowing employees to sign up for training sessions tailored to their needs. We try to ensure that each division can efficiently compile their reports and also utilize reports generated by others.

Ivana Karhanová: How many people in your company currently work with data or are expected to do so in the future?

Eva Netušilová: Before starting the implementation, we conducted a survey to identify who is working with data and how. We found that dozens of employees across various divisions and teams are involved, which allows us to tailor the new solution to their needs and experiences with existing master systems.

Ivana Karhanová: What lessons have you learned from your experiences in data governance? What is often underestimated or typically fails to work?

Eva Netušilová: Often, users need a quick resolution to their needs. When data is not readily and centrally available, users find their own ways to address the situation. This “temporary” approach can lead to inconsistencies and errors in data.

Ivana Karhanová: How can the creation of these shadow areas be prevented?

Eva Netušilová: It’s important to understand when and why users need data and to ensure they have easy and quick access to verified information. The goal is not to prevent users from using data but to provide them with tools and processes that allow them to use data efficiently and safely.

Ivana Karhanová: How would you convince top management of the need to invest in a data platform and data governance?

Eva Netušilová: It depends on the type of company and the data it needs to process. It’s crucial to demonstrate how data affects compliance with legislative requirements, emission tracking, and other key operations. Companies need a reliable single source of truth that is auditable and can serve as the basis for precise decision-making. Effective data governance directly influences the quality of decision-making and company management.

Ivana Karhanová: Thank you for the interview, and I look forward to our next meeting.