ManoMano's Journey to Enhanced Governance with CastorDoc

This customer story is contributed by Yoann Molina, Senior Data Governance Manager at ManoMano, a European leader in DIY, home improvement, and gardening products. Founded in 2013, ManoMano operates in six European markets, serving 50 million monthly visitors and 7 million active customers. With over 10 million products from 3,600 merchants, ManoMano continues to innovate and lead the online home improvement sector in Europe.

Introduction

"Our main goal is straightforward: we want to ensure that strategy and business teams have all the data they need to make smarter decisions." — Yoann Molina, Senior Data Governance Manager at ManoMano.

At ManoMano, I lead the data governance team, collaborating closely with data engineers, analysts, and data scientists to ensure our data is managed effectively across the company. As a marketplace that started in France, we’ve rapidly expanded to serve six countries, including Germany, the UK, Spain, and Italy, which has significantly increased our data needs.

Being a digital-first company, data is at the core of everything we do. From creating personalized customer experiences to making major business decisions, we rely on data daily. My role is to ensure that our data is organized, shared and easily accessible for everyone.

At ManoMano, we manage our data using a mix of both old and new systems:

1. An older, large system that uses MySQL for data storage.
2. Over 350 smaller, specialized applications utilizing Postgres databases.
3. Kafka for moving data events between systems, which we then store in Amazon S3.
4. Snowflake for data analysis.

Our tech team is structured as follows:

• 21 teams handling day-to-day data operations.
• 3 separate teams focusing on data engineering, data analytics and data science to inform strategic decisions.

Our main goal is straightforward: we want to ensure that strategy and business teams have all the data they need to make smarter decisions. It’s all about providing the right information to the right people at the right time.

The Problem: Scaling Data Documentation and Accessibility

"Before we implemented CastorDoc, we had almost no documentation on our tables. There was some information on certain KPIs, but it wasn't necessarily up to date." - Yoann Molina, Senior Data Governance Manager at ManoMano.

As ManoMano grew, we faced several challenges related to data management:

1. Lack of centralized documentation: With over 350 microservices and countless data sources, finding and understanding our data assets became increasingly difficult. It was like searching for a needle in a haystack.
2. Inconsistent data definitions: Different teams often interpreted the same data in different ways, which led to discrepancies in reporting and decision-making. It was like everyone was speaking a different language when it came to our data.
3. Difficulty in tracking data ownership: As our company restructured, keeping track of who owned and maintained different data assets became a real challenge. Some teams disappeared while new ones emerged, which complicated things further.
4. Limited visibility into data lineage: Understanding how data flowed through our complex system of monoliths, microservices, and data warehouses was becoming harder by the day.

These challenges were affecting our data teams' efficiency and hindering our ability to make quick, data-driven decisions. We needed a solution that would allow us to document our data assets comprehensively and make this information easily accessible to everyone in the company.

That's when we recognized the need to take stock of our data—essentially conducting an inventory. This is where CastorDoc came in. With CastorDoc's data catalog, we were able to centralize all our data in one system, making it easier to organize and visualize. We also set up a business dictionary to standardize not only data terminology, but also business definitions for the entire company.

The Solution: Implementing CastorDoc for Data Cataloging and Governance

"We chose CastorDoc because of its user-friendly interface and its ability to cater to both technical and business users" - Yoann Molina, Senior Data Governance Manager at ManoMano.

After evaluating several options, including some more complex, legacy solutions, we decided to implement CastorDoc in 2022. Other platforms felt overly complex and costly, while some were too technical for our needs at the time.

The decision was driven by CastorDoc's user-friendly interface, which aligned well with ManoMano's goal of making data accessible to all employees, regardless of their technical expertise.

The implementation process took about 3-4 months for the core features to be in place, with ongoing refinements and feature adoptions continuing throughout the first year. Here’s how we approached it:

1. Integrating CastorDoc with Our Tools: We focused on connecting CastorDoc with our existing systems, like Snowflake, our data warehouse, Qlik, our dataviz tool, and then Kafka, our event streaming platform. This was key to ensuring smooth data flow and easy access across the organization.
2. Documenting Key Data Assets: We knew we needed solid documentation, so we started with the most important and commonly used data assets.
3. Boosting Adoption with Support and Training: Strong backing from management and targeted training helped us get everyone on board with CastorDoc. My team trained key users across different departments so they could effectively use the tool.
4. Making It Easy to Use with Popular Tools: Integrating CastorDoc with tools like Slack and Holacracy made it easier for everyone to use in their daily work. The simplicity of using CastorDoc for communication, especially through Slack, has been really helpful, as it allowed team members to easily share info about data tables and assets.

The Impact: Enhanced Data Quality, Accessibility, and Efficiency

"In just two years, we've gone from almost zero documentation to over 80% of our main assets being documented" - Yoann Molina, Senior Data Governance Manager at ManoMano.

The implementation of CastorDoc has had a significant impact on ManoMano's data operations. Over the last two years, we've made substantial strides in how we handle our data. We've documented over 80% of our main data assets, which has completely transformed how we work with and access our data.

Now that we have solid documentation, data discovery has become much easier. Our team can quickly locate the information they need, saving time and avoiding mistakes from using outdated or wrong data. This has resolved many issues we previously faced with scattered data and unclear data paths.

One of the biggest improvements is the clarity around data ownership. CastorDoc has helped us define who is responsible for each data asset, which has been especially useful as teams change. When we went through recent restructuring, CastorDoc made everything smoother by keeping ownership clear.

Additionally, the ability to track data lineage with CastorDoc has been a game-changer. It allows us to trace the flow of data through different systems, helping us understand the broader impact of any changes and ensuring the accuracy of our data across the board.

While it’s hard to quantify all the time saved, the improved efficiency from these changes—whether it’s a few minutes saved in searching for data or faster ETL processes—has had a noticeable impact on our overall productivity.

Lastly, CastorDoc has greatly contributed to enhancing our data governance. By standardizing our data definitions and improving data quality, we’re able to rely on more accurate data for better decision-making.

Measuring Success:

To track the impact of CastorDoc, I have been monitoring two key metrics:

1. Percentage of Documented Main Data Assets: This shows how we’re improving data documentation and accessibility.
2. Number of Active Users: We currently have around 120-138 unique users each month, showing strong adoption across departments.

Seeing growth in these metrics is a clear sign of how CastorDoc positively influences our data culture.

What's Next

"Our next big focus is extending CastorDoc's use to our operational teams and improving the documentation of our microservices" - Yoann Molina, Senior Data Governance Manager at ManoMano.

Looking ahead, we have plans to use CastorDoc even more to enhance our data operations. As we continue to expand, one of our key goals is to make CastorDoc more accessible to our operational teams, particularly those working with microservices. These teams often handle fast-moving, complex data, and we recognize that they need customized training and documentation that suits their specific needs. By tailoring our approach, we aim to ensure that CastorDoc becomes a valuable tool for them, helping improve efficiency.

Another important step will be improving our ability to track data lineage. While we've already made progress in understanding how data moves through our systems, we plan to go even deeper, mapping the entire journey of our data—from operational sources all the way to analytics. This will give us a much clearer view of our data flows, making it easier to understand the origins of any issues and ensure the accuracy of our insights.

We’re also focused on enhancing our AI-powered search. CastorDoc’s current AI Assistant is useful, but we know there's room for improvement. By increasing the coverage and accuracy of the AI, we aim to make data discovery even faster and more precise. This will reduce the time spent searching for information and allow our teams to focus on using data to drive value, rather than just locating it.

Of course, none of this matters if the tool isn't widely adopted, so we’ll continue pushing for broader usage of CastorDoc across all departments. Our goal is to embed it deeply into our company’s culture, making it the go-to resource for anything data-related. We’ll ensure that CastorDoc supports not just the technical teams, but every decision-making process in the organization.